CN114371409A - Training method of battery state prediction model, and battery state prediction method and device - Google Patents
Training method of battery state prediction model, and battery state prediction method and device Download PDFInfo
- Publication number
- CN114371409A CN114371409A CN202210214884.5A CN202210214884A CN114371409A CN 114371409 A CN114371409 A CN 114371409A CN 202210214884 A CN202210214884 A CN 202210214884A CN 114371409 A CN114371409 A CN 114371409A
- Authority
- CN
- China
- Prior art keywords
- model
- battery
- training
- simulation
- neural network
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses a training method of a battery state prediction model, a battery state prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring an electrochemical model, wherein the electrochemical model is constructed by measurement operation data and attribute data of a battery; performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions; and taking the simulation operation data and the measurement operation data as training samples, inputting the training samples into the neural network, adjusting network parameters of the neural network and model parameters of the electrochemical model according to the output result of the neural network, and determining the neural network which meets the iteration stop condition as a battery state prediction model. In the invention, the training samples of the neural network are expanded and enriched by means of the electrochemical model, and the electrochemical model is optimized based on the output result of the neural network, so that the electrochemical model provides more accurate training samples for the neural network, and the model training accuracy is improved.
Description
Technical Field
The present invention relates to the field of battery technologies, and in particular, to a training method for a battery state prediction model, a battery state prediction method and apparatus, an electronic device, and a storage medium.
Background
With the development of new energy, the usage of rechargeable batteries such as lithium batteries is rapidly increasing. In the practical use of lithium batteries, the batteries often cannot be used in an ideal manner in a full life cycle due to the environment in which the batteries are located and the influence of factors such as series connection, parallel connection and the like. The abnormal life is attenuated, and even accidents such as smoking and spontaneous combustion occur. And thus is very important for the state detection of the battery.
A conventional Battery Management System (BMS) generally only roughly estimates the state of a battery through a statistical, filtering, or equivalent circuit model from measured values of voltage, current, temperature, etc. of the battery, and thus has low accuracy. Few schemes predict the state of the battery based on an electrochemical model, but because of limited calculation, the real-time and effective electrochemical model is difficult to establish, and the fault early warning of the battery is inaccurate. The state of the battery is predicted based on machine learning, but the prediction is limited by acquisition and marking of training samples, so that an accurate model cannot be obtained, and the state of the battery cannot be accurately predicted.
Disclosure of Invention
The invention provides a training method of a battery state prediction model, a battery state prediction method and device, electronic equipment and a storage medium, and aims to improve the accuracy of battery state detection.
The invention solves the technical problems through the following technical scheme:
in a first aspect, a training method for a battery state prediction model is provided, including:
acquiring an electrochemical model, wherein the electrochemical model is constructed by measurement operation data and attribute data of a battery;
performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions;
and taking the simulation operation data and the measurement operation data as training samples, inputting the training samples into a neural network, adjusting network parameters of the neural network and model parameters of the electrochemical model according to an output result of the neural network, and determining the neural network meeting an iteration stop condition as the battery state prediction model.
Optionally, the method further comprises:
acquiring measurement operation data and attribute data of the battery;
and fitting the measurement operation data according to the attribute data, and determining the electrochemical model according to a fitting result.
Optionally, performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions, including:
generating simulation condition data corresponding to different simulation conditions;
inputting the simulation condition data into the electrochemical model for charge and discharge simulation to obtain simulation operation data; the simulation operation data comprises operation state data of the battery under different operation states.
Optionally, adjusting the model parameters of the electrochemical model according to the output result of the neural network comprises:
calculating a loss error according to an output result of the neural network;
under the condition that the loss error falls into a preset error range, adjusting model parameters of the electrochemical model according to the loss error; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
Optionally, adjusting the model parameters of the electrochemical model according to the output result of the neural network comprises:
inputting a training sample corresponding to the output result of the neural network into the electrochemical model to obtain the output result of the electrochemical model;
under the condition that the difference value between the output result of the neural network and the output result of the electrochemical model falls into a preset difference value range, adjusting the model parameters of the electrochemical model according to the difference value; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
In a second aspect, a battery state prediction method is provided, including:
acquiring measurement data of a battery;
inputting the measurement data into a battery state prediction model; the battery state prediction model is obtained by training according to any one of the above training methods of the battery state prediction model;
and determining the state of the battery according to the output result of the battery state prediction model.
In a third aspect, a training apparatus for a battery state prediction model is provided, including:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring an electrochemical model, and the electrochemical model is constructed by measurement operation data and attribute data of a battery;
the simulation module is used for performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions;
and the training module is used for inputting the simulation operation data and the measurement operation data serving as training samples into a neural network, adjusting network parameters of the neural network and model parameters of the electrochemical model according to the output result of the neural network, and determining the neural network meeting the iteration stop condition as the battery state prediction model.
Optionally, the method further comprises:
the second acquisition module is used for acquiring the measured operation data and the attribute data of the battery;
and the fitting module is used for fitting the measurement operation data according to the attribute data and determining the electrochemical model according to the fitting result.
Optionally, the simulation module includes:
a generating unit for generating simulation condition data corresponding to different simulation conditions;
the simulation unit is used for inputting the simulation condition data into the electrochemical model to perform charge and discharge simulation to obtain simulation operation data; the simulation operation data comprises operation state data of the battery under different operation states.
Optionally, the training module comprises:
the calculation unit is used for calculating loss errors according to output results of the neural network;
the adjusting unit is used for adjusting the model parameters of the electrochemical model according to the loss error under the condition that the loss error falls into a preset error range; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
Optionally, the training module comprises:
the input unit is used for inputting a training sample corresponding to the output result of the neural network into the electrochemical model to obtain the output result of the electrochemical model;
the adjusting unit is used for adjusting the model parameters of the electrochemical model according to the difference value when the difference value between the output result of the neural network and the output result of the electrochemical model falls into a preset difference value range; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
In a fourth aspect, a battery state prediction apparatus is provided, including:
the acquisition module is used for acquiring measurement data of the battery;
the input module is used for inputting the measurement data into a battery state prediction model; the battery state prediction model is obtained by training according to the training device of the battery state prediction model;
and the determining module is used for determining the state of the battery according to the output result of the battery state prediction model.
In a fifth aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of the above when executing the computer program.
A sixth aspect provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of the above.
The positive progress effects of the invention are as follows: in the invention, the training samples of the neural network are expanded and enriched by means of the electrochemical model, and the electrochemical model is optimized based on the output result of the neural network, so that the electrochemical model provides more accurate training samples for the neural network, and the model training accuracy is improved.
Drawings
Fig. 1 is a flowchart of a training method of a battery state prediction model according to an exemplary embodiment of the present invention;
fig. 2 is a flowchart of a battery state prediction method according to an exemplary embodiment of the present invention;
FIG. 3 is a block diagram of a training apparatus for a battery state prediction model according to an exemplary embodiment of the present invention;
fig. 4 is a block diagram of a battery state prediction apparatus according to an exemplary embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention.
Detailed Description
The invention is further illustrated by the following examples, which are not intended to limit the scope of the invention.
Fig. 1 is a flowchart of a training method for a battery state prediction model according to an exemplary embodiment of the present invention, where the training method includes the following steps:
The measured operation data may be data measured during a test for charging and discharging the battery, and is preferably. The number of the batteries can be one or more, that is, an electrochemical model of the battery is constructed based on the measured operation data and the attribute data of one battery, or the electrochemical model of the battery is constructed based on the measured operation data and the attribute data of a plurality of batteries, and the plurality of batteries are of the same type. Charging and discharging of the battery may be performed under a plurality of simulation conditions. The simulation conditions may be characterized by, but are not limited to, parameters such as ambient temperature, ambient humidity, ambient pressure, and the like.
The measured operating data includes simulation condition data and operating state data.
The simulation condition data includes at least one of the following parameters: ambient temperature, ambient humidity, ambient pressure.
The operation State data includes at least one parameter capable of representing the performance of the battery, such as battery temperature, battery voltage, battery Power, SOC (State of charge), SOH (State of health), SOP (State of Power), SOS (State of safety estimation), and the like of the battery in different operation states.
The operation state of the battery includes a normal state, an abnormal state, and a fault state. The abnormal state means that the operating state data of the electric potential does not meet the normal state standard, but does not meet the fault state standard. The division of each state is set according to the actual situation.
The attribute data includes at least one of the following parameters: the battery comprises battery size, anode and cathode material parameters, diaphragm material parameters, electrolyte material parameters, mobility and battery material parameters calculated by adopting a first principle.
In one embodiment, the electrochemical model of the battery is constructed by fitting the measured operating data to the attribute data and determining the electrochemical model based on the fitting. The electrochemical model can estimate the running state data of the battery under different simulation conditions, including SOC, SOH, SOP, SOS and the like. The electrochemical model may be, but is not limited to, a Pseudo 2Dimension (P2D) lithium battery model.
And 102, performing charging and discharging simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions.
In one embodiment, in step 102, simulation condition data corresponding to different simulation conditions are generated, and the simulation condition data are input into an electrochemical model for charge and discharge simulation to obtain simulation operation data of the battery; the simulation operation data comprises operation state data of the battery under different simulation conditions.
Regarding the generation of the simulation condition data, in one embodiment, the parameter value ranges of the scene parameters representing the simulation conditions input by the user are obtained, sampling is performed in the parameter value ranges of the scene parameters, and the sampling results are arranged and combined to generate the simulation condition data of different simulation conditions. For example, if the parameter value range of the environmental temperature input by the user is [10 ℃, 50 ℃), the parameter value range of the environmental humidity is [ 20%, 80% ], and the sampling interval of the environmental temperature is 10 ℃ and the sampling interval of the environmental humidity is 10%, sampling is performed to obtain 5 temperature sampling points of 10 ℃, 20 ℃, 30 ℃, 40 ℃ and 50 ℃ and 7 humidity sampling points of 20%, 30%, 40%, 50%, 60%, 70% and 80%, 35 permutation and combination can be obtained (10 ℃, 20%), (10 ℃, 30%), … …, (50 ℃, 80%), and the charging and discharging simulation is performed on the electrochemical model for each permutation and combination to obtain the simulation operation data of the battery under the corresponding simulation condition.
It should be noted that the sampling interval may be set according to actual situations, and may be equal-interval sampling or random sampling, which is not limited in this embodiment of the present invention.
The simulation operation data obtained based on the electrochemical model is used as a training sample for training the neural network, and because the battery operation data under a plurality of simulation conditions are difficult to obtain through measurement, the simulation operation data under various simulation conditions can be obtained by means of the electrochemical model, so that the training sample of the neural network is expanded and enriched, and the accuracy of model training is improved. It can be understood that the more simulation conditions are set, the wider the range of the simulation operation data representing the battery in the full life cycle is, the higher the richness of the training samples is, and the more accurate the battery state prediction model can be obtained.
And 103, inputting the simulation operation data and the measurement operation data serving as training samples into the neural network, adjusting network parameters of the neural network and model parameters of the electrochemical model according to the output result of the neural network, and determining the neural network meeting the iteration stop condition as a battery state prediction model.
The neural network may be, but is not limited to, RNN (recurrent neural network), LSTM (long short term memory), transform (graph neural network), or the like.
In step 103, in the training process of the neural network, calculating a loss error according to an output result of the neural network, adjusting network parameters of the neural network based on the loss error, repeating the above process until the neural network meets an iteration stop condition, stopping training, and determining the trained neural network as a battery state prediction model. And in the training process of the neural network, the model parameters of the electrochemical model are adjusted based on the output result of the neural network, and the optimization of the electrochemical model is realized by continuously iterating and adjusting the model parameters of the electrochemical model.
The trained battery state prediction model can predict the state of the battery, analyze and judge the actual working condition of the battery, and give early warning to possible faults of the battery.
It is understood that the input parameters and the output parameters of the battery state prediction model are related to the training samples, for example, if the training samples include the SOC, the trained battery state prediction model can predict the SOC of the battery; if the training sample comprises SOH, the trained battery state prediction model can predict the SOH of the battery; if the training samples comprise SOC and SOH, the SOC and SOH of the battery can be predicted by the trained battery state prediction model; if the training samples comprise SOC, SOH, SOP and SOS, the trained battery state prediction model can predict the SOC, SOH, SOP and SOS of the battery; and so on.
In the embodiment of the invention, the training samples of the neural network are expanded and enriched by virtue of the electrochemical model, and the electrochemical model is optimized based on the output result of the neural network, so that the electrochemical model provides more accurate training samples for the neural network, and the model training accuracy is improved.
For the optimization of the electrochemical model, the electrochemical model can be optimized in each iteration process of the neural network, and the electrochemical model can also be optimized when the loss error of the neural network is within a preset error range or meets other conditions.
In one embodiment, in the training process of the neural network, the step of optimizing the electrochemical model specifically includes: and calculating a loss error according to the output result of the neural network, and adjusting the model parameters of the electrochemical model according to the loss error under the condition that the loss error falls into a preset error range. And then, simulation operation data can be obtained based on the electrochemical model obtained through model parameter adjustment, and the simulation operation data can be used as a training sample of the neural network. The electrochemical model obtained through model parameter adjustment is more accurate and can represent the characteristics of the battery, and a more accurate battery state prediction model can be obtained by training a neural network based on simulation operation data obtained through the electrochemical model obtained through model parameter adjustment.
The preset error range can be designed according to actual conditions. If the loss error falls into the preset error range, the electrochemical model is optimized probably because the training sample is not ideal, namely, the simulation operation data output by the electrochemical model is not ideal.
In one embodiment, in the training process of the neural network, the step of optimizing the electrochemical model specifically includes: in the iterative training process of the neural network, inputting a training sample corresponding to an output result of the neural network into the electrochemical model to obtain an output result of the electrochemical model, and under the condition that a difference value between the output result of the neural network and the output result of the electrochemical model falls into a preset difference value range, adjusting model parameters of the electrochemical model according to the difference value; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
Wherein, the preset difference range can be set according to the actual situation. If the difference value falls into the preset difference value range, probably because the training sample is not ideal, namely the simulation operation data output by the electrochemical model is not ideal, the electrochemical model is optimized.
In the embodiment of the invention, the output result based on the neural network and the output result of the electrochemical model are crossed and mutually optimized, which is equivalent to that the neural network has the characteristics of the electrochemical model, and the calculation speed of the neural network is much faster than that of the electrochemical model, so that the trained battery state prediction model not only can realize accurate prediction of the battery state, but also has higher calculation speed, and realizes real-time calculation of electrochemistry in battery detection.
Fig. 2 is a flowchart of a battery state prediction method according to an exemplary embodiment of the present invention, where the battery state prediction method includes the following steps:
The battery may be an in-use battery, a battery in a charged state, or an idle battery. The measurement data may include, but is not limited to, current, voltage, temperature, etc. of the battery.
The battery state prediction model is obtained by training according to the training method of the battery state prediction model provided by any one of the embodiments.
And step 203, determining the state of the battery according to the output result of the battery state prediction model.
The state of the battery may include, but is not limited to, a normal state, an abnormal state, a fault state, and may also be characterized by at least one of SOC, SOH, SOP, and SOS.
Corresponding to the training method of the battery state prediction model and the embodiment of the battery state prediction method, the invention also provides a training device of the battery state prediction model and an embodiment of a battery state prediction device.
Fig. 3 is a schematic block diagram of a training apparatus for a battery state prediction model according to an exemplary embodiment of the present invention, where the training apparatus includes:
the first obtaining module 31 is configured to obtain an electrochemical model, where the electrochemical model is constructed by measurement operation data and attribute data of a battery;
the simulation module 32 is used for performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions;
the training module 33 is configured to input the simulation operation data and the measurement operation data as training samples into a neural network, adjust network parameters of the neural network and model parameters of the electrochemical model according to an output result of the neural network, and determine the neural network meeting an iteration stop condition as the battery state prediction model.
Optionally, the method further comprises:
the second acquisition module is used for acquiring the measured operation data and the attribute data of the battery;
and the fitting module is used for fitting the measurement operation data according to the attribute data and determining the electrochemical model according to the fitting result.
Optionally, the simulation module includes:
a generating unit for generating simulation condition data corresponding to different simulation conditions;
the simulation unit is used for inputting the simulation condition data into the electrochemical model to perform charge and discharge simulation to obtain simulation operation data; the simulation operation data comprises operation state data of the battery under different operation states.
Optionally, the training module comprises:
the calculation unit is used for calculating loss errors according to output results of the neural network;
the adjusting unit is used for adjusting the model parameters of the electrochemical model according to the loss error under the condition that the loss error falls into a preset error range; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
Optionally, the training module comprises:
the input unit is used for inputting a training sample corresponding to the output result of the neural network into the electrochemical model to obtain the output result of the electrochemical model;
the adjusting unit is used for adjusting the model parameters of the electrochemical model according to the difference value when the difference value between the output result of the neural network and the output result of the electrochemical model falls into a preset difference value range; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
Fig. 4 is a block diagram of a battery state prediction apparatus according to an exemplary embodiment of the present invention, where the battery state prediction apparatus includes:
an obtaining module 41, configured to obtain measurement data of the battery;
an input module 42 for inputting the measurement data into a battery state prediction model; the battery state prediction model is obtained by training according to the training device of the battery state prediction model;
and a determining module 43, configured to determine the state of the battery according to an output result of the battery state prediction model.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the invention. One of ordinary skill in the art can understand and implement it without inventive effort.
Fig. 5 is a schematic diagram of an electronic device according to an exemplary embodiment of the present invention, and shows a block diagram of an exemplary electronic device 50 suitable for implementing an embodiment of the present invention. The electronic device 50 shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.
As shown in fig. 5, the electronic device 50 may be embodied in the form of a general purpose computing device, which may be, for example, a server device. The components of the electronic device 50 may include, but are not limited to: the at least one processor 51, the at least one memory 52, and a bus 53 connecting the various system components (including the memory 52 and the processor 51).
The bus 53 includes a data bus, an address bus, and a control bus.
The memory 52 may include volatile memory, such as Random Access Memory (RAM)521 and/or cache memory 522, and may further include Read Only Memory (ROM) 523.
The processor 51 executes various functional applications and data processing, such as the methods provided by any of the above embodiments, by running a computer program stored in the memory 52.
The electronic device 50 may also communicate with one or more external devices 54 (e.g., a keyboard, a pointing device, etc.). Such communication may be through an input/output (I/O) interface 55. Moreover, the model-generated electronic device 50 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet) via a network adapter 56. As shown, network adapter 56 communicates with the other modules of model-generated electronic device 50 over bus 53. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the model-generating electronic device 50, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.
It should be noted that although in the above detailed description several units/modules or sub-units/modules of the electronic device are mentioned, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the units/modules described above may be embodied in one unit/module according to embodiments of the invention. Conversely, the features and functions of one unit/module described above may be further divided into embodiments by a plurality of units/modules.
Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method provided in any of the above embodiments.
More specific examples, among others, that the readable storage medium may employ may include, but are not limited to: a portable disk, a hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible implementation manner, the embodiment of the present invention may also be implemented in a form of a program product, which includes program code for causing a terminal device to execute a method implementing any of the above-mentioned embodiments when the program product runs on the terminal device.
Where program code for carrying out the invention is written in any combination of one or more programming languages, the program code may be executed entirely on the user device, partly on the user device, as a stand-alone software package, partly on the user device and partly on a remote device or entirely on the remote device.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (14)
1. A training method of a battery state prediction model is characterized by comprising the following steps:
acquiring an electrochemical model, wherein the electrochemical model is constructed by measurement operation data and attribute data of a battery;
performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions;
and taking the simulation operation data and the measurement operation data as training samples, inputting the training samples into a neural network, adjusting network parameters of the neural network and model parameters of the electrochemical model according to an output result of the neural network, and determining the neural network meeting an iteration stop condition as the battery state prediction model.
2. The method for training a battery state prediction model according to claim 1, further comprising:
acquiring measurement operation data and attribute data of the battery;
and fitting the measurement operation data according to the attribute data, and determining the electrochemical model according to a fitting result.
3. The method for training the battery state prediction model according to claim 1, wherein performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions comprises:
generating simulation condition data corresponding to different simulation conditions;
inputting the simulation condition data into the electrochemical model for charge and discharge simulation to obtain simulation operation data; the simulation operation data comprises operation state data of the battery under different operation states.
4. The method for training a battery state prediction model according to any one of claims 1 to 3, wherein adjusting the model parameters of the electrochemical model according to the output result of the neural network comprises:
calculating a loss error according to an output result of the neural network;
under the condition that the loss error falls into a preset error range, adjusting model parameters of the electrochemical model according to the loss error; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
5. The method for training a battery state prediction model according to any one of claims 1 to 3, wherein adjusting the model parameters of the electrochemical model according to the output result of the neural network comprises:
inputting a training sample corresponding to the output result of the neural network into the electrochemical model to obtain the output result of the electrochemical model;
under the condition that the difference value between the output result of the neural network and the output result of the electrochemical model falls into a preset difference value range, adjusting the model parameters of the electrochemical model according to the difference value; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
6. A battery state prediction method, comprising:
acquiring measurement data of a battery;
inputting the measurement data into a battery state prediction model; the battery state prediction model is obtained by training according to the training method of the battery state prediction model in any one of claims 1-5;
and determining the state of the battery according to the output result of the battery state prediction model.
7. A training apparatus for a battery state prediction model, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring an electrochemical model, and the electrochemical model is constructed by measurement operation data and attribute data of a battery;
the simulation module is used for performing charge and discharge simulation based on the electrochemical model to obtain simulation operation data of the battery under different simulation conditions;
and the training module is used for inputting the simulation operation data and the measurement operation data serving as training samples into a neural network, adjusting network parameters of the neural network and model parameters of the electrochemical model according to the output result of the neural network, and determining the neural network meeting the iteration stop condition as the battery state prediction model.
8. The apparatus for training a battery state prediction model according to claim 7, further comprising:
the second acquisition module is used for acquiring the measured operation data and the attribute data of the battery;
and the fitting module is used for fitting the measurement operation data according to the attribute data and determining the electrochemical model according to the fitting result.
9. The apparatus for training a battery state prediction model according to claim 7, wherein the simulation module comprises:
a generating unit for generating simulation condition data corresponding to different simulation conditions;
the simulation unit is used for inputting the simulation condition data into the electrochemical model to perform charge and discharge simulation to obtain simulation operation data; the simulation operation data comprises operation state data of the battery under different operation states.
10. Training apparatus of a battery state prediction model according to any of claims 7-9, characterized in that the training module comprises:
the calculation unit is used for calculating loss errors according to output results of the neural network;
the adjusting unit is used for adjusting the model parameters of the electrochemical model according to the loss error under the condition that the loss error falls into a preset error range; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
11. Training apparatus of a battery state prediction model according to any of claims 7-9, characterized in that the training module comprises:
the input unit is used for inputting a training sample corresponding to the output result of the neural network into the electrochemical model to obtain the output result of the electrochemical model;
the adjusting unit is used for adjusting the model parameters of the electrochemical model according to the difference value when the difference value between the output result of the neural network and the output result of the electrochemical model falls into a preset difference value range; and the electrochemical model obtained through model parameter adjustment is used for generating simulation operation data serving as the training sample.
12. A battery state prediction apparatus, comprising:
the acquisition module is used for acquiring measurement data of the battery;
the input module is used for inputting the measurement data into a battery state prediction model; the battery state prediction model is obtained by training according to a training device of the battery state prediction model in any one of claims 1-5;
and the determining module is used for determining the state of the battery according to the output result of the battery state prediction model.
13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 6 when executing the computer program.
14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210214884.5A CN114371409B (en) | 2022-03-07 | 2022-03-07 | Training method of battery state prediction model, battery state prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210214884.5A CN114371409B (en) | 2022-03-07 | 2022-03-07 | Training method of battery state prediction model, battery state prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114371409A true CN114371409A (en) | 2022-04-19 |
| CN114371409B CN114371409B (en) | 2023-07-04 |
Family
ID=81146556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210214884.5A Active CN114371409B (en) | 2022-03-07 | 2022-03-07 | Training method of battery state prediction model, battery state prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114371409B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115436821A (en) * | 2022-09-02 | 2022-12-06 | 四川新能源汽车创新中心有限公司 | Method and device for identifying heat generation abnormity of battery system |
| CN116047314A (en) * | 2023-03-31 | 2023-05-02 | 泉州装备制造研究所 | Rechargeable battery health state prediction method |
| CN116070530A (en) * | 2023-03-07 | 2023-05-05 | 国网天津市电力公司城西供电分公司 | Power grid modeling real-time simulation method and device based on data driving |
| WO2024000756A1 (en) * | 2022-06-30 | 2024-01-04 | 上海玫克生储能科技有限公司 | Electrochemical model parameter identification method based on deep learning and heuristic algorithm |
| CN117393069A (en) * | 2023-11-06 | 2024-01-12 | 上海赫耳墨锶科技有限公司 | Method for determining electrolysis control data of target metal based on neural network |
| WO2024040608A1 (en) * | 2022-08-26 | 2024-02-29 | 西门子股份公司 | Model training method for energy management system, apparatus, and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103293485A (en) * | 2013-06-10 | 2013-09-11 | 北京工业大学 | Model-based storage battery SOC (state of charge) estimating method |
| CN108898218A (en) * | 2018-05-24 | 2018-11-27 | 阿里巴巴集团控股有限公司 | A kind of training method of neural network model, device and computer equipment |
| CN109143083A (en) * | 2018-11-07 | 2019-01-04 | 重庆大学 | A kind of electric vehicle lithium ion battery analysis lithium diagnostic method of data-driven |
| CN110058159A (en) * | 2019-04-29 | 2019-07-26 | 杭州电子科技大学 | A kind of lithium battery health status estimation method based on grey neural network |
| CN110568360A (en) * | 2019-09-12 | 2019-12-13 | 华中科技大学 | lithium battery aging diagnosis method based on fuzzy logic algorithm |
| CN111190111A (en) * | 2020-01-14 | 2020-05-22 | 北京智芯微电子科技有限公司 | Electrochemical energy storage battery state of charge estimation method, device and system |
| CN111695301A (en) * | 2020-06-16 | 2020-09-22 | 中国科学院深圳先进技术研究院 | Method and device for predicting battery charge state, storage medium and equipment |
| CN112149345A (en) * | 2020-08-26 | 2020-12-29 | 清华大学 | Battery management method and device, computer equipment and storage medium |
| CN113009349A (en) * | 2021-04-09 | 2021-06-22 | 哈尔滨工业大学 | Lithium ion battery health state diagnosis method based on deep learning model |
| CN113702845A (en) * | 2021-09-02 | 2021-11-26 | 重庆金康动力新能源有限公司 | Method and equipment for evaluating core parameters of retired lithium battery |
| CN113702855A (en) * | 2021-08-31 | 2021-11-26 | 北京航空航天大学 | Lithium battery pack health state online prediction method based on multi-physical-field simulation and neural network |
| CN113884924A (en) * | 2021-11-16 | 2022-01-04 | 中国科学院电工研究所 | Lithium ion battery internal short circuit diagnosis method and system based on convolutional neural network |
-
2022
- 2022-03-07 CN CN202210214884.5A patent/CN114371409B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103293485A (en) * | 2013-06-10 | 2013-09-11 | 北京工业大学 | Model-based storage battery SOC (state of charge) estimating method |
| CN108898218A (en) * | 2018-05-24 | 2018-11-27 | 阿里巴巴集团控股有限公司 | A kind of training method of neural network model, device and computer equipment |
| CN109143083A (en) * | 2018-11-07 | 2019-01-04 | 重庆大学 | A kind of electric vehicle lithium ion battery analysis lithium diagnostic method of data-driven |
| CN110058159A (en) * | 2019-04-29 | 2019-07-26 | 杭州电子科技大学 | A kind of lithium battery health status estimation method based on grey neural network |
| CN110568360A (en) * | 2019-09-12 | 2019-12-13 | 华中科技大学 | lithium battery aging diagnosis method based on fuzzy logic algorithm |
| CN111190111A (en) * | 2020-01-14 | 2020-05-22 | 北京智芯微电子科技有限公司 | Electrochemical energy storage battery state of charge estimation method, device and system |
| CN111695301A (en) * | 2020-06-16 | 2020-09-22 | 中国科学院深圳先进技术研究院 | Method and device for predicting battery charge state, storage medium and equipment |
| CN112149345A (en) * | 2020-08-26 | 2020-12-29 | 清华大学 | Battery management method and device, computer equipment and storage medium |
| CN113009349A (en) * | 2021-04-09 | 2021-06-22 | 哈尔滨工业大学 | Lithium ion battery health state diagnosis method based on deep learning model |
| CN113702855A (en) * | 2021-08-31 | 2021-11-26 | 北京航空航天大学 | Lithium battery pack health state online prediction method based on multi-physical-field simulation and neural network |
| CN113702845A (en) * | 2021-09-02 | 2021-11-26 | 重庆金康动力新能源有限公司 | Method and equipment for evaluating core parameters of retired lithium battery |
| CN113884924A (en) * | 2021-11-16 | 2022-01-04 | 中国科学院电工研究所 | Lithium ion battery internal short circuit diagnosis method and system based on convolutional neural network |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024000756A1 (en) * | 2022-06-30 | 2024-01-04 | 上海玫克生储能科技有限公司 | Electrochemical model parameter identification method based on deep learning and heuristic algorithm |
| WO2024040608A1 (en) * | 2022-08-26 | 2024-02-29 | 西门子股份公司 | Model training method for energy management system, apparatus, and storage medium |
| CN115436821A (en) * | 2022-09-02 | 2022-12-06 | 四川新能源汽车创新中心有限公司 | Method and device for identifying heat generation abnormity of battery system |
| CN115436821B (en) * | 2022-09-02 | 2024-01-30 | 四川新能源汽车创新中心有限公司 | Method and device for identifying abnormal heat generation of battery system |
| CN116070530A (en) * | 2023-03-07 | 2023-05-05 | 国网天津市电力公司城西供电分公司 | Power grid modeling real-time simulation method and device based on data driving |
| CN116070530B (en) * | 2023-03-07 | 2023-06-27 | 国网天津市电力公司城西供电分公司 | Power grid modeling real-time simulation method and device based on data driving |
| CN116047314A (en) * | 2023-03-31 | 2023-05-02 | 泉州装备制造研究所 | Rechargeable battery health state prediction method |
| CN116047314B (en) * | 2023-03-31 | 2023-08-18 | 泉州装备制造研究所 | Rechargeable battery health state prediction method |
| CN117393069A (en) * | 2023-11-06 | 2024-01-12 | 上海赫耳墨锶科技有限公司 | Method for determining electrolysis control data of target metal based on neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114371409B (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114371409B (en) | Training method of battery state prediction model, battery state prediction method and device | |
| Zhu et al. | A state of charge estimation method for lithium-ion batteries based on fractional order adaptive extended kalman filter | |
| CN108254696B (en) | Battery health state evaluation method and system | |
| WO2022253038A1 (en) | Method and system for predicting state of health of lithium battery on basis of elastic network, and device and medium | |
| CN113189495B (en) | Battery health state prediction method and device and electronic equipment | |
| CN109143094B (en) | Abnormal data detection method and device for power battery | |
| CN105334462A (en) | Online estimation method for battery capacity loss | |
| CN114705990B (en) | Method and system for estimating state of charge of battery cluster, electronic device and storage medium | |
| CN115114878B (en) | Method and device for online prediction of battery life of energy storage power station and storage medium | |
| CN111142025A (en) | Battery SOC estimation method and device, storage medium and electric vehicle | |
| JP2020125968A (en) | Battery degradation diagnosing device, battery degradation analysis circuit, and battery degradation diagnosing program | |
| Xiao et al. | Rapid measurement method for lithium‐ion battery state of health estimation based on least squares support vector regression | |
| CN116796623A (en) | Lithium battery SOC prediction method, system, equipment and medium | |
| CN204030697U (en) | Based on the battery management system of dynamic SOC estimating system | |
| CN114879070A (en) | Battery state evaluation method and related equipment | |
| Qian et al. | State of health estimation of lithium-ion battery using energy accumulation-based feature extraction and improved relevance vector regression | |
| CN116027218A (en) | Battery state evaluation method and electronic equipment | |
| CN114660461A (en) | Battery health state evaluation method and device, electronic equipment and storage medium | |
| CN117825970A (en) | Battery degradation analysis method, device, equipment and storage medium | |
| CN115951225A (en) | Battery equalization optimization capacity estimation method and device | |
| CN113466722B (en) | Method and device for determining measurement accuracy of battery state of charge and electronic equipment | |
| JP5323396B2 (en) | Input / output characteristic evaluation system and charge / discharge test apparatus incorporating the same | |
| Di Fonso et al. | Data-driven Modeling of Li-ion Battery based on the Manufacturer Specifications and Laboratory Measurements | |
| CN115015787B (en) | Battery SOC compensation method, device and equipment and computer readable storage medium | |
| WO2024115934A1 (en) | Battery deterioration state estimation method and battery deterioration state estimation device |
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 |