Calibration method of battery internal resistance tester

Abstract

The invention discloses a calibration method for a battery internal resistance tester, and the method comprises the following steps: S1, carrying out the N times of measurement of the standard resistance through a battery internal resistance tester in a standard temperature environment, and constructing a training sample; s2, for M standard resistors with different resistance values, repeatedly executing the step S1 to obtain M training samples; s3, in environments with different temperatures, the step S2 is executed repeatedly, and a training set is constructed; s4, constructing a battery internal resistance tester calibration model based on a neural network algorithm and training the battery internal resistance tester calibration model; and S5, when the battery internal resistance tester is used for carrying out actual measurement, carrying out N times of measurement on the to-be-tested battery, then constructing a feature vector of the to-be-tested battery internal resistance tester, inputting the feature vector into the trained battery internal resistance tester calibration model to obtain a deviation, and compensating a measurement result by using the obtained deviation. According to the method provided by the invention, the deviation at different temperatures can be predicted and compensated, and the precision of the battery internal resistance tester is effectively improved.

Description

Technical Field

[0001] The invention relates to the field of instrument calibration, in particular to a calibration method for a battery internal resistance tester. Background Art

[0002] The aging of batteries is often manifested in the physical characteristics of increased internal resistance. After the battery ages, the polarization resistance of the positive and negative electrodes increases, and the external manifestation is the increase in internal resistance. Therefore, the measurement of battery internal resistance is very important in practical engineering applications. This requires the use of a battery internal resistance tester to measure the battery internal resistance. Existing battery internal resistance testers generally use the voltage drop method for measurement, that is, the battery internal resistance is measured by measuring the voltage drop after the current flows through the battery, as well as the current value; the battery internal resistance tester can also test ordinary fixed resistors and standard resistors.

[0003] In order to ensure the measurement accuracy of the battery internal resistance tester, it is often necessary to calibrate the battery internal resistance tester. The existing calibration method generally uses the actual measurement result and the standard result to directly subtract to obtain the calibration value, and the calibration value can be used for direct compensation; however, its actual working process will be affected by temperature, and random fluctuations will occur when measuring the same target. Therefore, the accuracy of the existing calibration method is insufficient to a certain extent. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a calibration method for a battery internal resistance tester, which fully considers the influence of temperature and the fluctuations generated during measurement, constructs a battery internal resistance tester calibration model based on a neural network algorithm, performs training, and uses the trained model to predict deviations, uses the obtained deviations as calibration values, and performs compensation, thereby effectively improving the accuracy of the calibration.

[0005] The object of the present invention is achieved through the following technical scheme: a calibration method for a battery internal resistance tester, comprising the following steps: S1. Under standard temperature environment, use battery internal resistance tester to measure standard resistance N times, obtain N measurement results, calculate the average value, standard deviation and deviation k of the average value relative to the standard resistance, and construct training samples; S2. Repeat step S1 for M standard resistors with different resistance values ​​to obtain M training samples; S3. Repeat step S2 in different temperature environments, add the training samples to the same set, and complete the construction of the training set; S4. construct a battery internal resistance tester calibration model based on a neural network algorithm, and train the battery internal resistance tester calibration model using training samples in a training set; S5. When using the battery internal resistance tester for actual measurement, measure the battery to be tested N times, then construct a feature vector of the battery internal resistance tester to be tested, input it into the trained battery internal resistance tester calibration model, obtain the deviation, and use the obtained deviation to compensate the measurement result.

[0006] The beneficial effects of the present invention are as follows: the present invention fully considers the influence of temperature and the fluctuations generated during measurement, constructs sample features based on temperature, standard deviation of measurement and measurement results, and uses the measured average value and standard value to calculate the measurement deviation as the sample label, constructs a battery internal resistance tester calibration model based on a neural network algorithm for training, and uses the trained model to predict deviations, uses the obtained deviations as calibration values, and performs compensation, thereby effectively improving the accuracy of calibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0008] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0009] like Figure 1 As shown, a calibration method for a battery internal resistance tester is characterized in that it includes the following steps: S1. Under standard temperature environment, use battery internal resistance tester to measure standard resistance N times, obtain N measurement results, calculate the average value, standard deviation and deviation k of the average value relative to the standard resistance, and construct training samples; S101. Under a standard temperature environment, use a battery internal resistance tester to measure the standard resistance N times, and obtain N measurement results recorded as: ; S102. Calculate the average value of N measurement results And standard deviation S: ; in, represents the result of the nth measurement, ; S103. Calculate the deviation k of the measured average value relative to the standard resistance: ; in, Indicates the known resistance value of a standard resistor; S104. Construct training samples: Sample features are composed of temperature T, standard deviation S and N measurement results Composition, recorded as (T, S, ), the sample label is the deviation k from the mean value relative to the standard resistance.

[0010] S2. Repeat step S1 for M standard resistors with different resistance values ​​to obtain M training samples; S3. Repeat step S2 in different temperature environments, add the training samples to the same set, and complete the construction of the training set; In step S3, based on the standard temperature, the temperature is stepped according to the preset temperature. Gradually increase P steps and gradually decrease P steps to obtain 2P temperature points, which together with the standard temperature total 2P+1 temperature points, constituting the different temperatures; Since M training samples are obtained by repeatedly executing step S2 at each temperature, the final training set contains a total of (2P+1)*M training samples.

[0011] S4. construct a battery internal resistance tester calibration model based on a neural network algorithm, and train the battery internal resistance tester calibration model using training samples in a training set; The neural network algorithm adopts the RNN neural network algorithm.

[0012] In step S4, when the neural network algorithm is used to construct a battery internal resistance tester calibration model for training, the sample features in the sample set are used as the model input, and the labels corresponding to the sample features are used as the expected output for training. When all the training samples in the sample set are trained, it is considered that a trained battery internal resistance tester calibration model is obtained.

[0013] S5. When using the battery internal resistance tester for actual measurement, measure the battery to be tested N times, then construct a feature vector of the battery internal resistance tester to be tested, input it into the trained battery internal resistance tester calibration model, obtain the deviation, and use the obtained deviation to compensate the measurement result.

[0014] The step S5 comprises: S501. When using the battery internal resistance tester for actual measurement, firstly measure the battery to be tested N times, and obtain N measurement results recorded as: ; S502. Collect the ambient temperature when the battery internal resistance tester is actually measuring, and select the temperature point closest to the ambient temperature from the 2P+1 temperature points, and record it as ; S503. Calculation The average and standard deviation : ; in, It indicates the result of the nth measurement of the internal resistance of the battery to be tested. ; S504. Construct the characteristic vector of the internal resistance tester of the battery to be tested ( , , ), input the trained battery internal resistance tester calibration model, and the model outputs the deviation, recorded as ; S505. Using the obtained deviation The average value of the measurement results Compensation is performed to obtain the compensated measurement value, which is recorded as : .

[0015] The above is a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A calibration method for a battery internal resistance tester, characterized in that: The following steps are involved: S1. Under standard temperature environment, use battery internal resistance tester to measure standard resistance N times, obtain N measurement results, calculate the average value, standard deviation and deviation k of the average value relative to the standard resistance, and construct training samples; S2. Repeat step S1 for M standard resistors with different resistance values ​​to obtain M training samples; S3. Repeat step S2 in different temperature environments, add the training samples to the same set, and complete the construction of the training set; S4. construct a battery internal resistance tester calibration model based on a neural network algorithm, and train the battery internal resistance tester calibration model using training samples in a training set; S5. When using the battery internal resistance tester for actual measurement, measure the battery to be tested N times, then construct a feature vector of the battery internal resistance tester to be tested, input it into the trained battery internal resistance tester calibration model, obtain the deviation, and use the obtained deviation to compensate the measurement result.

2. The calibration method of a battery internal resistance tester according to claim 1, characterized in that: The step S1 comprises: S101. Under a standard temperature environment, use a battery internal resistance tester to measure the standard resistance N times, and obtain N measurement results recorded as: ; S102. Calculate the average value of N measurement results And standard deviation S: ； in, represents the result of the nth measurement, ; S103. Calculate the deviation k of the measured average value relative to the standard resistance: ； in, Indicates the known resistance value of a standard resistor; S104. Construct training samples: Sample features are composed of temperature T, standard deviation S and N measurement results Composition, recorded as (T, S, ), the sample label is the deviation k from the mean value relative to the standard resistance.

3. The calibration method of a battery internal resistance tester according to claim 1, characterized in that: In step S3, based on the standard temperature, the temperature is stepped according to the preset temperature. Gradually increase P steps and gradually decrease P steps to obtain 2P temperature points, which together with the standard temperature total 2P+1 temperature points, constituting the different temperatures; Since M training samples are obtained by repeatedly executing step S2 at each temperature, the final training set contains a total of (2P+1)*M training samples.

4. The calibration method of a battery internal resistance tester according to claim 1, characterized in that: The neural network algorithm adopts the RNN neural network algorithm.

5. The calibration method of a battery internal resistance tester according to claim 1, characterized in that: In step S4, when the neural network algorithm is used to construct a battery internal resistance tester calibration model for training, the sample features in the sample set are used as the model input, and the labels corresponding to the sample features are used as the expected output for training. When all the training samples in the sample set are trained, it is considered that a trained battery internal resistance tester calibration model is obtained.

6. The calibration method of a battery internal resistance tester according to claim 3, characterized in that: The step S5 comprises: S501. When using the battery internal resistance tester for actual measurement, firstly measure the battery to be tested N times, and obtain N measurement results recorded as: ; S502. Collect the ambient temperature when the battery internal resistance tester is actually measuring, and select the temperature point closest to the ambient temperature from the 2P+1 temperature points, and record it as ; S503. Calculation The average and standard deviation : ； in, It indicates the result of the nth measurement of the internal resistance of the battery to be tested. ; S504. Construct the characteristic vector of the internal resistance tester of the battery to be tested ( , , ), input the trained battery internal resistance tester calibration model, and the model outputs the deviation, recorded as ; S505. Use the obtained deviation to compensate the average value of the measurement result to obtain the compensated measurement value, which is recorded as : 。

Images