CN111413625A - Method and device for obtaining test current - Google Patents
Method and device for obtaining test current Download PDFInfo
- Publication number
- CN111413625A CN111413625A CN202010298806.9A CN202010298806A CN111413625A CN 111413625 A CN111413625 A CN 111413625A CN 202010298806 A CN202010298806 A CN 202010298806A CN 111413625 A CN111413625 A CN 111413625A
- Authority
- CN
- China
- Prior art keywords
- neural network
- current
- predicted
- end voltage
- test
- 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.)
- Pending
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
-
- 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]
- G01R31/3644—Constructional arrangements
- G01R31/3647—Constructional arrangements for determining the ability of a battery to perform a critical function, e.g. cranking
-
- 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]
- G01R31/385—Arrangements for measuring battery or accumulator variables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Secondary Cells (AREA)
Abstract
The embodiment of the application provides a method and a device for obtaining test current, wherein the method comprises the following steps: determining that a first predicted end voltage output by a first neural network meets a target end voltage; providing a first proposed starting current input to the first neural network corresponding to the first predicted ending voltage. The embodiment of the application acquires the test current value of the battery cell through the neural network, improves the intelligent degree of the battery test process, and shortens the test time of the battery (for example, the battery cell).
Description
Technical Field
The application relates to the field of battery testing, in particular to a method and a device for obtaining testing current.
Background
At present, the boundary searching test of the battery consumes time, labor and resources, and occupies resources and manpower of other test items. For example, one tester invests the battery test fully, the test time for completing one battery cell is 2-3 days, and the time is longer under normal conditions. Therefore, how to increase the speed of battery testing becomes an urgent problem to be solved.
Disclosure of Invention
An object of the embodiments of the present application is to provide a method and an apparatus for obtaining a test current, which can significantly improve the speed of a cell test current (or referred to as power) test and shorten the test time.
In a first aspect, an embodiment of the present application provides a method for obtaining a test current, where the method includes: determining that a first predicted end voltage output by a first neural network meets a target end voltage; a first suggested starting current for input to the first neural network corresponding to the first predicted ending voltage is taken as a test current.
The embodiment of the application acquires the test current value of the battery cell through the neural network, improves the intelligent degree of the battery test process, and shortens the test time of the battery (for example, the battery cell).
In some embodiments, the determining that the first predicted end voltage of the first neural network output satisfies the target end voltage comprises: inputting the set test temperature, the battery charge state, the continuous discharge time and the proposed initial current into the first neural network; obtaining the predicted end voltage obtained through the first neural network; and when the predicted end voltage meets the target end voltage, the current suggested starting current is the first suggested starting current.
The embodiment of the application further defines the input parameters of the first neural network to improve the training speed of the first neural network and the accuracy of predicting the predicted end voltage.
In some embodiments, the method further comprises: inputting a priori predicted voltages into the first neural network; the obtaining of the predicted end voltage of the current time through the first neural network includes: and the first neural network predicts the current predicted end voltage according to the test temperature, the battery charge state, the continuous discharge time, the current recommended starting current and the prior predicted voltage.
In the embodiment of the application, the priori predicted voltage is used as an input parameter of the neural network to obtain the predicted end voltage. Because the neural network training needs a large amount of data to be successful, the prior knowledge can simplify the structure of the model, and improve the training speed and the accuracy of model prediction.
In some embodiments, the method further comprises: determining that a second predicted end voltage output by a second neural network meets the target end voltage at least according to the first suggested start current and an end voltage obtained by actual testing; using a second suggested starting current for input to the second neural network corresponding to the second predicted ending voltage as the test current.
According to the embodiment of the application, the series neural network is adopted to obtain the proposed current, so that the accuracy of obtaining the test current is improved, and the time for obtaining the test current is shortened.
In some embodiments, said determining that a second predicted voltage output by a second neural network meets the target ending voltage based at least on the first proposed starting current and an actual tested ending voltage comprises: inputting the set test temperature, the battery state of charge, the continuous discharge time, the end voltage obtained by the actual test and the first suggested starting current into the second neural network; obtaining the predicted ending voltage obtained through the second neural network; and determining whether the predicted end voltage meets the target end voltage.
The embodiment of the application increases parameters for the input of the second neural network connected in series, and improves the accuracy of the obtained pulse current.
In some embodiments, the method further comprises: inputting a priori predicted voltage into the second neural network; the obtaining of the predicted end voltage obtained through the second neural network includes: and the second neural network predicts and obtains the predicted end voltage according to the test temperature, the battery charge state, the continuous discharge time, the end voltage obtained by the actual test, the first suggested initial current and the prior predicted voltage.
The input parameters of the second neural network in the embodiment of the application also comprise priori predicted voltage obtained according to the priori knowledge, and due to the fact that the priori knowledge is added to the input parameters of the neural network, the training times of the second trial network are reduced at one time, and the accuracy of the pulse current obtained by the second neural network is improved.
In some embodiments, the method further comprises: training the first neural network or the second neural network through historical data, wherein the first neural network and the second neural network are fully connected networks.
The first neural network and the second neural network of the embodiments of the present application may be fully connected neural networks.
In some embodiments, the method further comprises: resampling sample data of the first type battery through particle filtering and lossless Kalman filtering; adjusting the last layer parameter of the full-connection network by adopting the resampled sample data; or, adjusting an attenuation factor by using the resampled sample data.
According to the embodiment of the application, two filtering modes are adopted for sampling sample data of the new-type battery, and the performance of the neural network on different types of batteries is improved.
In some embodiments, the method further comprises: obtaining the attenuation factor of the second type battery according to the historical data; and obtaining the prior predicted voltage according to the attenuation factor.
According to the embodiment of the application, the priori prediction voltage is obtained through the attenuation factor, the structure of the neural network model can be simplified through the priori knowledge, and the training speed and the accuracy of model prediction are improved.
In a second aspect, an embodiment of the present application provides an apparatus for obtaining a test current, the apparatus including: a first neural network; a determination module configured to determine that a first predicted end voltage output by the first neural network satisfies a target end voltage; an output module configured to take a first suggested starting current corresponding to the first predicted ending voltage for input to the first neural network as a test current.
In some embodiments, the first neural network is configured to predict the predicted end voltage according to the input test temperature, the battery state of charge, the discharge duration and the recommended start current.
In some embodiments, the apparatus further comprises: and the second neural network is configured to predict and obtain the predicted end voltage according to the input test temperature, the battery charge state, the continuous discharge time, the end voltage obtained by actual test and the first suggested initial current.
In a third aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, can implement the method of the first aspect.
In a fourth aspect, an embodiment of the present application provides an information processing apparatus, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the method according to the first aspect.
In a fifth aspect, embodiments of the present application provide a computer program product, which when run on a computer, causes the computer to execute the method of the first aspect or any possible implementation manner of the first aspect.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a schematic diagram of a cell testing process in the prior art;
fig. 2 is a schematic diagram of a cell testing process provided in an embodiment of the present application;
FIG. 3 is a flowchart of a method for obtaining a test current according to an embodiment of the present disclosure;
fig. 4 is a schematic diagram of a cell testing process provided in an embodiment of the present application;
FIG. 5 is a block diagram of an apparatus for obtaining a test current according to an embodiment of the present disclosure;
fig. 6 is a flowchart of an information processing apparatus provided in an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
At present, the battery boundary-finding test consumes time and resources, occupies resources of other test items and occupies manpower, and the current required by the battery test is obtained by machine learning in the embodiment of the application, so that the boundary-finding test work is more automatic and intelligent, and the test efficiency is improved. The method for determining the boundary finding test of the battery is to evaluate whether the cut-off voltage of the battery reaches the cut-off condition under the selected multiplying power (or called current) to determine whether the selected pulse multiplying power (i.e. the suggested pulse current) is appropriate or not within a certain discharge duration under the condition of a fixed battery state of charge (SOC) and temperature.
The following describes a boundary finding test procedure of a battery according to the related art with reference to fig. 1.
Referring to fig. 1, the battery test of fig. 1 includes the following steps:
the duration (i.e., the duration of discharge of the battery) is selected S101.
And S102, adjusting the temperature to obtain the current test temperature.
And S103, adjusting the SOC of the battery to obtain the SOC required by the test.
S104, initial current selection is suggested.
S105, judging whether the lower limit cut-off voltage meets a target cut-off condition, namely judging whether the end voltage of the battery to be tested meets the lower limit cut-off voltage (or called a target end voltage) under the condition of the selected pulse multiplying power or called a suggested initial current, and executing S107 when the cut-off condition is met; otherwise, S106 is executed.
It should be noted that, under the condition of a fixed battery state of charge SOC point and temperature, in a certain discharge duration, if the battery voltage does not reach the cut-off condition under the selected rate (corresponding to a certain suggested starting current), the battery test is considered to fail.
And S106, selecting a larger pulse multiplying power (power) point upwards to continue testing (the difference between the two multiplying powers needs to be manually judged, and the action is obtained by workers according to work experience at present).
And S107, considering the current multiplying power (power) as the magnitude of the test current (power) at the SOC and the temperature until the cell voltage reaches a cut-off condition.
S108, judging whether other test conditions are set, and repeating the steps if the next SOC point is switched;
and S109, switching the temperature after all SOC points are measured at the temperature, and repeating the steps.
In order to improve the randomness and inaccuracy of selecting pulse multiplying power or testing current in the battery testing process of the related technology, the pulse multiplying power of S106 (a certain pulse multiplying power corresponds to a certain pulse current) is obtained through the neural network in the embodiment of the application, the testing current obtained through the embodiment of the application can effectively reduce the testing times of the cell voltage reaching the cut-off condition, the battery measuring speed is effectively improved, and the dependence degree of the battery testing of the related technology on manpower is eliminated.
The following describes, with reference to fig. 2, an exemplary embodiment of the present application at a specific stage in a battery testing process of the related art.
As shown in fig. 2, the test procedure of the battery using the technical solution of the embodiment of the present application includes the following steps.
S201, starting the test.
S202, acquiring the pulse current suggested to be adopted based on the neural network.
And S203, using the suggested current to carry out actual test to obtain the end voltage of the battery to be tested.
And S204, judging whether the ending voltage meets the requirement, if so, ending the current round of test on the battery, and if not, executing S205.
And S205, inputting the actually tested pulse current and the ending voltage into the neural network, and repeatedly executing the steps from S202 to S204.
It should be noted that some steps in S203-S206 of fig. 2 require human intervention. The technical solution of the embodiment of the present application corresponds to S202 of fig. 2. That is, the technical purpose of the embodiments of the present application is to provide a suggested pulse current for a battery test, by which the number of times of the battery test can be effectively reduced.
The process of obtaining the test current (i.e., S202 of fig. 2) in the embodiment of the present application is described below with reference to fig. 3.
To implement the method of fig. 3 the embodiment of the present application requires at least two neural networks, namely a first neural network and a second neural network.
The training process for the first and second neural networks is briefly described below. The first neural network and the second neural network are trained through historical data (for example, input data and corresponding output data), wherein the first neural network or the second neural network is trained to adjust parameters of the neural network according to the end voltage data of an actual test obtained according to the test condition (namely, the input data) so as to have the function of predicting the end voltage according to the test condition of the battery to be tested. Taking the first neural network as an example, the historical data of the first neural network includes the battery state of charge SOC, the test temperature, the sustained discharge time, the initial internal resistance, the initial voltage, the battery size (i.e., the battery capacity, in Ah) and the battery model (e.g., model a1, a2, B1 or B2) of the input neural network, and the historical data also includes the actual end voltage obtained by actual testing according to the test conditions (i.e., the test temperature, the sustained discharge time, the initial internal resistance, the initial voltage, the battery size, the initial weight, the battery model or model, and the battery state of charge). The input data types of the second neural network model include SOC, temperature, duration, initial internal resistance, initial voltage, battery size (i.e., battery capacity, unit Ah), battery model (e.g., model a1, a2, B1, or B2), last start current (i.e., first recommended start current resulting from performing the first actual test), and last end voltage (i.e., actual end voltage resulting from performing the first actual test), and the output is the predicted end voltage. That is, the second neural network has two more parameters than the input to the first neural network: last start current, last end voltage. Therefore, when training the second neural network, the test condition also needs to increase two parameters of the last starting current and the last ending voltage. The first suggested start current and the actual test derived end voltage for the last start current and last end voltage, respectively, below.
It should be noted that in order to train the first neural network and the second neural network, in the embodiment of the present application, a test data database needs to be established, and statistics on the cell test history data of the sample needs to be performed. The data in the test data database includes sets of input and output data, for example, the historical data includes sets of collected input data (e.g., SOC, temperature, duration (i.e., battery duration), initial internal resistance, initial voltage, battery size, initial weight, battery model, or model) and corresponding output data (e.g., end voltage corresponding to actual testing of various input data).
As shown in fig. 3, an embodiment of the present application provides a method for obtaining a test current (based on the trained neural network), where the method includes: s310, determining that a first predicted end voltage output by a first neural network meets a target end voltage; s320, using a first suggested start current corresponding to the first predicted end voltage for input to the first neural network as a test current.
In S310, the first neural network is a trained first neural network, and is used to obtain a predicted value of the charge or discharge end voltage. When the obtained predicted value satisfies the target end voltage (i.e., the lower limit cut-off voltage at the time of discharge), the first predicted end voltage of S310 is obtained.
Optionally, as an example, S310 may include: inputting the set test temperature, the battery charge state, the continuous discharge time and the proposed initial current into the first neural network; obtaining the predicted end voltage obtained through the first neural network; and when the predicted end voltage meets the target end voltage, the current suggested starting current is the first suggested starting current.
It should be noted that the parameters input into the first neural network may further include a model number, an initial internal resistance, an initial voltage, an initial weight, or a model of the battery to be tested, and the above examples only provide some parameters by way of example. For a certain type of battery to be tested, the suggested initial current input into the first neural network for the first time can be given a more accurate recommended value by counting the historical data of the test of the battery core of the sample of the type. If the predicted voltage end value obtained by current prediction of the current recommendation test does not meet the target end voltage, the isolation between the next recommended starting current and the current recommended starting current can be selected according to experience, and the current range of each recommended starting current can be set according to experience. For example, the respective proposed starting currents may be: 2.5A, 2.75A, 3.00A, or 3.25A, etc., each time suggesting an initial galvanic isolation of 0.25A in this example.
To improve the accuracy of the acquired test current, the method of fig. 3 may further include: inputting a priori predicted voltages into the first neural network; s310 may further include: and obtaining the predicted ending voltage of the time by predicting the first neural network according to the test temperature, the battery charge state, the continuous discharge time, the recommended starting current of the time and the prior predicted voltage.
To further improve the accuracy of the obtained test current, as an example, the method may further include, after S320: s330, determining that a second predicted end voltage output by a second neural network meets the target end voltage at least according to the first suggested start current and an end voltage obtained by actual test; s340, using a second suggested starting current corresponding to the second predicted ending voltage and input into the second neural network as the test current. It should be noted that after the step S320 in fig. 3 is executed, the relevant tester may perform an actual testing process on the battery to be tested according to the first suggested starting current. If the first suggested starting current does not satisfy the condition (for example, the ending voltage of the battery to be tested at the first suggested starting current does not satisfy the lower limit cut-off voltage condition of the discharge process) through practical tests, the method for acquiring the test current according to the embodiment of the present application may continue to perform S330 and S340 of the present example.
Since the first proposed starting current obtained by the first neural network may be inaccurate, the second proposed starting current obtained by the second neural network with two added input parameters (the first proposed starting current and the end voltage of the actual test) is more accurate.
Optionally, as an example, S330 may include: inputting the set test temperature, the battery state of charge, the continuous discharge time, the end voltage obtained by the actual test and the first suggested starting current into the second neural network; obtaining the predicted ending voltage of the time through the second neural network; and determining whether the predicted end voltage meets the target end voltage.
In order to improve the accuracy of the proposed starting current obtained by the second neural network, an a priori predicted voltage obtained according to a priori knowledge may be input into the second neural network, and as an example, S330 may further include: and obtaining the predicted end voltage of the current time by the second neural network according to the test temperature, the battery charge state, the pulse duration discharge time, the end voltage obtained by the actual test, the first suggested initial current and the prior predicted voltage.
It can be understood that, for the purpose of enabling the first neural network and the second neural network to predict the ending voltage according to the input data, the method of the embodiment of the present application further includes: training the first neural network or the second neural network through historical data, wherein the first neural network and the second neural network are fully connected networks.
In order to promote the application scenarios of the first neural network and the second neural network (for example, to meet the test of a new model of battery to be tested), as an example, the method further comprises the following steps: resampling sample data of the first type battery through particle filtering and lossless Kalman filtering; and adjusting the last layer parameter of the fully-connected network by adopting the resampled sample data, or adjusting an attenuation factor by adopting the resampled sample data.
To leverage a priori knowledge reduction to achieve a reduction in the process of training the neural network, as one example, the method further comprises: obtaining the attenuation factor of the second type battery according to the historical data; and obtaining the prior predicted voltage according to the attenuation factor.
Taking the discharge process of the battery as an example, the formula related to the attenuation factor is as follows:
V＝a-0.0005*Q+log(b-Q)
a＝4–0.25*I
b＝6000–250*I
where V is the end voltage of the battery discharge, I is the discharge current of the battery, and Q is the amount of battery discharge (which is the integral of the current over time, and can be specifically calculated from the pulse "duration discharge time" input into the first or second neural network) is the integral of I over time t. For different cells or different temperature decay factors, where the linear part: a-4-0.25 (1 +)/curve section: b 6000-.
Specifically, based on all the acquired battery sample data, the basic morphology of the curve is obtained by least squares calculation for the assumption of domain knowledge.
The process of testing the battery core by using the trained first neural network and the trained second neural network is described below with reference to fig. 4. It should be noted that, in fig. 4, the charge/discharge test 1 refers to a process of performing a first actual test of charging/discharging the battery cell by using a first recommended starting current output by the first neural network, and the charge/discharge test 2 refers to a process of performing a second actual test of charging/discharging the battery cell by using a second recommended starting current output by the second neural network.
The first neural network 430 in fig. 4 is configured to obtain the predicted end voltage of this time according to input parameters (for example, including SOC, temperature, model, sustained discharge time, initial internal resistance, initial voltage, and the suggested start current of this time in fig. 4, it is to be noted that only a part of the input parameters are shown as 4) obtained by the set test conditions, and then, in the embodiment of the present application, it is further determined whether the predicted end voltage output by the first neural network 430 in the prediction meets the target end voltage, and when the target end voltage is met, the suggested start current of this time is output as the first suggested start current for the first actual test. When the ending voltage of the battery to be tested is tested according to the first suggested starting current, if the ending voltage obtained by the actual test does not reach the standard, the ending voltage obtained by the actual test and the first suggested starting current are input into the second neural network 440 to be processed again. Different from the first neural network 430, the second neural network 440 includes an end voltage (i.e., an end voltage of the charge/discharge test 1 shown in fig. 4) obtained by the first actual test and a first suggested start current in addition to the input parameters of the first neural network 430, the second neural network 440 obtains the predicted end voltage according to the input parameters, and then the present application determines the input suggested start current corresponding to the predicted end voltage satisfying the target end voltage as the second suggested start current. The second recommended starting current is obtained for the second actual test (i.e., the charge/discharge test 2 of fig. 4), which is also to perform the test procedure as shown in fig. 1 (it should be noted that the pulse rate selection of fig. 1 is also the second recommended starting current) or fig. 2 (i.e., S203 and S204 of fig. 2) for the battery to be tested.
Referring to fig. 5, fig. 5 shows a device for obtaining a test current provided by an embodiment of the present application, it should be understood that the device corresponds to the embodiment of the method of fig. 3, and can perform various steps related to the embodiment of the method, and the specific functions of the device may be referred to the description above, and a detailed description is appropriately omitted herein to avoid redundancy. The device comprises at least one software functional module which can be stored in a memory in the form of software or firmware or be fixed in an operating system of the device, and the device for acquiring the test current may comprise: a first neural network 500 (corresponding to the first neural network 430 of fig. 4); a determining module 501 configured to determine that a first predicted end voltage output by the first neural network 500 meets a target end voltage; an output module 502 configured to take a first suggested start current for input to the first neural network corresponding to the first predicted end voltage as a test current.
As an example, the first neural network 500 is configured to predict the predicted end voltage according to the input test temperature, the battery state of charge, the discharge duration and the recommended start current.
As one example, the apparatus further includes a second neural network (not shown in fig. 5) configured to predict a present predicted end voltage based on the input test temperature, battery state of charge, pulse duration, actual test-derived end voltage, and the first proposed start current.
Embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, and the computer program can implement the method described in fig. 3 when being executed by a processor.
As shown in fig. 6, an information processing apparatus 600 according to an embodiment of the present application includes a memory 610, a processor 620, and a computer program stored in the memory 610 and executable on the processor 620, where the processor 620 may implement the method shown in fig. 3 when executing the computer program. For example, the processor 620 of fig. 6 reads a program from the memory 610 through the bus 630 and executes it.
Execution of the program by processor 620 may implement the method illustrated in fig. 3. For example, the processor executing the program may determine whether the first suggested starting current satisfies the target ending voltage, may implement a function of the first neural network (specifically, a function of predicting the voltage this time according to the input parameter), may implement a function of the second neural network (specifically, a function of predicting the voltage this time according to the input parameter).
Processor 520 may process digital signals and may include various computing structures. Such as a complex instruction set computer architecture, a structurally reduced instruction set computer architecture, or an architecture that implements a combination of instruction sets. In some examples, processor 620 may be a microprocessor.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (14)
1. A method of obtaining a test current, the method comprising:
determining that a first predicted end voltage output by a first neural network meets a target end voltage;
a first suggested starting current for input to the first neural network corresponding to the first predicted ending voltage is taken as a test current.
2. The method of claim 1, wherein the determining that the first predicted end voltage of the first neural network output satisfies the target end voltage comprises:
inputting the set test temperature, the battery charge state, the continuous discharge time and the proposed initial current into the first neural network;
obtaining the predicted end voltage obtained through the first neural network;
and when the predicted end voltage meets the target end voltage, the current suggested starting current is the first suggested starting current.
3. The method of claim 2, wherein the method further comprises:
inputting a priori predicted voltages into the first neural network;
the obtaining of the predicted end voltage of the current time through the first neural network includes: and obtaining the predicted ending voltage of the time by predicting the first neural network according to the test temperature, the battery charge state, the continuous discharge time, the recommended starting current of the time and the prior predicted voltage.
4. The method of obtaining a test current of claim 1, further comprising:
determining that a second predicted end voltage output by a second neural network meets the target end voltage at least according to the first suggested start current and an end voltage obtained by actual testing;
using a second suggested starting current for input to the second neural network corresponding to the second predicted ending voltage as the test current.
5. The method of deriving a test current according to claim 4, wherein said determining that a second predicted voltage output by a second neural network meets the target ending voltage based at least on the first proposed starting current and an actual tested ending voltage comprises:
inputting the set test temperature, the battery state of charge, the continuous discharge time, the end voltage obtained by the actual test and the first suggested starting current into the second neural network;
obtaining the predicted ending voltage obtained through the second neural network;
and determining whether the predicted end voltage meets the target end voltage.
6. The method of claim 5, wherein the method further comprises:
inputting a priori predicted voltage into the second neural network;
the obtaining of the predicted end voltage obtained through the second neural network includes: and obtaining the predicted end voltage of the current time predicted by the second neural network according to the test temperature, the battery charge state, the continuous discharge time, the end voltage obtained by the actual test, the first suggested initial current and the prior predicted voltage.
7. The method of claim 4, wherein the method further comprises: training the first neural network or the second neural network through historical data, wherein the first neural network and the second neural network are fully connected networks.
8. The method of claim 7, wherein the method further comprises:
resampling sample data of the first type battery through particle filtering and lossless Kalman filtering;
and adjusting and optimizing the last layer of parameters of the fully-connected network by adopting the resampled sample data.
9. The method of claim 3 or 6, further comprising:
obtaining the attenuation factor of the second type battery according to the historical data;
and obtaining the prior predicted voltage according to the attenuation factor.
10. An apparatus for obtaining a test current, the apparatus comprising:
a first neural network;
a determination module configured to determine that a first predicted end voltage output by the first neural network satisfies a target end voltage;
an output module configured to take a first suggested starting current corresponding to the first predicted ending voltage for input to the first neural network as a test current.
11. The apparatus of claim 10, wherein the first neural network is configured to predict a current predicted end voltage based on the input test temperature, battery state of charge, duration of discharge, and a current proposed start current.
12. The apparatus of claim 11, wherein the apparatus further comprises: and the second neural network is configured to predict and obtain the predicted end voltage according to the input test temperature, the battery charge state, the continuous discharge time, the end voltage obtained by actual test and the first suggested initial current.
13. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, is adapted to carry out the method of any one of claims 1 to 9.
14. An information processing apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program is operable to implement the method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010298806.9A CN111413625A (en) | 2020-04-16 | 2020-04-16 | Method and device for obtaining test current |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010298806.9A CN111413625A (en) | 2020-04-16 | 2020-04-16 | Method and device for obtaining test current |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111413625A true CN111413625A (en) | 2020-07-14 |
Family
ID=71491870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010298806.9A Pending CN111413625A (en) | 2020-04-16 | 2020-04-16 | Method and device for obtaining test current |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111413625A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201524077A (en) * | 2013-12-03 | 2015-06-16 | Nat Inst Chung Shan Science & Technology | Battery health status analysis method |
| CN108199109A (en) * | 2018-01-16 | 2018-06-22 | 上海应用技术大学 | The screening technique that a kind of retired power battery pack echelon utilizes |
| CN108802625A (en) * | 2018-06-20 | 2018-11-13 | 国网江苏省电力有限公司电力科学研究院 | A kind of SOC adaptive correction methods of secondary use battery |
| CN110515001A (en) * | 2019-09-07 | 2019-11-29 | 创新奇智(广州)科技有限公司 | A kind of two stages battery performance prediction technique based on charge and discharge |
-
2020
- 2020-04-16 CN CN202010298806.9A patent/CN111413625A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201524077A (en) * | 2013-12-03 | 2015-06-16 | Nat Inst Chung Shan Science & Technology | Battery health status analysis method |
| CN108199109A (en) * | 2018-01-16 | 2018-06-22 | 上海应用技术大学 | The screening technique that a kind of retired power battery pack echelon utilizes |
| CN108802625A (en) * | 2018-06-20 | 2018-11-13 | 国网江苏省电力有限公司电力科学研究院 | A kind of SOC adaptive correction methods of secondary use battery |
| CN110515001A (en) * | 2019-09-07 | 2019-11-29 | 创新奇智(广州)科技有限公司 | A kind of two stages battery performance prediction technique based on charge and discharge |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3709039A1 (en) | Battery life learning device, battery life prediction device, method and program | |
| KR102633335B1 (en) | Method for detecting abnomal cells and estimating SOH in Lithium-Ion battery pack | |
| WO2015041093A1 (en) | Device and method for evaluating performance of storage cell | |
| JP7300878B2 (en) | BATTERY EVALUATION SYSTEM, BATTERY EVALUATION METHOD AND PROGRAM | |
| US20210173012A1 (en) | Method and system for estimation of open circuit voltage of a battery cell | |
| CN104956233A (en) | Battery state estimating device | |
| CN114290960B (en) | Method and device for acquiring battery health of power battery and vehicle | |
| WO2018112818A1 (en) | Rapid prediction method for cycle life of battery and rapid prediction device therefor | |
| CN116381541B (en) | Health assessment method and system for energy storage lithium battery system | |
| CN110806540B (en) | Battery cell test data processing method, device and system and storage medium | |
| CN116466236A (en) | Battery remaining life prediction method, device, equipment and readable storage medium | |
| CN115453376A (en) | State monitoring method and system for power battery of electric automobile and electric automobile | |
| CN116626502A (en) | Battery capacity prediction method, device, equipment and storage medium | |
| CN115392134A (en) | Vehicle power battery comprehensive performance evaluation method and related device | |
| CN115663980B (en) | Aircraft battery management system, method, computer equipment and storage medium | |
| CN111413625A (en) | Method and device for obtaining test current | |
| CN116819364A (en) | Battery health determination method, application method, electronic device and storage medium | |
| CN117129895A (en) | Battery state of health calculation method, device, storage medium and vehicle | |
| CN110568362B (en) | Method and device for determining cell performance parameters | |
| CN113447821A (en) | Method for estimating state of charge of battery | |
| CN112394290A (en) | Method and device for estimating SOH of battery pack, computer equipment and storage medium | |
| CN116224091B (en) | Battery cell fault detection method and device, electronic equipment and storage medium | |
| CN116125281A (en) | Battery throughput prediction method and device, electronic equipment and storage medium | |
| CN117031296A (en) | Battery system performance evaluation method based on boundary structure | |
| CN116540096A (en) | Battery bulge and service life prediction method, electrochemical device and electric equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200714 |