A method for solving SOC denominator array in data loss mode

Abstract

This invention discloses a method for solving the denominator array of SOC under a data loss scenario, comprising: obtaining the initial denominator array VD by calculating the original capacity change corresponding to each SOC change position in a data loss scenario. 0 For VD 0 Perform clustering, and label the clustering results as a. L , will a L Assign to VD L ; and utilize VD L With VD 0 The difference between the values ​​is used to correct the capacity value. Based on the corrected capacity value, the virtual value corresponding to the one-dimensional array distribution state is calculated, and the mean square error J between it and the real value is calculated. When the mean square error J is greater than the threshold m1, the position where the difference is greater than the threshold m2 is retrieved and marked to the last sequence number range VD. L The elements are re-clustered to obtain a L+1 and a L+1 Assign to VD L+1 The process iterates repeatedly, adjusting the SOC denominator array until L is less than the threshold m1, at which point the process ends and outputs a one-dimensional SOC denominator array VD. G This invention solves the problem of calculating the denominator of SOC in scenarios with lost data, avoiding the problem of large errors in ampere-hour integral capacity leading to battery diagnostic failure.

Description

Technical Field

[0001] This invention relates to the field of battery application technology, and more specifically to a method for solving the denominator array of SOC under a data loss mode. Background Technology

[0002] The battery system of new energy vehicles has a function to display the State of Charge (SOC). This value represents the percentage of the battery system's remaining usable capacity relative to its full capacity. It is calculated by the mainboard of the Battery Management System (BMS) using current values ​​collected from the battery system's internal CAN data. The calculation mainly relies on ampere-hour integration and voltage or open-circuit voltage correction. During the entire discharge process of the battery, especially in lithium iron phosphate battery systems where the battery voltage plateau is too flat to allow for voltage or open-circuit voltage correction, the SOC typically decreases at a fixed rate through ampere-hour integration. That is, the capacity released for each unit decrease in SOC is usually a constant, and this capacity value is marked as the denominator of the SOC calculation.

[0003] The SOC denominator is primarily determined by the battery system capacity or the health of individual battery cells. In most vehicles, the SOC denominator remains constant throughout the discharge process. This principle allows for capacity correction when battery data is lost or for calculating battery health. However, in certain special scenarios, such as when the battery system has severe imbalances or contains cells with abnormal self-discharge, the voltage drops to the SOC threshold after the vehicle's battery depth reaches a certain level, triggering the BMS's end-voltage correction. In this case, the SOC denominator changes significantly, using a smaller denominator value to guide a rapid decrease in SOC. In this scenario, the SOC denominator during discharge is an array composed of several constants, each used for a varying duration. Because the SOC denominator is crucial for capacity correction when data is lost and for calculating battery health, accurately calculating the SOC denominator array is essential for ensuring accurate battery diagnostics.

[0004] Currently, existing methods for compensating for lost data include window averaging, nearest-neighbor filling, and interpolation fitting. However, these methods are not suitable for compensating for missing current data during battery discharge because the current changes rapidly in this scenario, failing to meet the requirements of consistent window averaging, nearest-neighbor filling, and linear interpolation, thus failing to accurately correct the capacity. Therefore, we provide a method for solving the SOC denominator array under data loss conditions. Summary of the Invention

[0005] This invention provides a method for solving the denominator array of SOC under a data loss mode, aiming to solve the shortcomings of existing data loss scenario compensation methods, such as large correction errors and inability to accurately correct capacity.

[0006] The present invention adopts the following technical solution:

[0007] A method for solving the denominator array in a data loss mode for SOC includes the following steps:

[0008] Step 1: Collect vehicle current, time, and state of charge (SOC) data. i The data is used to perform ampere-hour integration on the current and time data during the vehicle's discharge process to obtain the capacity array Q corresponding to each moment. 1*N and obtain SOC i The time Tc corresponding to the location where the change occurred 1*M Where N is the length of the current array and M is the state of charge (SOC). i The number of times changes occurred;

[0009] Step 2: Calculate each SOC i The change in the original capacity corresponding to the change in position is used to obtain the initial denominator array VD. 0 ;

[0010] Step 3: Apply the VD obtained in Step 2 0 Clustering is performed, and the numerical label of the clustering result is 'a'. L , will a L Assign to VD L VD L All elements are a L ;

[0011] Step 4: Using VD L With VD 0 The difference between them is used to correct the capacity value, and the virtual value corresponding to the distribution state of the one-dimensional array is calculated based on the corrected capacity value. ;

[0012] Step 5: Calculate the virtual With reality The mean square error J between them;

[0013] Step 6: When the mean square error J is greater than the threshold m1, perform the search. The position where the difference is greater than the threshold m2 is marked as k1; for VD L The elements at positions k1 to M are re-clustered to obtain a L+1 and a L+1 Assign to VD L+1 VD L +1 The element at the corresponding position from k1 to the last position is a. L+1 Other positional elements remain consistent with VD L The same applies, where L represents the number of iterations;

[0014] Step 7: Repeat steps 4 through 7 until the mean square error J is less than the threshold m1, then output the one-dimensional array VD of the SOC denominator.G , where G is the final iteration number.

[0015] In a preferred embodiment, the initial denominator array VD in step two above... 0 The method for obtaining q is as follows: calibrate the capacity values ​​Q(Tc(k+1)) and Q(Tc(k)) corresponding to times Tc(k+1) and Tc(k), respectively. Q(Tc(k+1)) - Q(Tc(k)) is q. i (k), obtain q corresponding to all times Tc. i That is, to form the initial SOC denominator array VD 0 .

[0016] In a preferred embodiment, the clustering in step three above can be selected as the average of the cluster with the largest average value in the clustering results; or the cluster with the largest average value in the clustering results can be selected as the target array, and a second outlier filtering can be performed to obtain the maximum value after filtering as the clustering result; or the cluster with the largest average value in the clustering results can be selected as the target array, and the probability density distribution can be calculated to obtain the element with the largest probability density distribution value as the clustering result.

[0017] In a preferred embodiment, step four above utilizes VD L With VD 0 Virtual difference calculation The formula is as follows:

[0018]

[0019] Where I is the current, t is the time, i is the time sequence number, and k is the sequence number of times the state of charge (SOC) changes. i (i) SOC i (i-1) represent the SOC at times i and i-1, respectively. i Value, q i Q represents the capacity value between two changes in State of Charge (SOC), and detaQ represents the compensation value for the lost capacity between the two changes in SOC. L The updated capacity value is calculated for the Lth iteration. During the update, while keeping the SOC constant, detaQ is compensated only once.

[0020] In a preferred embodiment, the objective function of the mean square error J in step five above is as follows: .

[0021] In a preferred embodiment, the objective function of the mean square error J in step five above can also be replaced by the sum of absolute error values ​​or the average of absolute error values.

[0022] In a preferred embodiment, the threshold m2 is greater than the threshold m1, and the threshold m1 is set to 0.01-0.05, and the threshold m2 is set to 0.02-0.05.

[0023] The solution method of this invention is applicable to new energy vehicle battery systems or energy storage battery systems.

[0024] As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following advantages:

[0025] This invention is based on the SOC calculation principle of a battery management system. By leveraging the changing patterns of SOC, it achieves accurate calculation of the SOC denominator through multiple iterations, obtaining a one-dimensional array of SOC denominator values. This can be used to solve the compensation problem of ampere-hour integral capacity in scenarios with lost data. It greatly overcomes the impact of lost data on ampere-hour integral capacity values ​​when batteries have problems such as self-discharge anomalies and imbalances. This allows for accurate diagnosis and early warning of self-discharge anomalies and imbalances even in scenarios with lost data, ensuring the accuracy of battery cloud diagnosis in this special scenario, improving the detection rate of battery health and safety hazards, and ensuring vehicle safety. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the solution method of the present invention.

[0027] Figure 2 This is a schematic diagram of the SOC denominator array obtained using the solution method of this invention. Detailed Implementation

[0028] The following reference Figure 1 Specific embodiments of the present invention will be described below. Many details are described below to provide a comprehensive understanding of the invention; however, those skilled in the art can implement the invention without these details. Well-known components, methods, and processes will not be described in detail hereafter.

[0029] This invention provides a method for solving the denominator array in SOC under a data loss mode, referring to... Figure 1 This includes the following steps:

[0030] 1. Collect vehicle current, time, and state of charge (SOC) data. i The data is used to perform ampere-hour integration on the current and time data during the vehicle's discharge process to obtain the capacity array Q corresponding to each moment. 1*N and obtain SOC i The time Tc corresponding to the location where the change occurred 1*M Where N is the length of the current array and M is the state of charge (SOC). i The number of times changes occurred.

[0031] 2. Calibrate the capacity values ​​Q(Tc(k+1)) and Q(Tc(k)) corresponding to times Tc(k+1) and Tc(k), respectively. Subtract the capacity value Q(Tc(k)) at time Tc(k+1) from the capacity value Q(Tc(k)) at time Tc(k+1) to obtain q. i (k), obtain q corresponding to all times Tc. i That is, to form the initial SOC denominator array VD 0 .

[0032] III. The VD obtained in step two 0 Clustering is performed, and the numerical label of the clustering result is 'a'. L , will a L Assign to VD L VD L All elements are a L .

[0033] IV. Using VD L With VD 0 The difference between the values ​​is used to correct the capacity value, and the virtual value corresponding to the distribution state of the one-dimensional array is calculated based on the corrected capacity value.

[0034] Specifically, in step four of this embodiment, VD is used. L With VD 0 Virtual difference calculation The formula is as follows:

[0035]

[0036] Where I is the current, t is the time, i is the time sequence number, and k is the sequence number of times the state of charge (SOC) changes. i (i) SOC i (i-1) represent the SOC at times i and i-1, respectively. i Value, q i Q represents the capacity value between two changes in State of Charge (SOC), and detaQ represents the compensation value for the lost capacity between the two changes in SOC. L The updated capacity value is calculated for the Lth iteration. During the update, while keeping the SOC constant, detaQ is compensated only once.

[0037] V. Calculating Virtual With reality The mean square error J between them. Specifically, the formula used to calculate the mean square error J in this embodiment is as follows: .

[0038] VI. When the mean square error J is greater than the threshold m1, the search is performed. The position where the difference is greater than the threshold m2 is marked as k1; for VD LThe elements at positions k1 to M are re-clustered to obtain a L+1 and a L+1 Assign to VD L+1 VD L+1 The element at the corresponding position from k1 to the last position is a. L+1 Other positional elements remain consistent with VD L The same applies; L represents the number of iterations.

[0039] The threshold m2 is greater than the threshold m1. The threshold m1 can be set to 0.01-0.05 depending on the application scenario; the threshold m2 can be set to 0.02-0.05 depending on the application scenario.

[0040] 7. Repeat steps four through seven until the mean square error J is less than the threshold m1, then output the one-dimensional array VD of the SOC denominator. G , where G is the final iteration number.

[0041] Figure 2 The figure shows a specific embodiment of the SOC denominator array solved using the method of this invention. The figure includes the following: The “” sign represents the missing SOC denominator, while the blue “·” sign represents the one-dimensional array of the SOC denominator obtained after 4 iterations.

[0042] In step three above, the clustering can be selected by taking the average of the cluster with the largest average value as the clustering result; or by taking the cluster with the largest average value as the target array, performing secondary outlier filtering, and obtaining the maximum value after filtering as the clustering result; or by taking the cluster with the largest average value as the target array, performing probability density distribution calculation, and obtaining the element with the largest probability density distribution value as the clustering result.

[0043] The objective function for the mean square error J in step five above can also be replaced by the sum of the absolute values ​​of the errors or the average of the absolute values ​​of the errors.

[0044] The above are merely specific embodiments of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial modifications made to the present invention using this concept shall be considered as infringing upon the protection scope of the present invention.

Claims

1. A method for solving the denominator array in a data loss mode in SOC, characterized in that, Includes the following steps: Step 1: Collect vehicle current, time, and state of charge (SOC) data. i The data is used to perform ampere-hour integration on the current and time data during the vehicle's discharge process to obtain the capacity array Q corresponding to each moment. 1*N and obtain SOC i The time Tc corresponding to the location where the change occurred 1*M Where N is the length of the current array and M is the state of charge (SOC). i The number of times changes occurred; Step 2: Calculate each SOC i The change in the original capacity corresponding to the change in position is used to obtain the initial denominator array VD. 0 ; Step 3: Apply the VD obtained in Step 2 0 Clustering is performed, and the numerical label of the clustering result is 'a'. L , will a L Assign to VD L VD L All elements are a L ; Step 4: Using VD L With VD 0 The difference between the values ​​is used to correct the capacity value, and VD is calculated based on the corrected capacity value. L Virtual corresponding to the distributed state ; Step 5: Calculate the virtual With reality The mean square error J between them; Step 6: When the mean square error J is greater than the threshold m1, perform the search. The position where the difference is greater than the threshold m2 is marked as k1; for VD L The elements at positions k1 to M are re-clustered to obtain a L+1 and a L+1 Assign to VD L+1 VD L+1 The element at the corresponding position from k1 to the last position is a. L+1 Other positional elements remain consistent with VD L The same applies, where L represents the number of iterations; Step 7: Repeat steps 4 through 7 until the mean square error J is less than the threshold m1, then output the one-dimensional array VD of the SOC denominator. G Where G is the final iteration number; The capacity released for each unit decrease in SOC is marked as the SOC denominator; In step four, VD is used L With VD 0 Virtual difference calculation The formula is as follows: Where I is the current, t is the time, i is the time sequence number, and k is the sequence number of times the state of charge (SOC) changes. i (i) SOC i (i-1) represent the SOC at times i and i-1, respectively. i Value, q i Q represents the capacity value between two changes in State of Charge (SOC), and detaQ represents the compensation value for the lost capacity between the two changes in SOC. L The updated capacity value is calculated for the Lth iteration. During the update, while keeping the SOC constant, detaQ is compensated only once.

2. The method for solving the SOC denominator array under the data loss mode as described in claim 1, characterized in that, In step two, the initial denominator array VD 0 The method for obtaining q is as follows: calibrate the capacity values ​​Q(Tc(k+1)) and Q(Tc(k)) corresponding to times Tc(k+1) and Tc(k), respectively. Q(Tc(k+1)) - Q(Tc(k)) is q. i (k), obtain q corresponding to all times Tc. i That is, to form the initial SOC denominator array VD 0 .

3. The method for solving the SOC denominator array under the data loss mode as described in claim 1, characterized in that: In step three, the clustering selection process involves choosing the average value of the cluster with the largest average value as the clustering result; or selecting the cluster with the largest average value as the target array, performing secondary outlier filtering, and obtaining the maximum value after filtering as the clustering result; or selecting the cluster with the largest average value as the target array, performing probability density distribution calculation, and obtaining the element with the largest probability density distribution value as the clustering result.

4. The method for solving the SOC denominator array under the data loss mode as described in claim 1, characterized in that, The objective function for the mean square error J in step five is as follows: .

5. The method for solving the SOC denominator array under the data loss mode as described in claim 1, characterized in that: The objective function for the mean square error J in step five is the sum of the absolute values ​​of the errors or the average of the absolute values ​​of the errors.

6. The method for solving the SOC denominator array under the data loss mode as described in claim 1, characterized in that: The threshold m2 is greater than the threshold m1, and the threshold m1 is set to 0.01-0.05, while the threshold m2 is set to 0.02-0.

05.

7. The method for solving the SOC denominator array under the data loss mode as described in claim 1, characterized in that: This solution method can be applied to new energy vehicle battery systems or energy storage battery systems.

Images