CN117973238B - Intelligent prediction method and device for charge and discharge time of battery cell formation process - Google Patents
Intelligent prediction method and device for charge and discharge time of battery cell formation process Download PDFInfo
- Publication number
- CN117973238B CN117973238B CN202410391485.5A CN202410391485A CN117973238B CN 117973238 B CN117973238 B CN 117973238B CN 202410391485 A CN202410391485 A CN 202410391485A CN 117973238 B CN117973238 B CN 117973238B
- Authority
- CN
- China
- Prior art keywords
- battery cell
- cell
- data
- duration
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 175
- 230000015572 biosynthetic process Effects 0.000 title claims abstract description 78
- 238000013178 mathematical model Methods 0.000 claims abstract description 143
- 238000007599 discharging Methods 0.000 claims abstract description 140
- 238000012549 training Methods 0.000 claims abstract description 88
- 238000004458 analytical method Methods 0.000 claims abstract description 85
- 230000001419 dependent effect Effects 0.000 claims description 6
- 230000032823 cell division Effects 0.000 claims description 5
- 238000007405 data analysis Methods 0.000 abstract description 14
- 238000012545 processing Methods 0.000 abstract description 5
- 238000004364 calculation method Methods 0.000 description 16
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 14
- 229910052744 lithium Inorganic materials 0.000 description 14
- 230000006870 function Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 238000012417 linear regression Methods 0.000 description 6
- NBIIXXVUZAFLBC-UHFFFAOYSA-N Phosphoric acid Chemical compound OP(O)(O)=O NBIIXXVUZAFLBC-UHFFFAOYSA-N 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 229910000147 aluminium phosphate Inorganic materials 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000009417 prefabrication Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Secondary Cells (AREA)
Abstract
The invention relates to the field of battery core data processing, and discloses an intelligent prediction method and device for battery core formation flow charge and discharge time, which can take determined analysis parameters as a data analysis direction, take a mathematical model as a training core, perform data analysis on historical data of all first battery cores, so as to analyze and determine the influence relationship between the historical data of each first battery core and the analysis parameters, fully utilize the historical data of each first battery core and improve the data utilization rate of the historical data of the first battery core; then, when the input battery cell is analyzed through the mathematical model which is completed through training, the charging and discharging operation corresponding to the battery cell can be rapidly and accurately predicted according to the influence relation determined by the front part, and the prediction speed and the prediction accuracy of the charging and discharging time are improved; in addition, the battery cells can be paired by calling a predetermined process flow file, so that the charge and discharge time of the battery cells can be rapidly and accurately predicted.
Description
Technical Field
The invention relates to the technical field of cell data processing, in particular to an intelligent prediction method and device for charging and discharging time of a cell formation process.
Background
In the process of producing the battery cells, the formation procedure is a particularly important ring, and in the formation procedure, the formation procedures for different battery cells are also different. At present, a common formation technology generally adopts a preset formation process flow file, and the power supply is controlled through the content of the process flow file, so that the battery cell is charged and discharged for a specified duration. However, the actual formation process flow file is a general file, and even for different cells (such as cells of different cell types), the same process flow is adopted, which requires that the cells entering the formation process flow are guaranteed to be highly consistent in quality, so that the formation process flow file can be matched for each cell. However, in the actual process, the quality of each cell is necessarily different, and the use of the same process file may cause uneven final product quality of the cells.
A common solution is now by setting some cut-off conditions: and stopping the current process for the battery cell exceeding the set value in the formation process, and forcedly entering the next process, so that the voltage and current value of the battery cell is prevented from exceeding the allowable range to a certain extent. Such a solution may result in the same batch of cells being in different process flows, and then the cell that performs the process most quickly needs to wait for the cell that performs the end most slowly. In the formation process, the standing time after the battery cells are charged and discharged is also an important circle, so that the standing time of the battery cells can not meet the execution requirement due to the different processes of the same batch, and finally the quality of the battery cells can still be reduced. It is important to provide a corresponding solution for solving the technical problems of the prior art that the technological process file is not adapted and the standing time of the battery cell is not up to standard in the battery cell formation operation.
Disclosure of Invention
The invention provides an intelligent prediction method and device for charging and discharging time of a battery cell formation process, which can generate process flow files adapting to different types of battery cells, and improve the operation accuracy of charging and discharging executed by different battery cells in the formation process and the setting adaptation degree of standing time.
In order to solve the technical problems, the first aspect of the invention discloses an intelligent prediction method for charging and discharging time of a battery cell formation process, which comprises the following steps:
Determining all first electric cores needing to execute preset electric core prediction operation, and acquiring historical data corresponding to each first electric core, wherein the historical data corresponding to each first electric core is used for determining electric core formation data of the first electric core, and the electric core formation data of the first electric core are used for predicting charge and discharge time of the first electric core;
Determining a plurality of analysis parameters corresponding to each first battery cell according to the historical data corresponding to each first battery cell, wherein each analysis parameter is a parameter affecting the charge and discharge time length of the first battery cell for executing charge and discharge operation;
Inputting all the historical data into a preset mathematical model by taking all the analysis parameters as references so as to train the mathematical model;
the mathematical model after training is used for analyzing the battery cell input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell.
As an alternative embodiment, in the first aspect of the present invention, all the analysis parameters include at least one analysis parameter of a cell weight, an initial voltage, an internal resistance value, and a theoretical capacitance of each of the first cells, and are recorded as first sub-parameters;
After determining a plurality of analysis parameters corresponding to each first battery cell according to the historical data corresponding to each first battery cell, the method further comprises:
Dividing all the first electric cores into a plurality of electric core sets by taking the electric core identifier corresponding to each first electric core as a dividing reference, wherein each electric core set is matched with one electric core identifier, and the electric core identifier corresponding to each first electric core is used for indicating the model and/or the type of the first electric core;
and for each first cell in each cell set, taking the first sub-parameter corresponding to the first cell as a reference, extracting the history data matched with each first sub-parameter from the history data corresponding to the first cell, recording the history data as target history data, and executing data association or data filling on the first cell and the target history data corresponding to the first cell so as to update all the cell sets.
As an optional implementation manner, in the first aspect of the present invention, all the analysis parameters further include a single charging duration of each of the first electric cells, a number of times of charging corresponding to the single charging duration of the electric cell, a single discharging duration of the electric cell, and a number of times of discharging corresponding to the single discharging duration of the electric cell, which are recorded as second sub-parameters;
Inputting all the historical data into a preset mathematical model by taking all the analysis parameters as references so as to train the mathematical model, wherein the method comprises the following steps of:
For each first cell, inputting the target historical data corresponding to the first cell into a preset mathematical model, and executing model training operation on the mathematical model by taking the first sub-parameter corresponding to the first cell as an independent variable and the second sub-parameter corresponding to the first cell as a dependent variable to obtain a training result corresponding to the mathematical model;
after determining that the model training is performed on all of the first sub-parameters, all of the second sub-parameters by the mathematical model, it is determined that a model training operation for the mathematical model is completed.
As an optional implementation manner, in the first aspect of the present invention, each target history data has a corresponding data tag, where the data tag is used to indicate that the target history data is recorded after the first battery cell performs a charging operation or a discharging operation;
Before the target historical data corresponding to each first cell is input into a preset mathematical model, the method further comprises:
according to the data labels corresponding to each target historical data and the matching parameters in each second sub-parameter, matching operation is carried out on all the historical data and all the second sub-parameters, and a plurality of paired data are obtained;
updating all the target historical data according to all the pairing data;
Wherein the matching parameters include the number of charging and the number of discharging; each of the pairing data includes one of the target history data and one of the second sub-parameters.
As an optional implementation manner, in the first aspect of the present invention, after determining that the mathematical model completes training, the method further includes:
determining standing data corresponding to each first battery cell from historical data corresponding to each first battery cell, wherein the standing data corresponding to each first battery cell comprises a plurality of sub-standing time periods, and each sub-standing time period is the time period of standing after the first battery cell performs charging operation or discharging operation;
for all the first electric cores in each electric core set, generating a process flow file for the electric core set according to the first sub-parameters, the second sub-parameters, the target historical data and the sub-standing time length corresponding to each first electric core in the electric core set;
The process flow files comprise characteristic parameters of each first battery cell, wherein the characteristic parameters of each first battery cell are used for predicting the predicted charging duration corresponding to the charging operation of the first battery cell and the first standing duration after the charging operation is performed, or used for predicting the predicted discharging duration corresponding to the discharging operation of the first battery cell and the second standing duration after the discharging operation is performed.
As an optional implementation manner, in the first aspect of the present invention, after determining that the mathematical model completes training, the method further includes:
determining all the current battery cells needing to execute target prediction operation, marking the current battery cells as second battery cells, and acquiring battery cell data corresponding to each second battery cell;
The target prediction operation is used for predicting the predicted charging duration corresponding to the charging operation executed by the battery cell or the predicted discharging duration corresponding to the discharging operation executed by the battery cell; the corresponding electric core data of each second electric core comprises the electric core type of the second electric core, and the corresponding electric core data of each second electric core also comprises at least one data of the electric core weight, initial voltage, internal resistance value, theoretical capacitance, charging times and discharging times of the second electric core;
inputting the cell data corresponding to all the second cells into the mathematical model and executing the target prediction operation to obtain a predicted duration corresponding to each second cell, wherein the predicted duration comprises the predicted discharge duration or the predicted charge duration;
According to the cell division parameters corresponding to each second cell, performing division operation on all the second cells by combining the set time length deviation to obtain a plurality of cell division batches;
The battery cell dividing parameters corresponding to each second battery cell comprise the battery cell type corresponding to the second battery cell, the prediction duration and preset operation matters; the preset operation item is used for indicating that the second battery cell currently needs to execute the discharging operation or the charging operation.
As an optional implementation manner, in the first aspect of the present invention, each process flow file includes a plurality of sub-flow files, and each sub-flow file corresponds to one of the first electrical cores; the method further comprises the steps of:
For all the sub-process files in the same cell set, dividing all the sub-process files into a process file set corresponding to each deviation interval according to a preset deviation interval by taking the accumulated duration value corresponding to each sub-process file as a reference;
The accumulated time length value corresponding to each first battery cell is obtained by calculating the accumulated charging time length and the accumulated discharging time length of the first battery cell;
And the flow file sets are used for determining the predicted data of the battery cells to be analyzed, wherein the battery cells to be analyzed are the battery cells needing to execute the battery cell charging and discharging duration prediction operation, and the predicted data comprise the predicted discharging duration and the predicted charging duration corresponding to the battery cells to be analyzed.
The invention discloses an intelligent prediction device for charging and discharging time of a battery cell formation process, which comprises the following components:
The determining module is used for determining all first battery cells needing to execute preset battery cell prediction operation;
the acquisition module is used for acquiring historical data corresponding to each first battery cell, wherein the historical data corresponding to each first battery cell is used for determining battery cell formation data of the first battery cell, and the battery cell formation data of the first battery cell are used for predicting the charge and discharge time length of the first battery cell;
the determining module is further configured to determine, according to the historical data corresponding to each first electrical core, a plurality of analysis parameters corresponding to each first electrical core, where each analysis parameter is a parameter affecting a charge-discharge duration of the first electrical core when the first electrical core performs a charge-discharge operation;
the training module is used for inputting all the historical data into a preset mathematical model by taking all the analysis parameters as references so as to train the mathematical model;
the mathematical model after training is used for analyzing the battery cell input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell.
As an alternative embodiment, in the second aspect of the present invention, all the analysis parameters include at least one analysis parameter of a cell weight, an initial voltage, an internal resistance value, and a theoretical capacitance of each of the first cells, and are recorded as first sub-parameters;
the apparatus further comprises:
the first dividing module is used for dividing all the first electric cores into a plurality of electric core sets by taking the electric core identification corresponding to each first electric core as a dividing reference after the determining module determines a plurality of analysis parameters corresponding to each first electric core according to the historical data corresponding to each first electric core, wherein each electric core set is matched with one electric core identification, and the electric core identification corresponding to each first electric core is used for indicating the model and/or the type of the first electric core;
the extraction module is used for extracting the historical data matched with each first sub-parameter from the historical data corresponding to the first battery cell by taking the first sub-parameter corresponding to the first battery cell as a reference for each first battery cell in each battery cell set, and recording the historical data as target historical data;
And the association filling module is used for performing data association or data filling on the first battery cell and the corresponding target historical data so as to update all the battery cell sets.
As an optional implementation manner, in the second aspect of the present invention, all the analysis parameters further include a single charging duration of each of the first electric cells, a number of times of charging corresponding to the single charging duration of the electric cell, a single discharging duration of the electric cell, and a number of times of discharging corresponding to the single discharging duration of the electric cell, which are recorded as second sub-parameters;
the training module inputs all the historical data into a preset mathematical model by taking all the analysis parameters as references, and the mode for training the mathematical model specifically comprises the following steps:
For each first cell, inputting the target historical data corresponding to the first cell into a preset mathematical model, and executing model training operation on the mathematical model by taking the first sub-parameter corresponding to the first cell as an independent variable and the second sub-parameter corresponding to the first cell as a dependent variable to obtain a training result corresponding to the mathematical model;
after determining that the model training is performed on all of the first sub-parameters, all of the second sub-parameters by the mathematical model, it is determined that a model training operation for the mathematical model is completed.
As an optional implementation manner, in the second aspect of the present invention, each target history data has a corresponding data tag, where the data tag is used to indicate that the target history data is recorded after the first battery cell performs a charging operation or a discharging operation;
The training module inputs all the historical data into a preset mathematical model by taking all the analysis parameters as references, and the mode for training the mathematical model specifically further comprises the following steps:
Before the target historical data corresponding to each first battery cell is input into a preset mathematical model, matching operation is carried out on all the historical data and all the second sub-parameters according to the data label corresponding to each target historical data and combining the matching parameters in each second sub-parameter, so that a plurality of paired data are obtained;
updating all the target historical data according to all the pairing data;
Wherein the matching parameters include the number of charging and the number of discharging; each of the pairing data includes one of the target history data and one of the second sub-parameters.
As an optional implementation manner, in the second aspect of the present invention, the determining module is further configured to determine, from the historical data corresponding to each first electrical core, standing data corresponding to each first electrical core after determining that the mathematical model completes training, each standing data corresponding to each first electrical core includes a plurality of sub-standing periods, where each sub-standing period is a period of time for which the first electrical core stands after performing a charging operation or a discharging operation;
the apparatus further comprises:
The generating module is used for generating a process flow file for each first electric core in each electric core set according to the first sub-parameter, the second sub-parameter, the target historical data and the sub-standing duration corresponding to each first electric core in the electric core set;
The process flow files comprise characteristic parameters of each first battery cell, wherein the characteristic parameters of each first battery cell are used for predicting the predicted charging duration corresponding to the charging operation of the first battery cell and the first standing duration after the charging operation is performed, or used for predicting the predicted discharging duration corresponding to the discharging operation of the first battery cell and the second standing duration after the discharging operation is performed.
As an optional implementation manner, in the second aspect of the present invention, the determining module is further configured to determine, after determining that the mathematical model completes training, all the cells that are currently required to perform the target prediction operation, and record the determined cells as the second cells;
the acquisition module is further used for acquiring the cell data corresponding to each second cell;
The target prediction operation is used for predicting the predicted charging duration corresponding to the charging operation executed by the battery cell or the predicted discharging duration corresponding to the discharging operation executed by the battery cell; the corresponding electric core data of each second electric core comprises the electric core type of the second electric core, and the corresponding electric core data of each second electric core also comprises at least one data of the electric core weight, initial voltage, internal resistance value, theoretical capacitance, charging times and discharging times of the second electric core;
the apparatus further comprises:
The prediction module is used for inputting the cell data corresponding to all the second cells into the mathematical model and executing the target prediction operation to obtain a predicted duration corresponding to each second cell, wherein the predicted duration comprises the predicted discharge duration or the predicted charge duration;
the second dividing module is used for executing dividing operation on all the second electric cores according to the electric core dividing parameters corresponding to each second electric core and combining the set time length deviation to obtain a plurality of electric core dividing batches;
The battery cell dividing parameters corresponding to each second battery cell comprise the battery cell type corresponding to the second battery cell, the prediction duration and preset operation matters; the preset operation item is used for indicating that the second battery cell currently needs to execute the discharging operation or the charging operation.
As an optional implementation manner, in the second aspect of the present invention, each process flow file includes a plurality of sub-flow files, and each sub-flow file corresponds to one of the first electrical cores;
the first dividing module is further configured to divide, for all the sub-process files in the same electrical core set, all the sub-process files into a process file set corresponding to each deviation interval according to a preset deviation interval, with an accumulated duration value corresponding to each sub-process file as a reference;
The accumulated time length value corresponding to each first battery cell is obtained by calculating the accumulated charging time length and the accumulated discharging time length of the first battery cell;
And the flow file sets are used for determining the predicted data of the battery cells to be analyzed, wherein the battery cells to be analyzed are the battery cells needing to execute the battery cell charging and discharging duration prediction operation, and the predicted data comprise the predicted discharging duration and the predicted charging duration corresponding to the battery cells to be analyzed.
The third aspect of the invention discloses another intelligent prediction device for charging and discharging time of a battery cell formation process, which comprises the following components:
A memory storing executable program code;
A processor coupled to the memory;
the processor calls the executable program codes stored in the memory to execute the intelligent prediction method of the charge and discharge time of the battery cell formation process disclosed in the first aspect of the invention.
In a fourth aspect, the present invention discloses a computer storage medium, where computer instructions are stored, where the computer instructions are used to execute the intelligent prediction method of charge and discharge time of the cell formation process disclosed in the first aspect of the present invention when the computer instructions are called.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
In the embodiment of the invention, an intelligent prediction method for charging and discharging time of a battery cell formation process is provided, and the method comprises the following steps: determining all first electric cores needing to execute preset electric core prediction operation, and acquiring historical data corresponding to each first electric core, wherein the historical data corresponding to each first electric core is used for determining electric core formation data of the first electric core, and the electric core formation data of the first electric core are used for predicting charge and discharge time of the first electric core; according to the historical data corresponding to each first battery cell, determining a plurality of analysis parameters corresponding to each first battery cell, wherein each analysis parameter is a parameter affecting the charge and discharge time length of the first battery cell for executing charge and discharge operation; inputting all historical data into a preset mathematical model by taking all analysis parameters as references so as to train the mathematical model; the mathematical model after training is used for analyzing the battery cell input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell. Therefore, by implementing the invention, the historical data of a plurality of first electric cores can be automatically determined and acquired and used as the basic data for subsequent data analysis, so that a plurality of analysis parameters are determined on the basis of the historical data of each first electric core, the analysis parameters are used as indexes for subsequent analysis of all the historical data, all the historical data are used as training data of a model, and all the historical data are input into a set mathematical model to train the mathematical model; in the mode, the analysis parameters are taken as the data analysis direction, the mathematical model is taken as the training core, and the historical data of all the first electric cores are subjected to data analysis, so that the influence relationship between the historical data of each first electric core and the analysis parameters is analyzed and determined, the historical data of each first electric core is fully utilized, and the data utilization rate of the historical data of the first electric core is improved; then, when the input battery cell is analyzed through the mathematical model which is completed through training, the prediction of the charge and discharge time length corresponding to the battery cell charge and discharge operation can be rapidly and accurately performed according to the influence relation determined by the front part, and the prediction speed and the prediction accuracy of the charge and discharge time length are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow diagram of an intelligent prediction method for charging and discharging time in a cell formation flow disclosed in the embodiment of the invention;
FIG. 2 is a schematic flow chart of another intelligent prediction method for charge and discharge time in a cell formation flow according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an intelligent prediction device for charging and discharging time in a cell formation process according to an embodiment of the present invention;
Fig. 4 is a schematic structural diagram of an intelligent prediction device for charging and discharging time in another cell formation process according to an embodiment of the present invention;
Fig. 5 is a schematic structural diagram of an intelligent prediction device for charging and discharging time in a cell formation process according to another embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or article that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or article.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The invention discloses an intelligent prediction method and device for charge and discharge time of a battery cell formation process, which can automatically determine and acquire historical data of a plurality of first battery cells as basic data for subsequent data analysis, so that a plurality of analysis parameters are determined on the basis of the historical data of each first battery cell, the analysis parameters are used as indexes for subsequent analysis of all the historical data, all the historical data are used as training data of a model, and all the historical data are input into a set mathematical model to train the mathematical model; in the mode, the analysis parameters are taken as the data analysis direction, the mathematical model is taken as the training core, and the historical data of all the first electric cores are subjected to data analysis, so that the influence relationship between the historical data of each first electric core and the analysis parameters is analyzed and determined, the historical data of each first electric core is fully utilized, and the data utilization rate of the historical data of the first electric core is improved; then, when the input battery cell is analyzed through the mathematical model which is completed through training, the prediction of the charge and discharge time length corresponding to the battery cell charge and discharge operation can be rapidly and accurately performed according to the influence relation determined by the front part, and the prediction speed and the prediction accuracy of the charge and discharge time length are improved. The following will describe in detail.
Example 1
Referring to fig. 1, fig. 1 is a flow chart of an intelligent prediction method for charging and discharging time in a cell formation flow according to an embodiment of the invention. The method for intelligently predicting the charge and discharge time of the cell formation process described in fig. 1 can be applied to an intelligent prediction device for the charge and discharge time of the cell formation process, and the embodiment of the invention is not limited. As shown in fig. 1, the intelligent prediction method of the charge and discharge time of the battery cell formation process may include the following operations:
101. Determining all first electric cores needing to execute preset electric core prediction operation, and acquiring historical data corresponding to each first electric core, wherein the historical data corresponding to each first electric core is used for determining electric core formation data of the first electric core, and the electric core formation data of the first electric core are used for predicting charge and discharge time of the first electric core.
102. And determining a plurality of analysis parameters corresponding to each first battery cell according to the historical data corresponding to each first battery cell, wherein each analysis parameter is a parameter affecting the charge and discharge time length of the first battery cell for executing the charge and discharge operation.
103. And inputting all historical data into a preset mathematical model by taking all analysis parameters as references so as to train the mathematical model.
In the embodiment of the invention, the all historical data can comprise the historical data corresponding to a plurality of parameters such as the weight, the initial voltage, the internal resistance value, the theoretical capacitance and the like of the battery core corresponding to each first battery core, and also comprise the single-time charging duration and the single-time discharging duration which are recorded corresponding to each first battery core.
In the embodiment of the invention, the mathematical model is subsequently used for predicting the charge and discharge time length of the battery cell, namely the mathematical model is a prediction model of the charge and discharge time length of the battery cell; then, when step 103 is performed, after the above-mentioned history data is input as model training data into a mathematical model, the output result corresponding to the mathematical model includes the predicted charge duration and/or the predicted discharge duration corresponding to each cell.
In the embodiment of the invention, in the training process of a mathematical model, the judgment of the model training progress is involved, and at the moment, the difference value between the predicted charging time length and the corresponding recorded single charging time length and the difference value between the predicted discharging time length and the corresponding recorded single discharging time length are calculated through a set loss function; taking the two difference values as a reference, and judging the model training progress; further, when the two differences reach a preset difference threshold (e.g., reach a minimum value), the model is determined to complete training.
In the embodiment of the invention, the mathematical model which is trained is used for analyzing the battery cell which is input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell.
Therefore, by implementing the intelligent prediction method of the charge and discharge time of the cell formation process described in fig. 1, the historical data of a plurality of first cells can be automatically determined and obtained to be used as the basic data for subsequent data analysis, so that a plurality of analysis parameters are determined based on the historical data of each first cell, the analysis parameters are used as indexes for subsequent analysis of all the historical data, all the historical data are used as training data of a model, and all the historical data are input into a set mathematical model to train the mathematical model; in the mode, the analysis parameters are taken as the data analysis direction, the mathematical model is taken as the training core, and the historical data of all the first electric cores are subjected to data analysis, so that the influence relationship between the historical data of each first electric core and the analysis parameters is analyzed and determined, the historical data of each first electric core is fully utilized, and the data utilization rate of the historical data of the first electric core is improved; then, when the input battery cell is analyzed through the mathematical model which is completed through training, the prediction of the charge and discharge time length corresponding to the battery cell charge and discharge operation can be rapidly and accurately performed according to the influence relation determined by the front part, and the prediction speed and the prediction accuracy of the charge and discharge time length are improved.
In an alternative embodiment, all analysis parameters include at least one analysis parameter of the cell weight, the initial voltage, the internal resistance value and the theoretical capacitance of each first cell, and are recorded as first sub-parameters;
After determining the plurality of analysis parameters corresponding to each first cell according to the historical data corresponding to each first cell in step 102, the method further includes:
Dividing all the first electric cores into a plurality of electric core sets by taking the electric core identifier corresponding to each first electric core as a dividing reference, wherein each electric core set is matched with one electric core identifier, and the electric core identifier corresponding to each first electric core is used for indicating the model and/or the type of the first electric core;
And for each first cell in each cell set, extracting historical data matched with each first sub-parameter from the historical data corresponding to the first cell by taking the first sub-parameter corresponding to the first cell as a reference, recording the historical data as target historical data, and executing data association or data filling on the first cell and the corresponding target historical data so as to update all cell sets.
In this alternative embodiment, the cell weight, the initial voltage, the internal resistance value, and the theoretical capacitance of each first cell mentioned above are parameters that have an associated relationship with the charge duration/discharge duration corresponding to the first cell performing the charge/discharge operation.
In this alternative embodiment, after a large number of history data of the first battery cells are acquired, in order to improve the convenience of later calling the history data, a previous data processing needs to be performed on all the history data; at this time, the first cells are preferably divided by taking the cell identifier (cell model/cell type) as a division reference, that is, the first-level division is performed on all the first cells based on the cell identifier; then, on the basis of the determined first sub-parameters, extracting the history data matched with each first sub-parameter from the complex history data recorded by each first cell, and simultaneously executing data association or data filling operation to update all cell sets (actually updating each first cell and the corresponding target history data); that is, the first sub-parameters are used as the reference to conduct the second-level data arrangement on all the first battery cells and the corresponding historical data.
In the optional embodiment, a processing scheme for extracting the target historical data corresponding to each first battery cell is set, and the first-level division is performed on all the first battery cells through the battery cell identifiers, so that the data record simplicity of all the first battery cells is primarily improved; and then, carrying out second-level data arrangement on all the first electric cores and the corresponding historical data thereof through the first sub-parameters, thereby further improving the data recording accuracy, the detail degree and the data conciseness of each first electric core and the corresponding target historical data thereof.
In another optional embodiment, all analysis parameters further include a single-time charging duration of the battery cell of each first battery cell, a number of times of charging corresponding to the single-time charging duration of the battery cell, a single-time discharging duration of the battery cell, and a number of times of discharging corresponding to the single-time discharging duration of the battery cell, which are recorded as second sub-parameters;
Inputting all historical data into a preset mathematical model by taking all analysis parameters as references to train the mathematical model, wherein the method comprises the following steps of:
inputting target historical data corresponding to each first battery cell into a preset mathematical model, taking a first sub-parameter corresponding to the first battery cell as an independent variable, and taking a second sub-parameter corresponding to the first battery cell as a dependent variable, and executing model training operation on the mathematical model to obtain a training result corresponding to the mathematical model;
after determining that model training is performed on all the first sub-parameters and all the second sub-parameters by the mathematical model, it is determined that model training operations for the mathematical model are completed.
In this alternative embodiment, the mathematical model may be a linear regression model in actual use.
In this optional embodiment, the analysis parameter further includes a plurality of second sub-parameters for training the mathematical model, and by using the plurality of second sub-parameters and the first sub-parameter corresponding to the first cell as independent variables, an initial voltage value, an initial internal resistance value, a theoretical capacitance, and a theoretical capacitance corresponding to the first cell can be recorded as independent variables x of the mathematical model; taking a second sub-parameter corresponding to the second battery cell as a dependent variable; in practical training, the process of charging and discharging the first battery cell is divided into two processes, namely, the single-time charging duration of the battery cell corresponding to the first battery cell is recorded as a first independent variable y_charge_n, and the single-time discharging duration of the battery cell corresponding to the first battery cell is recorded as a second independent variable y_discharge_n. Wherein n represents the single charging time length of each battery cell, the charging times and discharging times corresponding to the single discharging time length of the battery cell; or the n may also be used to represent the sequence, for example, y_charge_1, y_charge_2 to y_charge_n represent the single charging duration of the first, second to nth first cells.
In this alternative embodiment, taking the example of performing the charging operation on the battery cell, the mathematical model training and the calculation formula adopted in the actual application after the training is completed are as follows:
ypredict_charge_n= x·w + [1,1,1,1,1]·w0
argmin f(w)=1/N∑( ypredict- yi)2
In this alternative embodiment, it should be noted that, the y predict_charge_n is the charging duration required by the nth charging of the current battery cell to be predicted; x is the data corresponding to the first sub-parameter of the current cell to be predicted; w and w 0 are predetermined matrices, wherein w and w 0 are used as model parameters of a mathematical model, and are obtained by analyzing the above-mentioned historical data; the argmin f (w) function is a minimum value determining function and is used for judging whether the model converges or not in the model training process; y i is used for indicating the actual charging duration corresponding to the ith battery cell (for example, the predicted charging duration y predict_charge_2 of the 2 nd battery cell is calculated through a first formula; meanwhile, the actual charging duration y 2 of the 2 nd battery cell is actually collected); n represents the number of all cells.
In this optional embodiment, optionally, y i may be further configured to indicate an actual charging duration corresponding to the n-th charging operation performed on the same battery cell (for example, a predicted charging duration corresponding to the second charging operation performed on the battery cell a is calculated by using the first formula and is y predict_charge_2; and meanwhile, an actual charging duration y 2 corresponding to the second charging operation performed on the battery cell a is actually collected); at this time, N represents the number of charging operations corresponding to the current statistics of the battery cell to perform the charging operation.
In this alternative embodiment, the calculation of the loss function value is repeated before the mathematical model is not trained, and if the loss function value is greater than the preset loss function threshold or the minimum loss function value is not reached, the model is represented as not being trained to be converged, and the model parameters of the mathematical model are further updated according to the calculated loss function value. For the calculation formula corresponding to argmin f (w), argmin f (w) is specifically used for determining the value corresponding to w when f (w) reaches the minimum value, where the minimum value of f (w) indicates that the loss function value is the minimum, which means that the mathematical model is trained to be converged, and the value corresponding to w when f (w) reaches the minimum value is the model parameter corresponding to the mathematical model that completes the training.
In this alternative embodiment, when the mathematical model determines that training is completed and the mathematical model is actually used to predict the battery charging duration, w and w 0 are matrix parameters that match the second sub-parameters of the first battery to be analyzed; for example, historical data of the cell type a is analyzed in advance to obtain matrix parameters w and w 0 corresponding to the cell type a.
In this alternative embodiment, after the first sub-parameter and the second sub-parameter are input into the mathematical model, a model that can predict the charge/discharge duration required by the nth charge/discharge of different cells (i.e., the cell type/happy difference) is obtained through training calculation of the mathematical model.
In this alternative embodiment, the training process of the set mathematical model is described below:
After the historical data of all the first battery cells are obtained, extracting initial voltage values, initial current values, first battery cell weights, theoretical capacitance and first battery cell types corresponding to each first battery cell from the historical data of the same type (such as ternary lithium battery cells), wherein the 5 data serve as influence factors for influencing the charge and discharge time of the first battery cells in the analysis; simultaneously extracting the actual charge time length and the actual discharge time length of each first battery cell; and then analyzing and obtaining the corresponding association coefficient between the actual charging time length, the actual discharging time length and different influence factors of each first battery cell, and further illustrating the following steps:
Taking the discharge operation of a ternary lithium battery cell as an example, for a certain ternary lithium battery cell, determining that when the discharge operation is performed on the ternary lithium battery cell from historical data of the ternary lithium battery cell, the corresponding initial voltage value is 3.6v, the initial current value (actually the discharge current) is 1C, the weight of the battery cell is 45g, the theoretical capacitance is 3600mAh, and recording that the discharge duration of the ternary lithium battery cell under the influence factors is 1 hour 45s, wherein the 1 hour 45s corresponds to 3645s; setting the corresponding battery cell type of the ternary lithium battery cell as 1; that is, x= {3.6,1, 45, 360,1}, the argument x is a 5-dimensional array composed of 5 influencing factors; simultaneously recording the obtained actual discharge duration y_discharge_1= {3645}; and by analogy, taking historical data corresponding to the influence factors as an independent variable x, taking the discharge time length actually recorded as an independent variable y_discharge_n, and combining a preset data model (adopting an existing linear regression model), so that the linear relation between x and y_discharge_n can be obtained by analysis.
In this alternative embodiment, by analyzing the historical data corresponding to the cell type, the values of the parameters w and w 0 for indicating the linear relationship between the arguments x and y _ discharge _ n for that cell type can be calculated, for example, w is calculated to be [1,1 ]: w 0 is [1, 1].
Wherein, when the charging operation is performed on the battery cell, the calculation formula for calculating the predicted charging time length of the battery cell by the mathematical model (linear regression model) is as follows:
ypredict_charge_n= xcharge_n·wcharge_n+ [1,1,1,1,1]·w0
Wherein x charge_n represents the data of the input of the battery cell (the data of the input of the battery cell corresponds to the data of the first sub-parameter) which needs to be analyzed at present for the charging time of the battery cell; w charge_n is an influence parameter matched with the cell type of the cell;
Similarly, when a discharge operation is performed on the cell, the calculation formula for calculating the predicted discharge time length of the cell by the mathematical model (linear regression model) is as follows:
ypredict_discharge_n= xdischarge_n·wdischarge_n+ [1,1,1,1,1]·w0
Wherein x discharge_n represents the data of the input of the battery cell (the data of the input of the battery cell and the data corresponding to the first sub-parameter) which is needed to be analyzed at present; w discharge_n is an influencing parameter matched with the cell type of the cell.
In this alternative embodiment, as mentioned above, the meaning of n in charge_n and discharge_n may actually refer to the nth cell, such as the 2 nd cell in a whole batch of cells; n may also refer to the number of charging/discharging operations corresponding to the same cell, for example, the charging 2 nd time or the discharging 2 nd time, where the actual meaning of n is adjusted according to the actual application requirement, and the embodiment of the present invention is not limited.
Taking the predicted discharge time length of the battery cell as an example, substituting the data represented by x= {3.6,1, 45, 3600,1} recorded by the currently analyzed ternary lithium battery cell into the calculation formula corresponding to the execution discharge operation, and calculating to obtain the predicted discharge time length of the battery cell for the ternary lithium battery cell as 3650.6, wherein the discharge time of the actual battery cell is 3645s, and the time difference of the predicted discharge time length and the actual discharge time length is 4.6s, and the time difference is within the set time length difference range.
In this alternative embodiment, it should be noted that, for the sake of convenience in calculation, the input x data may be directly acquired historical data as illustrated above, or may be data obtained by performing numerical conversion on the originally acquired historical data, which is not limited in the embodiment of the present invention.
In this alternative embodiment, for the cells of the same cell type (when the model training is actually performed, the historical data corresponding to the cells of a whole batch is input), after the linear regression model is input, a plurality of cell prediction charging durations (y_charge_1, y_charge_2,.+ -., y_charge_n) and cell prediction discharging durations (y_discharge_1, y_discharge_2,.+ -., y_discharge_n) of the first cells are obtained; meanwhile, during the model training of the linear regression model, the loss calculation of the model is involved, and a plurality of loss calculation values are obtained; at this time, selecting one of the multiple loss calculation values with the smallest value, and recording a corresponding model with the smallest value to complete training; the calculation formula for selecting the minimum loss value from the plurality of loss calculation values is as follows:
argmin f(w)=1/N∑( ypredict- yi)2
Wherein, the y predict refers to a predicted battery cell duration, which may be a predicted battery cell charging duration or a predicted battery cell discharging duration; y i refers to the i-th cell actual charging duration or the cell actual discharging duration, where N is the number of cells; or y i refers to the same actual charging duration or the same actual discharging duration of the battery cell corresponding to the ith charging or discharging, and at this time, N is the charging frequency or the discharging frequency of the battery cell; the formula right side 1/N sigma (y predict- yi)2 is a loss calculation formula adopted by a training mathematical model; the formula left side argmin f (w) is used for determining the value corresponding to the model parameter w when the value f (w) with the smallest loss value is selected from a plurality of corresponding calculated loss values aiming at the same batch of electric cores (different electric core types and different influence factors are mentioned above and corresponding to the calculated w is different).
In this alternative embodiment, it is noted that, for a ternary lithium cell, the capacitance is 3600mAh, after analyzing the historical data corresponding to the type of cell, the calculated influence coefficient w= [1,1 ]: w 0 = [1, 1]; for a ternary lithium battery core, the capacitance is 1800mAh, and the influence coefficients obtained through corresponding calculation can be w= [1,1,1,2,1] and w 0 = [1, 1]; that is, the actual values of w and w 0 need to be adjusted according to the actual calculation result, and w 0 are not fixed values.
In the optional embodiment, training analysis can be performed on the first sub-parameters and the second sub-parameters input for all the first battery cells through the set mathematical model, so that the analysis accuracy of relevant information of all the first battery cells is improved; the accurate mathematical model after the training is completed can be used for accurately predicting the charge and discharge time length of the battery cell, and the accuracy and the prediction efficiency for performing the charge and discharge time length prediction of the battery cell through the mathematical model are improved.
In yet another optional embodiment, each target history data has a corresponding data tag, where the data tag is used to indicate that the target history data is recorded after the first cell performs the charging operation or the discharging operation;
before the target historical data corresponding to each first cell is input into the preset mathematical model, the method further includes:
According to the data labels corresponding to each target historical data and the matching parameters in each second sub-parameter, matching operation is carried out on all the historical data and all the second sub-parameters, and a plurality of paired data are obtained;
updating all target historical data according to all the paired data;
the matching parameters comprise the charging times and the discharging times; each pairing data includes a project label history data and a second sub-parameter.
In this alternative embodiment, since the mathematical model of the setting starts from both charging and discharging directions when analyzing data, and further, when actually analyzing data, a data matching operation is set in advance, taking the data tag as a reference, all data belonging to the charging operation are integrated, all data belonging to the discharging operation are integrated, so as to obtain two distinct batches of data (a plurality of paired data), and in each batch of data, specifically, one batch of data with the data tag being the charging operation further includes a plurality of paired data, and each paired data is integrated and recorded by taking the matching parameter in the second sub-parameter as a reference. Specifically, for example, in the batch data of the charging operation, the number of times of charging is 10 (this may refer to the 10 th time of charging performed in the existing record, or may be 10 times of accumulated charging recorded), the plurality of paired data records have the first battery cell A, B, C, D, and the duration of single charging of the battery cell corresponding to the number of times of charging is 10 times is the same, and the data of the number of times of discharging is not described herein again. Specific pairing modes the embodiments of the present invention are not limited.
Therefore, in the optional embodiment, a data matching scheme based on the data tag is set, so that the integration fineness and accuracy of each first battery cell and corresponding target historical data are further improved, the training speed of the model during subsequent input of the mathematical model is improved to a certain extent, and the probability of model training errors is reduced.
Example two
Referring to fig. 2, fig. 2 is a flow chart of another intelligent prediction method for charge and discharge time in a cell formation flow according to an embodiment of the invention. The method for intelligently predicting the charge and discharge time of the cell formation process described in fig. 2 can be applied to an intelligent prediction device for the charge and discharge time of the cell formation process, and the embodiment of the invention is not limited. As shown in fig. 2, the intelligent prediction method of the charge and discharge time of the cell formation process may include the following operations:
201. Determining all first electric cores needing to execute preset electric core prediction operation, and acquiring historical data corresponding to each first electric core, wherein the historical data corresponding to each first electric core is used for determining electric core formation data of the first electric core, and the electric core formation data of the first electric core are used for predicting charge and discharge time of the first electric core.
202. And determining a plurality of analysis parameters corresponding to each first battery cell according to the historical data corresponding to each first battery cell, wherein each analysis parameter is a parameter affecting the charge and discharge time length of the first battery cell for executing the charge and discharge operation.
203. And inputting all historical data into a preset mathematical model by taking all analysis parameters as references so as to train the mathematical model.
In the embodiment of the present invention, for other descriptions of step 201 to step 203, please refer to other specific descriptions of step 101 to step 103 in the first embodiment, and the description of the embodiment of the present invention is omitted.
204. After the mathematical model is determined to complete training, the standing data corresponding to each first cell is determined from the historical data corresponding to each first cell.
In the embodiment of the invention, the standing data corresponding to each first battery cell comprises a plurality of sub-standing time periods, and each sub-standing time period is the time period of the first battery cell after the first battery cell performs the charging operation or the discharging operation;
205. and for all the first electric cores in each electric core set, generating a process flow file aiming at the electric core set according to the first sub-parameters, the second sub-parameters, the target historical data and the sub-standing time length corresponding to each first electric core in the electric core set.
In the embodiment of the invention, all the process flow files comprise characteristic parameters aiming at each first battery cell, wherein the characteristic parameters of each first battery cell are used for predicting the predicted charging duration corresponding to the charging operation of the first battery cell and the first standing duration after the charging operation is performed, or are used for predicting the predicted discharging duration corresponding to the discharging operation of the first battery cell and the second standing duration after the discharging operation is performed.
Therefore, by implementing the intelligent prediction method of the charge and discharge time of the cell formation process described in fig. 2, the process flow file for each cell set can be generated by combining the determined standing data of each first cell and taking the cell set as a unit; the generated process flow file comprises characteristic parameters for each first battery cell, so that when the battery cell charge and discharge duration prediction needs to be executed subsequently, the battery cell charge and discharge duration can be determined rapidly only by calling the process flow file to pair the battery cells, and the speed and accuracy of predicting the battery cell charge and discharge duration are improved; in addition, the process flow file is generated, the setting of the standing time length after the battery cell performs the charge/discharge operation is also involved, the matched standing time length is directly used in the follow-up process, and the convenience and accuracy of setting the charge/discharge time and the standing time corresponding to the charge/discharge operation performed on the battery cell are improved.
In an alternative embodiment, after determining that the mathematical model has completed training, the method further comprises:
Determining all the current battery cells needing to execute target prediction operation, marking the current battery cells as second battery cells, and acquiring battery cell data corresponding to each second battery cell;
The target prediction operation is used for predicting the predicted charging duration corresponding to the charging operation executed by the battery cell or the predicted discharging duration corresponding to the discharging operation executed by the battery cell; the corresponding battery cell data of each second battery cell comprises the battery cell type of the second battery cell, and the corresponding battery cell data of each second battery cell also comprises at least one data of the battery cell weight, initial voltage, internal resistance value, theoretical capacitance, charging times and discharging times of the second battery cell;
Inputting the electric core data corresponding to all the second electric cores into a mathematical model and executing target prediction operation to obtain predicted duration corresponding to each second electric core, wherein the predicted duration comprises predicted discharge duration or predicted charge duration;
According to the corresponding cell dividing parameters of each second cell, dividing all the second cells by combining the set time length deviation to obtain a plurality of cell dividing batches;
the battery cell dividing parameters corresponding to each second battery cell comprise the battery cell type, the predicted duration and the preset operation items corresponding to the second battery cell; the preset operation item is used for indicating that the second battery cell is required to execute a discharging operation or a charging operation currently.
In this optional embodiment, the performing a dividing operation on all the second cells according to the cell dividing parameter corresponding to each second cell and the set duration deviation to obtain a plurality of cell dividing batches includes:
When the set time deviation is a plurality of time intervals, dividing all the second cells by taking the type of the cell corresponding to each second cell and preset operation matters as a first hierarchical division standard to obtain a plurality of first cell batches;
And carrying out division on each first cell batch by taking the predicted time length corresponding to each second cell and the time length intervals as a second hierarchical division standard to obtain a division result corresponding to each first cell batch, wherein the division result corresponding to each first cell batch comprises a plurality of second cell batches serving as cell division batches.
In this optional embodiment, the foregoing dividing operation is performed on all the second electrical cores according to the electrical core dividing parameter corresponding to each second electrical core in combination with the set duration deviation to obtain a plurality of electrical core dividing batches, and in practical application, the dividing may refer to dividing all the second electrical cores with the same electrical core type and the same preset operation item, and then subdividing all the second electrical cores with predicted durations within the same interval threshold value or within the set duration deviation. Specifically, the first level of division can be performed on all ternary lithium battery cores and phosphoric acid ferroelectric cores needing to be charged, and further, the ternary lithium battery cores and phosphoric acid ferroelectric cores which are charged for a period of time of [30, 60 ] minutes and [60, 90) minutes are further subdivided according to the set period deviation; or taking the deviation of the total charging time length not exceeding 5% as a dividing standard, assuming that 5 cells with the total charging time lengths of 290, 295, 300, 305 and 310 exist in all ternary lithium cells, taking 300 as an average value, the deviation of the total charging time length of the 5 cells not exceeding 5% can be divided into a batch, and the actual cell batch dividing mode is not limited in the embodiment of the invention.
It can be seen that in this optional embodiment, after the trained mathematical model is obtained, the second electrical core to be subjected to the target prediction operation and the electrical core data of each second electrical core can be automatically determined, and then the target prediction operation is performed based on the mathematical model, so that the degree of intellectualization of performing the target prediction operation for the second electrical core, the operation efficiency, and the accuracy of the obtained prediction duration are improved; then, batch division can be carried out on all the second electric cores by combining the set electric core division parameters and the duration deviation, and when the charging/discharging operation is actually carried out on the second electric cores, all the second electric cores in the unified batch of operation cannot have the condition of overlarge charging/discharging duration deviation, so that the condition of reducing the quality of the second electric cores occurs; namely, the accurate batch division can lead the formation time length of the same batch to be consistent, the waiting time of the same batch of electric cores can not be wasted, and the production speed of the production line is higher, thereby improving the productivity, and finally being beneficial to reducing the loss of the second electric core and improving the production quality of the second electric core; in addition, for the battery cell with new specification, the scheme can be used for carrying out the prefabrication process flow, so that the trial-and-error cost of a new product is reduced.
In another alternative embodiment, each process flow file includes a plurality of sub-flow files, each sub-flow file corresponding to a first cell; the method further comprises the steps of:
For all the sub-process files in the same cell set, dividing all the sub-process files into a process file set corresponding to each deviation interval according to a preset deviation interval by taking the accumulated duration value corresponding to each sub-process file as a reference;
The accumulated time length value corresponding to each first battery cell is obtained by calculating the accumulated charging time length and the accumulated discharging time length of the first battery cell;
all the flow file sets are used for determining prediction data of the battery cell to be analyzed, wherein the battery cell to be analyzed is a battery cell needing to execute battery cell charging and discharging duration prediction operation, and the prediction data comprise predicted discharging duration and predicted charging duration corresponding to the battery cell to be analyzed.
In this optional embodiment, after the process flow file for each cell set is generated, an integration scheme for all the sub-flow files in each process flow file is further set, and all the sub-flow files are stored in a divided manner based on the deviation interval, so that when the flow file is subsequently called, the flow file set can be used as a guide to call, the problem that the difficulty and the complexity of data call are high due to scattered recording of the sub-flow files is solved, and the convenience in the subsequent operation of executing the target prediction operation (the prediction of the charge/discharge duration and the determination of the static duration of the cell) on the cell through the flow file set is further improved.
Example III
Referring to fig. 3, fig. 3 is a schematic structural diagram of an intelligent prediction device for charging and discharging time in a cell formation process according to an embodiment of the present invention. The intelligent prediction device of the charge and discharge time of the battery cell formation process can be an intelligent prediction terminal, equipment, a system or a server of the charge and discharge time of the battery cell formation process, the server can be a local server, a remote server or a cloud server (also called cloud server), and when the server is a non-cloud server, the non-cloud server can be in communication connection with the cloud server, and the embodiment of the invention is not limited. As shown in fig. 3, the intelligent prediction apparatus for charging and discharging time of a cell formation process may include a determining module 301, an obtaining module 302, and a training module 303, where:
The determining module 301 is configured to determine all first cells that need to perform a preset cell prediction operation.
The obtaining module 302 is configured to obtain historical data corresponding to each first electrical core, where the historical data corresponding to each first electrical core is used to determine electrical core formation data of the first electrical core, and the electrical core formation data of the first electrical core is used to predict a charge-discharge duration of the first electrical core.
The determining module 301 is further configured to determine, according to the historical data corresponding to each first electrical cell, a plurality of analysis parameters corresponding to each first electrical cell, where each analysis parameter is a parameter affecting a charge-discharge duration of the first electrical cell for performing the charge-discharge operation.
The training module 303 is configured to input all the historical data into a preset mathematical model based on all the analysis parameters, so as to train the mathematical model.
The mathematical model after training is used for analyzing the battery cell input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell.
Therefore, the intelligent prediction device for implementing the battery cell formation flow charge-discharge time described in fig. 3 can automatically determine and acquire the historical data of a plurality of first battery cells as the basic data for subsequent data analysis, so that a plurality of analysis parameters are determined based on the historical data of each first battery cell, the analysis parameters are used as indexes for subsequent analysis of all the historical data, all the historical data are used as training data of a model, and all the historical data are input into a set mathematical model to train the mathematical model; in the mode, the analysis parameters are taken as the data analysis direction, the mathematical model is taken as the training core, and the historical data of all the first electric cores are subjected to data analysis, so that the influence relationship between the historical data of each first electric core and the analysis parameters is analyzed and determined, the historical data of each first electric core is fully utilized, and the data utilization rate of the historical data of the first electric core is improved; then, when the input battery cell is analyzed through the mathematical model which is completed through training, the prediction of the charge and discharge time length corresponding to the battery cell charge and discharge operation can be rapidly and accurately performed according to the influence relation determined by the front part, and the prediction speed and the prediction accuracy of the charge and discharge time length are improved.
In an alternative embodiment, all analysis parameters include at least one analysis parameter of the cell weight, the initial voltage, the internal resistance value and the theoretical capacitance of each first cell, and are recorded as first sub-parameters;
referring to fig. 4, fig. 4 is a schematic structural diagram of an intelligent prediction device for charging and discharging time in another cell formation process according to an embodiment of the invention. As shown in fig. 4, the apparatus further includes a first partitioning module 304, an extracting module 305, and an association filling module 306, where:
The first dividing module 304 is configured to divide, after the determining module 301 determines a plurality of analysis parameters corresponding to each first electrical core according to historical data corresponding to each first electrical core, all the first electrical cores into a plurality of electrical core sets with a corresponding electrical core identifier of each first electrical core as a dividing reference, each electrical core set matching one electrical core identifier, and each electrical core identifier corresponding to each first electrical core being used for indicating a model and/or a type of the first electrical core;
The extracting module 305 is configured to extract, for each first cell in each cell set, historical data matched with each first sub-parameter from the historical data corresponding to the first cell based on the first sub-parameter corresponding to the first cell, and record the historical data as target historical data;
and the association filling module 306 is configured to perform data association or data filling on the first battery cell and the corresponding target history data thereof so as to update all battery cell sets.
In the optional embodiment, a processing scheme for extracting the target historical data corresponding to each first battery cell is set, and the first-level division is performed on all the first battery cells through the battery cell identifiers, so that the data record simplicity of all the first battery cells is primarily improved; and then, carrying out second-level data arrangement on all the first electric cores and the corresponding historical data thereof through the first sub-parameters, thereby further improving the data recording accuracy, the detail degree and the data conciseness of each first electric core and the corresponding target historical data thereof.
In another optional embodiment, all analysis parameters further include a single-time charging duration of the battery cell of each first battery cell, a number of times of charging corresponding to the single-time charging duration of the battery cell, a single-time discharging duration of the battery cell, and a number of times of discharging corresponding to the single-time discharging duration of the battery cell, which are recorded as second sub-parameters;
Optionally, the training module 303 inputs all the historical data into a preset mathematical model based on all the analysis parameters, so as to train the mathematical model specifically including:
inputting target historical data corresponding to each first battery cell into a preset mathematical model, taking a first sub-parameter corresponding to the first battery cell as an independent variable, and taking a second sub-parameter corresponding to the first battery cell as a dependent variable, and executing model training operation on the mathematical model to obtain a training result corresponding to the mathematical model;
after determining that model training is performed on all the first sub-parameters and all the second sub-parameters by the mathematical model, it is determined that model training operations for the mathematical model are completed.
In the optional embodiment, training analysis can be performed on the first sub-parameters and the second sub-parameters input for all the first battery cells through the set mathematical model, so that the analysis accuracy of relevant information of all the first battery cells is improved; the accurate mathematical model after the training is completed can be used for accurately predicting the charge and discharge time length of the battery cell, and the accuracy and the prediction efficiency for performing the charge and discharge time length prediction of the battery cell through the mathematical model are improved.
In yet another optional embodiment, each target history data has a corresponding data tag, where the data tag is used to indicate that the target history data is recorded after the first cell performs the charging operation or the discharging operation;
the training module 303 inputs all the historical data into a preset mathematical model based on all the analysis parameters, and the method for training the mathematical model specifically further includes:
Before inputting target historical data corresponding to each first battery cell into a preset mathematical model, matching all historical data with all second sub-parameters according to a data tag corresponding to each target historical data and combining the matching parameters in each second sub-parameter to obtain a plurality of paired data;
updating all target historical data according to all the paired data;
the matching parameters comprise the charging times and the discharging times; each pairing data includes a project label history data and a second sub-parameter.
Therefore, in the optional embodiment, a data matching scheme based on the data tag is set, so that the integration fineness and accuracy of each first battery cell and corresponding target historical data are further improved, the training speed of the model during subsequent input of the mathematical model is improved to a certain extent, and the probability of model training errors is reduced.
In another optional embodiment, the determining module 301 is further configured to determine, after the training of the mathematical model is determined, from the historical data corresponding to each first electrical core, standing data corresponding to each first electrical core, where each standing data corresponding to each first electrical core includes a plurality of sub-standing periods, and each sub-standing period is a period of time for which the first electrical core is standing after performing the charging operation or the discharging operation;
as shown in fig. 4, the apparatus further comprises a generating module 307, wherein:
The generating module 307 is configured to generate, for all the first electrical cores in each electrical core set, a process flow file for the electrical core set according to the first sub-parameter, the second sub-parameter, the target history data, and the sub-standing duration corresponding to each first electrical core in the electrical core set;
The process flow files comprise characteristic parameters of each first battery cell, wherein the characteristic parameters of each first battery cell are used for predicting the predicted charging duration corresponding to the charging operation of the first battery cell and the first standing duration after the charging operation is performed, or used for predicting the predicted discharging duration corresponding to the discharging operation of the first battery cell and the second standing duration after the discharging operation is performed.
It can be seen that, in this alternative embodiment, the determined standing data of each first electrical core can be combined, and the process flow file for each electrical core set can be generated by taking the electrical core set as a unit; the generated process flow file comprises characteristic parameters for each first battery cell, so that when the battery cell charge and discharge duration prediction needs to be executed subsequently, the battery cell charge and discharge duration can be determined rapidly only by calling the process flow file to pair the battery cells, and the speed and accuracy of predicting the battery cell charge and discharge duration are improved; in addition, the process flow file is generated, the setting of the standing time length after the battery cell performs the charge/discharge operation is also involved, the matched standing time length is directly used in the follow-up process, and the convenience and accuracy of setting the charge/discharge time and the standing time corresponding to the charge/discharge operation performed on the battery cell are improved.
In yet another alternative embodiment, the determining module 301 is further configured to determine, after determining that the mathematical model has completed training, all the cells currently required to perform the target prediction operation, and record the cells as the second cells;
The acquiring module 302 is further configured to acquire cell data corresponding to each second cell;
The target prediction operation is used for predicting the predicted charging duration corresponding to the charging operation executed by the battery cell or the predicted discharging duration corresponding to the discharging operation executed by the battery cell; the corresponding battery cell data of each second battery cell comprises the battery cell type of the second battery cell, and the corresponding battery cell data of each second battery cell also comprises at least one data of the battery cell weight, initial voltage, internal resistance value, theoretical capacitance, charging times and discharging times of the second battery cell;
as shown in fig. 4, the apparatus further includes a prediction module 308 and a second partitioning module 309, where:
the prediction module 308 is configured to input the cell data corresponding to all the second cells into a mathematical model and perform a target prediction operation, so as to obtain a predicted duration corresponding to each second cell, where the predicted duration includes a predicted discharge duration or a predicted charge duration;
the second dividing module 309 is configured to perform a dividing operation on all the second electrical cores according to the electrical core dividing parameter corresponding to each second electrical core, and combine the set duration deviation, so as to obtain a plurality of electrical core dividing batches;
the battery cell dividing parameters corresponding to each second battery cell comprise the battery cell type, the predicted duration and the preset operation items corresponding to the second battery cell; the preset operation item is used for indicating that the second battery cell is required to execute a discharging operation or a charging operation currently.
It can be seen that in this optional embodiment, after the trained mathematical model is obtained, the second electrical core to be subjected to the target prediction operation and the electrical core data of each second electrical core can be automatically determined, and then the target prediction operation is performed based on the mathematical model, so that the degree of intellectualization of performing the target prediction operation for the second electrical core, the operation efficiency, and the accuracy of the obtained prediction duration are improved; then, batch division can be carried out on all the second electric cores by combining the set electric core division parameters and the duration deviation, and when the charging/discharging operation is actually carried out on the second electric cores, all the second electric cores in the unified batch of operation cannot have the condition of overlarge charging/discharging duration deviation, so that the quality of the second electric cores is reduced; that is, the accurate batch division is finally beneficial to reducing the loss of the second battery cell and improving the production quality of the second battery cell.
In another alternative embodiment, each process flow file includes a plurality of sub-flow files, each sub-flow file corresponding to a first cell;
The first dividing module 304 is further configured to divide, for all the sub-process files in the same electrical core set, all the sub-process files into a process file set corresponding to each deviation interval according to a preset deviation interval and based on an accumulated duration value corresponding to each sub-process file;
The accumulated time length value corresponding to each first battery cell is obtained by calculating the accumulated charging time length and the accumulated discharging time length of the first battery cell;
all the flow file sets are used for determining prediction data of the battery cell to be analyzed, wherein the battery cell to be analyzed is a battery cell needing to execute battery cell charging and discharging duration prediction operation, and the prediction data comprise predicted discharging duration and predicted charging duration corresponding to the battery cell to be analyzed.
In this optional embodiment, after the process flow file for each cell set is generated, an integration scheme for all the sub-flow files in each process flow file is further set, and all the sub-flow files are stored in a divided manner based on the deviation interval, so that when the flow file is subsequently called, the flow file set can be used as a guide to call, the problem that the difficulty and the complexity of data call are high due to scattered recording of the sub-flow files is solved, and the convenience in the subsequent operation of executing the target prediction operation (the prediction of the charge/discharge duration and the determination of the static duration of the cell) on the cell through the flow file set is further improved.
Example IV
Referring to fig. 5, fig. 5 is a schematic structural diagram of an intelligent prediction device for charging and discharging time in a cell formation process according to another embodiment of the present invention. As shown in fig. 5, the intelligent prediction device for charging and discharging time of the battery cell formation process may include:
A memory 401 storing executable program codes;
A processor 402 coupled with the memory 401;
the processor 402 invokes executable program codes stored in the memory 401 to execute steps in the intelligent prediction method of the charge and discharge time of the cell formation process described in the first embodiment or the second embodiment of the present invention.
Example five
The embodiment of the invention discloses a computer storage medium which stores computer instructions for executing the steps in the intelligent prediction method of the charge and discharge time of the cell formation flow described in the first embodiment or the second embodiment of the invention when the computer instructions are called.
Example six
An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer storage medium storing a computer program, and the computer program is operable to cause a computer to execute steps in the intelligent prediction method of the charge-discharge time of the cell formation flow described in the first embodiment or the second embodiment.
The apparatus embodiments described above are merely illustrative, in which the modules illustrated as separate components may or may not be physically separate, and the components shown as modules may or may not be physical, i.e., may be located in one place, or may be distributed over multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above detailed description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer storage medium including Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), one-time programmable Read-Only Memory (OTPROM), electrically erasable programmable Read-Only Memory (EEPROM), compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM) or other optical disc Memory, magnetic disc Memory, tape Memory, or any other medium readable by a computer that can be used to carry or store data.
Finally, it should be noted that: the embodiment of the invention discloses an intelligent prediction method and device for charging and discharging time of a battery cell formation process, which are disclosed by the embodiment of the invention only as a preferred embodiment of the invention, and are only used for illustrating the technical scheme of the invention, but not limiting the technical scheme; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme recorded in the various embodiments can be modified or part of technical features in the technical scheme can be replaced equivalently; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (9)
1. An intelligent prediction method for charging and discharging time of a cell formation process is characterized by comprising the following steps:
Determining all first electric cores needing to execute preset electric core prediction operation, and acquiring historical data corresponding to each first electric core, wherein the historical data corresponding to each first electric core is used for determining electric core formation data of the first electric core, and the electric core formation data of the first electric core are used for predicting charge and discharge time of the first electric core;
Determining a plurality of analysis parameters corresponding to each first battery cell according to the historical data corresponding to each first battery cell, wherein each analysis parameter is a parameter affecting the charge and discharge time length of the first battery cell for executing charge and discharge operation; all the analysis parameters comprise at least one analysis parameter of the cell weight, the initial voltage, the internal resistance value and the theoretical capacitance of each first cell, and the analysis parameters are recorded as first sub-parameters; the analysis parameters further comprise a single-time charging duration of the battery cell of each first battery cell, a charging frequency corresponding to the single-time charging duration of the battery cell, a single-time discharging duration of the battery cell and a discharging frequency corresponding to the single-time discharging duration of the battery cell, and the single-time charging duration of the battery cell, the charging frequency, the single-time discharging duration and the discharging frequency, the discharging frequency and the discharging frequency are recorded as second sub-parameters;
Inputting all the historical data into a preset mathematical model by taking all the analysis parameters as references so as to train the mathematical model;
the mathematical model after training is used for analyzing the battery cell input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell;
after determining that the mathematical model has completed training, the method further comprises:
determining standing data corresponding to each first battery cell from historical data corresponding to each first battery cell, wherein the standing data corresponding to each first battery cell comprises a plurality of sub-standing time periods, and each sub-standing time period is the time period of standing after the first battery cell performs charging operation or discharging operation;
Generating a process flow file for each preset battery cell set according to the first sub-parameters, the second sub-parameters, the target historical data and the sub-standing time length corresponding to each first battery cell in the battery cell set for all the first battery cells in the battery cell set; the set division standard corresponding to each preset battery cell set is the battery cell identification corresponding to all the first battery cells; the target historical data corresponding to each first battery cell is the historical data matched with each first sub-parameter, which is extracted from the historical data corresponding to the first battery cell;
The process flow files comprise characteristic parameters of each first battery cell, wherein the characteristic parameters of each first battery cell are used for predicting the predicted charging duration corresponding to the charging operation of the first battery cell and the first standing duration after the charging operation is performed, or used for predicting the predicted discharging duration corresponding to the discharging operation of the first battery cell and the second standing duration after the discharging operation is performed.
2. The intelligent prediction method of charge and discharge time in a cell formation process according to claim 1, wherein after determining a plurality of analysis parameters corresponding to each of the first cells according to the historical data corresponding to each of the first cells, the method further comprises:
Dividing all the first electric cores into a plurality of electric core sets by taking the electric core identifier corresponding to each first electric core as a dividing reference, wherein each electric core set is matched with one electric core identifier, and the electric core identifier corresponding to each first electric core is used for indicating the model and/or the type of the first electric core;
and for each first cell in each cell set, taking the first sub-parameter corresponding to the first cell as a reference, extracting the history data matched with each first sub-parameter from the history data corresponding to the first cell, recording the history data as target history data, and executing data association or data filling on the first cell and the target history data corresponding to the first cell so as to update all the cell sets.
3. The intelligent prediction method of charge and discharge time in a cell formation process according to claim 2, wherein the step of inputting all the historical data into a preset mathematical model based on all the analysis parameters to train the mathematical model comprises:
For each first cell, inputting the target historical data corresponding to the first cell into a preset mathematical model, and executing model training operation on the mathematical model by taking the first sub-parameter corresponding to the first cell as an independent variable and the second sub-parameter corresponding to the first cell as a dependent variable to obtain a training result corresponding to the mathematical model;
after determining that the model training is performed on all of the first sub-parameters, all of the second sub-parameters by the mathematical model, it is determined that a model training operation for the mathematical model is completed.
4. The intelligent prediction method of the charge and discharge time of the battery cell formation process according to claim 3, wherein each target historical data has a corresponding data tag, and the data tag is used for indicating that the target historical data is the historical data recorded after the first battery cell performs the charge operation or the discharge operation;
Before the target historical data corresponding to each first cell is input into a preset mathematical model, the method further comprises:
according to the data labels corresponding to each target historical data and the matching parameters in each second sub-parameter, matching operation is carried out on all the historical data and all the second sub-parameters, and a plurality of paired data are obtained;
updating all the target historical data according to all the pairing data;
Wherein the matching parameters include the number of charging and the number of discharging; each of the pairing data includes one of the target history data and one of the second sub-parameters.
5. The method of intelligent prediction of charge-discharge time in a cell formation process according to any one of claims 1-4, wherein after determining that the mathematical model has completed training, the method further comprises:
Determining all the current battery cells needing to execute target prediction operation, marking the current battery cells as second battery cells, and acquiring battery cell data corresponding to each second battery cell; the target prediction operation is used for predicting the predicted charging duration corresponding to the charging operation executed by the battery cell or the predicted discharging duration corresponding to the discharging operation executed by the battery cell; the corresponding electric core data of each second electric core comprises the electric core type of the second electric core, and the corresponding electric core data of each second electric core also comprises at least one data of the electric core weight, initial voltage, internal resistance value, theoretical capacitance, charging times and discharging times of the second electric core;
inputting the cell data corresponding to all the second cells into the mathematical model and executing the target prediction operation to obtain a predicted duration corresponding to each second cell, wherein the predicted duration comprises the predicted discharge duration or the predicted charge duration;
According to the cell division parameters corresponding to each second cell, performing division operation on all the second cells by combining the set time length deviation to obtain a plurality of cell division batches;
The battery cell dividing parameters corresponding to each second battery cell comprise the battery cell type corresponding to the second battery cell, the prediction duration and preset operation matters; the preset operation item is used for indicating that the second battery cell currently needs to execute the discharging operation or the charging operation.
6. The intelligent prediction method of charge and discharge time in a cell formation process according to claim 5, wherein each process flow file comprises a plurality of sub-flow files, and each sub-flow file corresponds to one first cell; the method further comprises the steps of:
For all the sub-process files in the same cell set, dividing all the sub-process files into a process file set corresponding to each deviation interval according to a preset deviation interval by taking the accumulated duration value corresponding to each sub-process file as a reference;
The accumulated time length value corresponding to each first battery cell is obtained by calculating the accumulated charging time length and the accumulated discharging time length of the first battery cell;
And the flow file sets are used for determining the predicted data of the battery cells to be analyzed, wherein the battery cells to be analyzed are the battery cells needing to execute the battery cell charging and discharging duration prediction operation, and the predicted data comprise the predicted discharging duration and the predicted charging duration corresponding to the battery cells to be analyzed.
7. An intelligent prediction device for charging and discharging time of a cell formation process is characterized by comprising:
The determining module is used for determining all first battery cells needing to execute preset battery cell prediction operation;
the acquisition module is used for acquiring historical data corresponding to each first battery cell, wherein the historical data corresponding to each first battery cell is used for determining battery cell formation data of the first battery cell, and the battery cell formation data of the first battery cell are used for predicting the charge and discharge time length of the first battery cell;
The determining module is further configured to determine, according to the historical data corresponding to each first electrical core, a plurality of analysis parameters corresponding to each first electrical core, where each analysis parameter is a parameter affecting a charge-discharge duration of the first electrical core when the first electrical core performs a charge-discharge operation; all the analysis parameters comprise at least one analysis parameter of the cell weight, the initial voltage, the internal resistance value and the theoretical capacitance of each first cell, and the analysis parameters are recorded as first sub-parameters; the analysis parameters further comprise a single-time charging duration of the battery cell of each first battery cell, a charging frequency corresponding to the single-time charging duration of the battery cell, a single-time discharging duration of the battery cell and a discharging frequency corresponding to the single-time discharging duration of the battery cell, and the single-time charging duration of the battery cell, the charging frequency, the single-time discharging duration and the discharging frequency, the discharging frequency and the discharging frequency are recorded as second sub-parameters;
the training module is used for inputting all the historical data into a preset mathematical model by taking all the analysis parameters as references so as to train the mathematical model;
the mathematical model after training is used for analyzing the battery cell input into the mathematical model so as to predict the charge and discharge time length corresponding to the charge and discharge operation executed by the battery cell;
The determining module is further configured to determine, after the mathematical model is determined to complete training, standing data corresponding to each first electrical core from historical data corresponding to each first electrical core, where each standing data corresponding to each first electrical core includes a plurality of sub-standing periods, and each sub-standing period is a period of time for which the first electrical core is standing after performing a charging operation or a discharging operation;
the apparatus further comprises:
The generating module is used for generating a process flow file for each preset battery cell set according to the first sub-parameter, the second sub-parameter, the target historical data and the sub-standing duration corresponding to each first battery cell in the battery cell set; the set division standard corresponding to each preset battery cell set is the battery cell identification corresponding to all the first battery cells; the target historical data corresponding to each first battery cell is the historical data matched with each first sub-parameter, which is extracted from the historical data corresponding to the first battery cell;
The process flow files comprise characteristic parameters of each first battery cell, wherein the characteristic parameters of each first battery cell are used for predicting the predicted charging duration corresponding to the charging operation of the first battery cell and the first standing duration after the charging operation is performed, or used for predicting the predicted discharging duration corresponding to the discharging operation of the first battery cell and the second standing duration after the discharging operation is performed.
8. An intelligent prediction device for charging and discharging time of a cell formation process is characterized by comprising:
A memory storing executable program code;
A processor coupled to the memory;
The processor invokes the executable program code stored in the memory to perform the intelligent prediction method of the charge and discharge time of the cell formation process as claimed in any one of claims 1 to 6.
9. A computer storage medium storing computer instructions which, when invoked, are adapted to perform the intelligent method of predicting charge and discharge times of a cell formation process according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410391485.5A CN117973238B (en) | 2024-04-02 | 2024-04-02 | Intelligent prediction method and device for charge and discharge time of battery cell formation process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410391485.5A CN117973238B (en) | 2024-04-02 | 2024-04-02 | Intelligent prediction method and device for charge and discharge time of battery cell formation process |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117973238A CN117973238A (en) | 2024-05-03 |
CN117973238B true CN117973238B (en) | 2024-06-04 |
Family
ID=90849902
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410391485.5A Active CN117973238B (en) | 2024-04-02 | 2024-04-02 | Intelligent prediction method and device for charge and discharge time of battery cell formation process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117973238B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113095000A (en) * | 2021-06-08 | 2021-07-09 | 蜂巢能源科技有限公司 | Method and device for obtaining battery cell discharge capacity, storage medium and electronic equipment |
WO2022242653A1 (en) * | 2021-05-17 | 2022-11-24 | 蜂巢能源科技股份有限公司 | Method and apparatus for determining charging/discharging performance of battery cell, storage medium, and electronic device |
CN115587527A (en) * | 2022-08-31 | 2023-01-10 | 广东邦普循环科技有限公司 | Battery life prediction method, system, terminal device and computer readable medium |
CN116359763A (en) * | 2023-06-01 | 2023-06-30 | 深圳和润达科技有限公司 | Intelligent analysis method and device for chemical component capacity energy consumption |
CN117129875A (en) * | 2023-08-25 | 2023-11-28 | 合肥工业大学 | Method, system, equipment and medium for training and predicting battery capacity prediction model |
CN117741442A (en) * | 2023-12-11 | 2024-03-22 | 阳光电源股份有限公司 | Cell temperature prediction method, device, apparatus, storage medium and program product |
-
2024
- 2024-04-02 CN CN202410391485.5A patent/CN117973238B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022242653A1 (en) * | 2021-05-17 | 2022-11-24 | 蜂巢能源科技股份有限公司 | Method and apparatus for determining charging/discharging performance of battery cell, storage medium, and electronic device |
CN113095000A (en) * | 2021-06-08 | 2021-07-09 | 蜂巢能源科技有限公司 | Method and device for obtaining battery cell discharge capacity, storage medium and electronic equipment |
CN115587527A (en) * | 2022-08-31 | 2023-01-10 | 广东邦普循环科技有限公司 | Battery life prediction method, system, terminal device and computer readable medium |
WO2024045567A1 (en) * | 2022-08-31 | 2024-03-07 | 广东邦普循环科技有限公司 | Battery life prediction method and system, terminal device, and computer readable medium |
CN116359763A (en) * | 2023-06-01 | 2023-06-30 | 深圳和润达科技有限公司 | Intelligent analysis method and device for chemical component capacity energy consumption |
CN117129875A (en) * | 2023-08-25 | 2023-11-28 | 合肥工业大学 | Method, system, equipment and medium for training and predicting battery capacity prediction model |
CN117741442A (en) * | 2023-12-11 | 2024-03-22 | 阳光电源股份有限公司 | Cell temperature prediction method, device, apparatus, storage medium and program product |
Also Published As
Publication number | Publication date |
---|---|
CN117973238A (en) | 2024-05-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110161425B (en) | Method for predicting remaining service life based on lithium battery degradation stage division | |
CN114114049B (en) | Lithium ion battery life prediction method based on sample migration | |
CN109001640B (en) | Data processing method and device for power battery | |
CN111999648A (en) | Lithium battery residual life prediction method based on long-term and short-term memory network | |
CN113406496B (en) | Battery capacity prediction method, system, device and medium based on model migration | |
CN116359763B (en) | Intelligent analysis method and device for chemical component capacity energy consumption | |
CN116360377B (en) | Data processing method for battery cell formation component capacity and distributed control system | |
WO2024060539A1 (en) | Annual carbon emission amount estimation method and device for power battery | |
CN108594117B (en) | Method and device for determining battery capacity, terminal equipment and computer storage medium | |
CN116047300A (en) | Controller for predicting characteristic parameters of battery and method thereof | |
CN117154844A (en) | Energy supply control method and device for energy storage system | |
CN117554846B (en) | Lithium battery life prediction method and system considering constraint conditions | |
CN114519470A (en) | Method and device for predicting medium-and-long-term power generation amount of wind power | |
CN117632905B (en) | Database management method and system based on cloud use records | |
CN117973238B (en) | Intelligent prediction method and device for charge and discharge time of battery cell formation process | |
CN114611799A (en) | Method, apparatus, device, storage medium and program product for predicting generated power | |
CN117872146A (en) | Method, device, equipment and storage medium for monitoring battery abnormality | |
CN117289167A (en) | Battery remaining life prediction method, device and medium based on multiple neural network | |
CN113459897B (en) | Electric vehicle charging big data-based state of charge correction and prediction method | |
CN114859231A (en) | Method for predicting remaining life of battery based on wiener process and extreme learning machine | |
DE102017103348A1 (en) | Operating strategy management system and method for operating an electrical energy storage | |
CN117332901B (en) | New energy small time scale power prediction method adopting layered time aggregation strategy | |
CN110008245A (en) | One is suitable for the method for equipment fault early-warning model time segment search | |
CN117991109B (en) | Model training and charging state determining method, device, equipment and storage medium | |
CN117201502B (en) | Intelligent cloud server access method and system based on artificial intelligence |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |