Transformer substation power supply system battery fault rapid positioning method and system based on impedance characteristics

Abstract

The invention discloses a transformer substation power supply system battery fault rapid positioning method based on online electrochemical impedance spectroscopy measurement and impedance characteristics, and the method comprises the steps: selecting a periodic square wave as a standard excitation current for impedance measurement; and completing time-frequency conversion of voltage and current by using Morlet wavelets so as to obtain an impedance frequency spectrum, fitting to obtain capacitance-resistance parameters in the impedance frequency spectrum, and carrying out cumulative probability distribution operation to detect the battery fault of the power supply system of the transformer substation. Compared with an existing fault detection method, the method has the advantages of being high in linearity and accuracy and high in applicability, compared with an existing transformer substation power system battery fault detection method, battery performance under different conditions can be evaluated, precision is high, and operation is easy. The battery pack fault rapid positioning method based on online electrochemical impedance spectroscopy measurement and impedance characteristics has the advantages of being simple, low in cost, high in efficiency, high in precision and capable of being online.

Description

Technical Field

[0001] The invention relates to the technical field of lithium-ion batteries, and in particular to a method and system for quickly locating a battery fault in a power supply system of a substation based on impedance characteristics. Background Art

[0002] With the rapid development of energy storage systems, new energy vehicles and other fields, lithium battery packs are used more and more widely as core components. With the increase in the use of lithium batteries, the number of reports of their failures has also gradually increased, which requires researchers and companies to conduct in-depth research on the fault location and analysis technology of lithium battery packs. When a failure occurs, since lithium batteries are a nonlinear and complex electrochemical system with complex internal mechanisms and lack of accurate mathematical models, it is difficult to diagnose and separate the fault mechanism.

[0003] The existing fault diagnosis scheme is: through the fault diagnosis technology of power lithium batteries, it is responsible for detecting and evaluating slow-changing faults during the operation of lithium batteries, and predicting and identifying sudden faults. A mature fault diagnosis system not only records the time of occurrence of the fault and the external working conditions, but also can evaluate the health status of the battery system based on the external characteristic vector during battery operation, warn of potential fault risks, and identify the internal mechanism of the fault that has occurred. At present, the fault location and analysis technology of lithium battery packs has made certain progress, but it still needs to be continuously studied and improved. With the continuous development of related technologies, the fault location and analysis of lithium batteries in the future will be more accurate, efficient and intelligent.

[0004] Therefore, quickly and accurately estimating the battery pack fault location and taking corresponding measures can increase the life of the battery pack and prevent the battery pack from suffering greater losses due to single cell failure. Based on this, it is imperative for technicians in this field to propose a fast and stable method for quickly locating battery pack faults. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for quickly locating battery faults in a substation power supply system based on impedance characteristics. This method and system are based on online electrochemical impedance spectrum measurement and impedance characteristics. A periodic square wave is selected as a standard excitation current for impedance measurement, and Morlet wavelet is used to complete the time-frequency conversion of voltage and current to obtain an impedance spectrum and fit the resistance value therein, and cumulative probability distribution calculation is performed to obtain the time and location of the fault. The method has high accuracy and fast calculation speed, and can be better applied in practice.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for quickly locating a battery fault in a substation power system based on impedance characteristics comprises the following steps:

[0008] Step 1, performing a real-time electrochemical impedance spectroscopy (EIS) test on the battery pack;

[0009] Step 2, performing ECM (error correction model) fitting on the EIS test results in step 1 to obtain a corresponding resistance data set;

[0010] Step 3, the resistance data set obtained in step 2 is put into cumulative probability distribution operation to obtain the cost performance index CPD calculation result;

[0011] Step 4, judging whether a fault occurs according to the calculation result in step 3, and stopping the battery operation if a fault occurs;

[0012] Step 5, mapping the time when the fault occurs in step 4 to the experimental time length in step 2, so that the type and location of the fault can be determined.

[0013] Furthermore, in step 1, a periodic square wave is selected as a standard excitation current for impedance measurement, and Morlet wavelet is used to complete the time-frequency conversion of voltage and current to obtain an impedance spectrum.

[0014] Furthermore, in step 2, the real-time EIS test results of step 1 are fitted into a second-order equivalent model through ECM to form a resistance data set; and data with obvious changes are selected from the resistance data set as data features, and the fault location of the battery pack is determined by the selected data features.

[0015] Furthermore, the resistance data set includes a main circuit resistance data set R0 and a first RC unit resistance data set R1

[0016] The main circuit resistance data set R0 is expressed as:

[0017]

[0018] The first RC unit resistance data set R1 is represented by:

[0019]

[0020] Wherein, m means that the battery pack has m battery cells; R 0(m) (t) represents the tth R0 resistance data point of the mth battery cell in the battery pack; R 1(m) (t) represents the tth R1 resistance data point of the mth battery cell.

[0021] Furthermore, the step 3 puts the resistance data set into a cumulative probability distribution algorithm (CPD), including the following steps:

[0022] Step 3-1: Obtain the maximum R corresponding to each battery cell through the resistance data sets R0 and R1 obtained by ECM fitting 0max and R 1max sequence;

[0023]

[0024] Where R 0max (t), R 1max (t) is the maximum resistance of the battery cell in the battery pack at the tth data point;

[0025] The maximum R obtained from each battery cell 0max and R 1max The sequence is

[0026] R 0max = {R 0max (1)…R 0max (t)}(5)

[0027] R 1max = {R 1max (1)…R 1max (t)}(6)

[0028] Step 3-2: 0max and R 1max The data in the sequence dataset are rearranged in sets D0 and D1 in increasing order, namely:

[0029]

[0030] Among them, R 0[1] For R 0max The smallest data in a sequence data set; R 0[t] For R 0max The largest data set in the sequence data set; R 1[1] For R 1max The smallest data in a sequence data set; R 1[t] For R 1max The largest data set in the sequence data set;

[0031] Step 3-3: Introduce numerical interval C

[0032]

[0033] Wherein, L represents the number of battery cells in the battery pack, which is L; I = 0, 1, 2, ..., L-1; C I+1 Indicates the C interval corresponding to the I+1th battery cell;

[0034] Step 3-4: Based on the sets D0 and D1 established in step 3-2 and the numerical interval C established in step 3-3, calculate the number of data points falling into each numerical interval in turn, and calculate the probability of a data point falling into the corresponding data interval. The calculation formula is as follows:

[0035]

[0036] Where P I+1 is the corresponding probability; K I+1 is the number of data points falling into the corresponding numerical interval; t represents the number of data points;

[0037] Step 3-5: Summarize the calculation results in step 3-4 to obtain the CPD calculation results of the resistance data sets R0 and R1, namely:

[0038]

[0039] Furthermore, in step 4, the calculation result in step 3 is compared with a set threshold value. When the calculation result is lower than the set threshold value, it indicates that a fault exists within the voltage range.

[0040] Furthermore, the threshold is determined by different battery pack experiments and is more accurately set according to its fault value. In actual tests, it is found that when the probability P of the calculation result in step 3 falling into the corresponding data interval is greater than 0.0075, a fault will exist; at this time, the corresponding threshold is 0.0075.

[0041] Furthermore, the types of faults described in step 5 are short circuit faults and open circuit faults; when a short circuit fault occurs, the selected characteristic data R0', R1' will decrease sharply; when an open circuit fault occurs, the selected characteristic data R0', R1' will increase sharply.

[0042] The present invention also proposes a system for quickly locating battery faults in a substation power system based on impedance characteristics, the system comprising:

[0043] Electrochemical impedance spectroscopy (EIS) test module, used to perform real-time electrochemical impedance spectroscopy (EIS) testing on battery packs;

[0044] The error correction model ECM fitting module is used to fit the test results obtained by the electrochemical impedance spectroscopy test module to obtain the corresponding resistance data set;

[0045] The cumulative probability distribution operation module is used to operate the resistance data set obtained by the error correction model ECM fitting module;

[0046] A battery fault detection module is used to make judgments based on the results obtained by the cumulative probability distribution calculation module and detect battery faults;

[0047] The battery fault location module is used to determine the fault type and location.

[0048] The beneficial effects of the present invention are as follows: compared with the existing fault detection methods, the method has the characteristics of online linearity, high accuracy and strong applicability, and compared with the existing substation power system battery fault detection methods, it is helpful to evaluate the battery performance under different conditions, has high accuracy and is easy to operate. The method for rapid location of substation power system battery faults based on online electrochemical impedance spectroscopy measurement and impedance characteristics has the advantages of simplicity, low cost, high efficiency, high accuracy and online, which reduces the consumption of manpower and material resources, lays a foundation for research such as determination of substation battery group fault location, is of great significance to battery fault determination, and has certain engineering practice value. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present invention will be further described below in conjunction with the accompanying drawings:

[0050] Figure 1 It is a schematic flow diagram of the method of the present invention;

[0051] Figure 2 is the equivalent circuit diagram obtained after the error correction model ECM fitting in Example 1;

[0052] Figure 3 is a diagram of the fault detection result obtained by the method of the present invention in Example 2;

[0053] Figure 4 This is a block diagram of the system structure of the present invention in Example 3. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0055] See attached Figure 1 The present invention provides a method for quickly locating a battery pack fault based on online electrochemical impedance spectroscopy measurement and cumulative probability distribution algorithm, comprising the following steps:

[0056] Step 1, performing a real-time electrochemical impedance spectroscopy (EIS) test on the battery pack;

[0057] Step 2, performing ECM (error correction model) fitting on the EIS test result in step 1 to obtain a corresponding resistance data set, wherein the resistance data set includes a main circuit resistance data set R0 and a first RC unit resistance data set R1;

[0058] Step 3, the resistance data set obtained in step 2 is put into cumulative probability distribution operation to obtain the cost performance index CPD calculation result;

[0059] Step 4, judging whether a fault occurs according to the calculation result in step 3, and stopping the battery operation if a fault occurs;

[0060] Step 5, mapping the time when the fault occurs in step 4 to the experimental time length in step 2, so as to determine the type and location of the fault; that is, determining the location of the abnormal data interval in the main circuit data set and the first RC unit resistance data set R1 through the calculation results in step 3 combined with the time of the fault occurrence.

[0061] Embodiment 1:

[0062] In step 1, a periodic square wave is selected as the standard excitation current for impedance measurement, and Morlet wavelet is used to complete the time-frequency conversion of voltage and current to obtain the impedance spectrum.

[0063] See attached Figure 2 In step 2, the real-time EIS test results of step 1 are fitted into a second-order equivalent model through ECM to form a resistance data set; and data with obvious changes are selected from the resistance data set as data features, and the fault location of the battery pack is determined by the selected data features.

[0064] The main circuit resistance data set R0 is expressed as:

[0065]

[0066] The first RC unit resistance data set R1 is represented by:

[0067]

[0068] Wherein, m means that the battery pack has m battery cells; R 0(m) (t) represents the tth R0 resistance data point of the mth battery cell in the battery pack; R 1(m) (t) represents the tth R1 resistance data point of the mth battery cell.

[0069] Step 3: Put the resistance data set into the cumulative probability distribution algorithm (CPD), including the following steps:

[0070] Step 3-1: Obtain the maximum R corresponding to each battery cell through the resistance data sets R0 and R1 obtained by ECM fitting0max and R 1max sequence;

[0071]

[0072] Where R 0max (t), R 1max (t) is the maximum resistance of the battery cell in the battery pack at the tth data point;

[0073] The maximum R obtained from each battery cell 0max and R 1max The sequence is

[0074] R 0max = {R 0max (1)…R 0max (t)}(5)

[0075] R 1max = {R 1max (1)…R 1max (t)}(6)

[0076] Step 3-2: 0max and R 1max The data in the sequence dataset are rearranged in sets D0 and D1 in increasing order, namely:

[0077]

[0078] Among them, R 0[1] For R 0max The smallest data in a sequence data set; R 0[t] For R 0max The largest data set in the sequence data set; R 1[1] For R 1max The smallest data in a sequence data set; R 1[t] For R 1max The largest data set in the sequence data set;

[0079] Step 3-3: Introduce numerical interval C

[0080]

[0081] Wherein, L represents the number of battery cells in the battery pack, which is L; I = 0, 1, 2, ..., L-1; C I+1 Indicates the C interval corresponding to the I+1th battery cell;

[0082] Step 3-4: Based on the sets D0 and D1 established in step 3-2 and the numerical interval C established in step 3-3, calculate the number of data points falling into each numerical interval in turn, and calculate the probability of a data point falling into the corresponding data interval. The calculation formula is as follows:

[0083]

[0084] Where P I+1 is the corresponding probability; K I+1 is the number of data points falling into the corresponding numerical interval; t represents the number of data points;

[0085] Step 3-5: Summarize the calculation results in step 3-4 to obtain the CPD calculation results of the resistance data sets R0 and R1, namely:

[0086]

[0087] In step 4, the calculation result in step 3 is compared with a set threshold value. When the calculation result is lower than the set threshold value, it indicates that a fault exists within the voltage range.

[0088] Embodiment 2:

[0089] In this implementation, four new lithium-ion batteries are selected and connected in series to form a battery pack. The failure is simulated by replacing the batteries. The specific experimental steps are as follows:

[0090] Real-time EIS detection of four batteries;

[0091] The voltage and current obtained by real-time monitoring are converted into impedance spectrum by using Morlet wavelet to obtain the time-frequency conversion of voltage and current and fit the impedance spectrum. The fitting circuit diagram is shown in Figure 2 As shown, the main circuit resistance data set R0 and the first RC unit resistance data set R1 are obtained; wherein the corresponding data are obtained as follows:

[0092]

[0093] The main circuit resistance data set R0 and the first RC unit resistance data set R1 collected from the four batteries are put into cumulative probability distribution calculation, wherein the data sets D0 and D1 of the largest main circuit resistance data set R0 and first RC unit resistance data set R1 sequence put into calculation are as follows:

[0094]

[0095] The numerical interval C introduced for D0 is as follows:

[0096]

[0097] The numerical interval C introduced for D1 is as follows:

[0098]

[0099] The above interval C and resistance data set R are used to obtain the fault location results. The results are as follows: Figure 3 As shown above, it can be seen that at 780s, the main circuit resistance R0′ and the first RC unit resistance data set R1′ of the No. 3 battery in the measured battery pack increased sharply, indicating that the battery pack failed, and it was further determined that an open circuit failure occurred in the battery pack. This example proves that this method has a strong sensitivity to internal battery pack failures and can accurately identify the time location and type of failures.

[0100] Embodiment 3:

[0101] This embodiment proposes a system for quickly locating battery faults in a substation power system based on impedance characteristics, and the system includes:

[0102] Electrochemical impedance spectroscopy (EIS) test module, used to perform real-time electrochemical impedance spectroscopy (EIS) testing on battery packs;

[0103] The error correction model ECM fitting module is used to fit the test results obtained by the electrochemical impedance spectroscopy test module to obtain the corresponding resistance data set;

[0104] The cumulative probability distribution operation module is used to operate the resistance data set obtained by the error correction model ECM fitting module;

[0105] A battery fault detection module is used to make judgments based on the results obtained by the cumulative probability distribution calculation module and detect battery faults;

[0106] The battery fault location module is used to determine the fault type and location.

[0107] Furthermore, the electrochemical impedance spectroscopy (EIS) test module selects a periodic square wave as a standard excitation current for impedance measurement, and uses Morlet wavelet to complete the time-frequency conversion of voltage and current to obtain an impedance spectrum.

[0108] Furthermore, the error correction model ECM fitting module fits the EIS test results of the electrochemical impedance spectroscopy EIS test module into a second-order equivalent model through ECM to form a resistance data set; and selects data with obvious changes in the resistance data set as data features, and uses the selected data features to determine the fault location of the battery pack.

[0109] The above describes in detail the preferred specific embodiments of the present invention. The method has the advantages of simplicity, low cost, high efficiency, high precision and online operation, reduces the consumption of manpower and material resources, lays the foundation for research such as determining the location of battery faults in substations, is of great significance for determining battery faults, and has certain engineering practice value. It should be understood that ordinary technicians in this field can make many modifications and changes based on the concept of the present invention without creative labor. Therefore, all technical solutions that can be obtained by technicians in this technical field through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for quickly locating battery faults in a substation power system based on impedance characteristics, characterized in that: The method comprises the following steps: Step 1, performing a real-time electrochemical impedance spectroscopy (EIS) test on the battery pack; Step 2, performing ECM error correction model fitting on the EIS test results in step 1 to obtain a corresponding resistance data set; Step 3, the resistance data set obtained in step 2 is put into cumulative probability distribution operation to obtain the cost performance index CPD calculation result; Step 4, judging whether a fault occurs according to the calculation result in step 3, and stopping the battery operation if a fault occurs; Step 5, mapping the time when the fault occurs in step 4 to the experimental time length in step 2, so that the type and location of the fault can be determined.

2. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 1 is characterized in that: In step 1, a periodic square wave is selected as a standard excitation current for impedance measurement, and Morlet wavelet is used to complete the time-frequency conversion of voltage and current to obtain an impedance spectrum.

3. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 1 is characterized in that: In step 2, the real-time EIS test results of step 1 are fitted into a second-order equivalent model through ECM to form a resistance data set; and data with obvious changes are selected from the resistance data set as data features, and the fault location of the battery pack is determined by the selected data features.

4. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 1 is characterized in that: The resistance data set includes a main circuit resistance data set R0 and a first RC unit resistance data set R1; The main circuit resistance data set R0 is expressed as: The first RC unit resistance data set R1 is represented by: Wherein, m means that the battery pack has m battery cells; R 0(m) (t) represents the tth R0 resistance data point of the mth battery cell in the battery pack; R 1(m) (t) represents the tth R1 resistance data point of the mth battery cell.

5. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 4 is characterized in that: The step 3 inputs the resistance data set into the cumulative probability distribution algorithm CPD, including the following steps: Step 3-1: Obtain the maximum R corresponding to each battery cell through the resistance data sets R0 and R1 obtained by ECM fitting 0max and R 1max sequence; Where R 0max (t), R 1max (t) is the maximum resistance of the battery cell in the battery pack at the tth data point; The maximum R obtained from each battery cell 0max and R 1max The sequence is R 0max ={R 0max (1)…R 0max (t)}(5) R 1max ={R 1max (1)…R 1max (t)}(6) Step 3-2: 0max and R 1max The data in the sequence dataset are rearranged in sets D0 and D1 in increasing order, namely: Among them, R 0[1] For R 0max The smallest data in a sequence data set; R 0[t] For R 0max The largest data set in the sequence data set; R 1[1] For R 1max The smallest data in a sequence data set; R 1[t] For R 1max The largest data set in the sequence data set; Step 3-3: Introduce numerical interval C Wherein, L represents the number of battery cells in the battery pack, which is L; I = 0, 1, 2, ..., L-1; C I+1 Indicates the C interval corresponding to the I+1th battery cell; Step 3-4: Based on the sets D0 and D1 established in step 3-2 and the numerical interval C established in step 3-3, calculate the number of data points falling into each numerical interval in turn, and calculate the probability of a data point falling into the corresponding data interval. The calculation formula is as follows: Where P I+1 is the corresponding probability; K I+1 is the number of data points falling into the corresponding numerical interval; t represents the number of data points; Step 3-5: Summarize the calculation results in step 3-4 to obtain the CPD calculation results of the resistance data sets R0 and R1, namely:

6. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 5 is characterized in that: In step 4, the calculation result in step 3 is compared with a set threshold value. When the calculation result is lower than the set threshold value, it indicates that a fault exists within the voltage range.

7. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 6 is characterized in that: The threshold is set by a fault value measured by a battery pack experiment.

8. The method for quickly locating battery faults in a substation power supply system based on impedance characteristics according to claim 7 is characterized in that: The threshold is 0.0075.

9. A system using the method for quickly locating battery faults in a substation power supply system based on impedance characteristics as claimed in claim 1, characterized in that: The system includes: Electrochemical impedance spectroscopy (EIS) test module, used to perform real-time electrochemical impedance spectroscopy (EIS) testing on battery packs; The error correction model ECM fitting module is used to fit the test results obtained by the electrochemical impedance spectroscopy test module to obtain the corresponding resistance data set; The cumulative probability distribution operation module is used to operate the resistance data set obtained by the error correction model ECM fitting module; A battery fault detection module is used to make judgments based on the results obtained by the cumulative probability distribution calculation module and detect battery faults; The battery fault location module is used to determine the fault type and location.

10. The fast location system for battery faults in a power supply system of a substation based on impedance characteristics according to claim 9 is characterized in that: The electrochemical impedance spectroscopy (EIS) test module selects periodic square waves as the standard excitation current for impedance measurement, and uses Morlet wavelets to complete the time-frequency conversion of voltage and current to obtain the impedance spectrum.

11. The fast location system for battery faults in a power supply system of a substation based on impedance characteristics according to claim 9 is characterized in that: The error correction model ECM fitting module fits the EIS test results of the electrochemical impedance spectroscopy (EIS) test module into a second-order equivalent model through ECM to form a resistance data set; and selects data with obvious changes in the resistance data set as data features, and uses the selected data features to determine the fault location of the battery pack.

Images