A method and device for determining a battery parameter identification result, and a storage medium

Abstract

The application discloses a battery parameter identification result determination method and device, and a storage medium. The method comprises the following steps: obtaining M groups of parameter identification results of a battery under the same working condition, each group of parameter identification results corresponding to one value of K electrochemical parameters to be identified included in the battery; taking each group of parameter identification results as a sample point, and forming a first sample set by the M groups of parameter identification results; searching for a sample point with the minimum sum of distances from all other sample points in the first sample set, and recording the sample point as a minimum distance sum point of the first sample set; obtaining a maximum density point of the first sample set according to the minimum distance sum point of the first sample set, and taking the values of the K electrochemical parameters of the maximum density point as the final identification result of the electrochemical parameters to be identified. The values of the maximum density point of the sample space are taken as the battery parameter identification result, so that the reliability and stability of the identification result can be improved.

Description

Technical Field

[0001] This invention relates to the field of battery technology, and in particular to a method and apparatus for determining battery parameter identification results, as well as a storage medium. Background Technology

[0002] Lithium-ion batteries have many electrochemical parameters, and the effective identification of these electrochemical parameters is the key to diagnosing and studying the state of lithium-ion batteries at the mesoscopic level.

[0003] Currently, parameter identification programs can achieve online identification of electrochemical parameters for a single lithium battery module at the sub-minute level. A single identification often outputs a large number of results, such as 200 rows of parameter identification results, i.e., 200 sets of 24 electrochemical parameters, corresponding to 200 points in a 24-dimensional parameter space. Theoretically, the same battery cell under the same operating condition has only one set of definite electrochemical parameters, but in practice, the multiple sets of results output often differ significantly. How to utilize a large number of identification results to determine the final electrochemical parameters uploaded to the client is crucial to the reliability and trustworthiness of the identification results. The commonly used method is the mean method, which calculates the average value of the corresponding parameters based on all rows of identification results, and then uses the average value of the entire set of parameters as the electrochemical parameter identification result for that operating condition data of a specific energy storage system.

[0004] However, the identification results for some parameters fluctuate significantly across different rows. For these parameters, the mean of the identification results from different runs of the identification program also fluctuates greatly. This means that the reliability of the identification results obtained using the mean method is unreliable. Summary of the Invention

[0005] One of the objectives of this invention is to overcome at least some of the shortcomings of the prior art and to provide a method, apparatus, and storage medium for determining battery parameter identification results.

[0006] This invention provides a method for determining battery parameter identification results, comprising:

[0007] Obtain M sets of parameter identification results for the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery.

[0008] Each set of parameter identification results is taken as a sample point, and the first sample set is formed based on all the obtained sample points.

[0009] From the first sample set, search for the sample point with the smallest sum of distances to all other sample points, and record it as the minimum distance sum point in the first sample set;

[0010] The maximum density point of the first sample set is obtained based on the minimum distance and point of the first sample set, and the final identification result of the electrochemical parameter to be identified is obtained based on the values ​​of K electrochemical parameters of the maximum density point.

[0011] The present invention also provides a method for determining battery parameter identification results, comprising:

[0012] Obtain M sets of parameter identification results for the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery.

[0013] Each set of parameter identification results is taken as a sample point, and the first sample set is formed based on all the obtained sample points.

[0014] The first sample set is divided into N sub-sample sets, and each subsequent sub-sample set contains the data of the previous sub-sample set. The last sub-sample set is the first sample set.

[0015] Obtain the minimum distance and point sum of the top L positions in each subsample set;

[0016] The minimum distance and points of the first L positions of the N sub-sample sets are used to form a second sample set;

[0017] The maximum density point of the first sample set is determined based on the second sample set, and the final identification result of the electrochemical parameter to be identified is obtained based on the values ​​of K electrochemical parameters of the maximum density point.

[0018] In some embodiments, determining the maximum density point of the first sample set based on the second sample set includes:

[0019] Count the number of repetitions in the second sample set and their frequency.

[0020] If there exists a point that is repeated N times, and the ranking of the minimum distance sum of the repeated point in the N sub-sample sets gradually moves forward, and stabilizes at the first position as the sample size increases, then the repeated point is the point with the highest density in the first sample set.

[0021] In some embodiments, including:

[0022] If there exists a point that is repeated less than N times but more than N / 2 times, and the ranking of the minimum distance sum of the repeated point in the N sub-sample sets gradually moves forward and stabilizes at the first position as the sample size increases, then the repeated point is the point with the highest density in the first sample set.

[0023] In some embodiments, the method includes: when the maximum density point of the first sample set cannot be determined based on the second sample set, obtaining more samples, updating the first sample set, obtaining a second sample set based on the updated first sample set, until the maximum density point of the first sample set is determined based on the second sample set.

[0024] In some embodiments, the first sample set is divided into N sub-sample sets with equal arithmetic progression, and the number of the N sub-sample sets can form an arithmetic progression sequence.

[0025] In some embodiments, obtaining the minimum distance and point sum of the top L positions in a subsample set includes:

[0026] Calculate the sum of distances for each sample point in the subsample set, where the sum of distances for each sample point is the sum of the distances between that sample point and all other sample points in the subsample set;

[0027] Based on the sum of distances to all sample points in the subsample set, the minimum distance sum of the L positions in the sorted subsample set is obtained.

[0028] In some embodiments, forming a first sample set based on all acquired sample points includes:

[0029] All the obtained sample points are normalized according to the upper and lower limits of the parameter identification space of each electrochemical parameter, and the first sample set is formed based on all the normalized sample points.

[0030] The method of obtaining the final identification result of the electrochemical parameter to be identified based on the values ​​of K electrochemical parameters at the maximum density point includes:

[0031] The values ​​of the K electrochemical parameters at the maximum density point are inversely normalized to the corresponding actual electrochemical values, and the actual electrochemical values ​​are used as the final identification results of the electrochemical parameters to be identified.

[0032] The present invention also provides a device for determining battery parameter identification results, comprising:

[0033] The sample acquisition module is used to acquire M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery.

[0034] The first sample set formation module is used to take each set of parameter identification results as a sample point and form the first sample set based on all the acquired sample points.

[0035] The search module is used to search for the sample point with the smallest sum of distances to all other sample points in the first sample set, and record it as the minimum distance sum point in the first sample set.

[0036] The identification result determination module is used to obtain the maximum density point of the first sample set based on the minimum distance and point of the first sample set, and to obtain the final identification result of the electrochemical parameter to be identified based on the values ​​of K electrochemical parameters of the maximum density point.

[0037] The present invention also provides a device for determining battery parameter identification results, comprising:

[0038] The sample acquisition module is used to acquire M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery.

[0039] The first sample set formation module is used to take each set of parameter identification results as a sample point and form the first sample set based on all the acquired sample points.

[0040] The grouping module is used to divide the first sample set into N sub-sample sets, where each sub-sample set contains the data of the previous sub-sample set, and the data of the last sub-sample set is the same as the data of the first sample set.

[0041] The search module is used to obtain the minimum distance and point of the top L positions in each subsample set; and to form a second sample set by combining the minimum distance and point of the top L positions in the N subsample sets.

[0042] The identification result determination module is used to determine the maximum density point of the first sample set based on the second sample set, and to obtain the final identification result of the electrochemical parameter to be identified based on the values ​​of K electrochemical parameters of the maximum density point.

[0043] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for determining the battery parameter identification result as described in any of the preceding claims.

[0044] Compared with the prior art, the method, apparatus, and storage medium for determining battery parameter identification results provided by the present invention have at least one of the following beneficial effects:

[0045] 1. By using the value of the maximum density point in the sample space as the battery parameter identification result, the present invention can improve the reliability and stability of the parameter identification result.

[0046] 2. This invention can find a stable maximum density point by automatically increasing the sample size, thereby improving the reliability of parameter identification results.

[0047] 3. By normalizing the electrochemical parameters in all sample points, this invention enables electrochemical parameters of different magnitudes to exert an equal influence in the calculation of distances. Attached Figure Description

[0048] The preferred embodiments will now be described in a clear and easy-to-understand manner, in conjunction with the accompanying drawings, to further explain the above-mentioned characteristics, technical features, advantages, and implementation methods of the present invention.

[0049] Figure 1 This is a flowchart of an embodiment of the method for determining battery parameter identification results of the present invention;

[0050] Figure 2 This is a flowchart of another embodiment of the method for determining battery parameter identification results of the present invention;

[0051] Figure 3 This is a schematic diagram of the structure of an embodiment of the battery parameter identification result determination device of the present invention;

[0052] Figure 4 This is a schematic diagram of another embodiment of the battery parameter identification result determination device of the present invention;

[0053] Figure 5 This is a schematic diagram of another embodiment of the battery parameter identification result determination device of the present invention. Detailed Implementation

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.

[0055] To keep the drawings concise, each figure only schematically shows the parts relevant to the invention, and these do not represent the actual structure of the product. Furthermore, to facilitate understanding, in some figures, only one of components with the same structure or function is schematically depicted, or only one is labeled. In this document, "one" not only means "only one," but can also mean "more than one."

[0056] One embodiment of the present invention, such as Figure 1 As shown, a method for determining battery parameter identification results includes:

[0057] Step S100: Obtain the M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of one of the K electrochemical parameters to be identified in the battery.

[0058] Step S110 takes each set of parameter identification results as a sample point and forms the first sample set based on all the obtained sample points.

[0059] Specifically, taking lithium-ion batteries as an example, their electrochemical models include numerous electrochemical parameters. Some of these parameters can be directly measured, while others need to be identified through model parameter identification. Model parameter identification refers to determining unknown parameters in the electrochemical model based on readily available external performance data of the battery. The electrochemical parameters that need to be identified vary depending on the model or application scenario. Here, it is assumed that the battery requires identification of K electrochemical parameters, where K>=1.

[0060] The battery referred to in this embodiment is not limited to lithium-ion batteries, but can also be other types of batteries such as sodium-ion batteries.

[0061] Online identification of electrochemical parameters can be achieved by running a parameter identification program, often outputting a large number of identification results in a single run. Each set of identification results is equivalent to a sample point in a sample space. To determine reliable parameter identification results, a relatively large sample size is generally required; M is usually quite large. The required sample size for parameter identification results may vary depending on the operating conditions. Setting M to be larger than a preset data size can be done based on the experience of professionals.

[0062] A sufficient number of samples can be obtained through multiple identifications. Once a sufficient number of samples are obtained, the final identification result can be determined based on the obtained samples.

[0063] Taking the daily data identification of a certain field station as an example: the cluster current, cluster SOC, cluster voltage and other operating conditions are good. The operating condition data is recorded every 10 seconds. The operating condition data of about one day (a total of 8442 operating condition data) is selected. The parameter identification program runs for about 40-50 minutes and can obtain about 100,000 lines of identification results. These can form a sufficiently rich sample size.

[0064] Step S120: From the first sample set, search for the sample point with the smallest sum of distances to all other sample points, and record it as the minimum distance sum point of the first sample set;

[0065] Step S130: The maximum density point of the first sample set is obtained based on the minimum distance and point of the first sample set. The final identification result of the electrochemical parameter to be identified is obtained based on the values ​​of the K electrochemical parameters of the maximum density point.

[0066] Specifically, each set of parameter identification results is treated as a point in the sample set, and each point has K-dimensional data. In the K-dimensional parameter space, a large number of K-dimensional coordinate points are distributed, and the density of these points in space has a certain distribution. A particular point necessarily corresponds to the peak value of the density distribution function; this point is the point of maximum density. This means that the most probable coordinate value in this parameter space is, in other words, the most likely value of the parameter identification result for this working condition.

[0067] How to find this point? One possible approach is to find a point in the entire parameter space that satisfies the condition that the sum of the distances to all other points is the minimum among all the sums of distances. This indicates that the point is the centroid of the point distribution space, which is also the point of maximum density.

[0068] Specifically, in the first sample set containing M sample points in a K-dimensional data space, the sum of the distances between any point and all the remaining sample points is calculated. Then, based on the sum of the distances of all sample points, the sample point with the smallest sum of distances is found. This point with the smallest sum of distances is taken as the maximum density point in the first sample set. The K-dimensional coordinates of the maximum density point are used as the final identification result of the battery parameters.

[0069] Because the sample size is sufficient, a stable and reliable maximum density point can be obtained. Using the value of the maximum density point in the sample space as the battery parameter identification result can improve the reliability and stability of the identification result.

[0070] In one embodiment, step S110, forming a first sample set based on all acquired sample points, includes: normalizing all acquired sample points according to the upper and lower limits of the parameter identification space of each electrochemical parameter, and forming a first sample set based on all normalized sample points.

[0071] Step S130 obtains the final identification result of the electrochemical parameter to be identified based on the values ​​of the K electrochemical parameters at the maximum density point, including: inversely normalizing the values ​​of the K electrochemical parameters at the maximum density point to the corresponding actual electrochemical values, and using the obtained actual electrochemical values ​​as the final identification result of the electrochemical parameter to be identified.

[0072] Specifically, there are many electrochemical parameters to identify, and these parameters may have different value ranges. For example, some parameters may have a minimum value as low as 1E-8, while others may have a maximum value as high as 1E4. Without normalization, when calculating the distance between points composed of parameters of different magnitudes, parameters of lower magnitudes will be overwhelmed by parameters of higher magnitudes. To ensure that electrochemical parameters of different magnitudes or with significant differences in magnitude have an equal impact in the distance calculation process, it is necessary to normalize the values ​​of each electrochemical parameter in the sample points, unifying the values ​​of each electrochemical parameter to the same range. For example, the actual value range of parameter 1 is A1 to B1, and the actual value range of parameter 2 is A2 to B2. After normalization, they are both unified to decimals between 0 and 1. The value of parameter 1 can be converted to between 0 and 1 by calculating (parameter 1 - A1) / (B1 - A1), and the value of parameter 2 can be converted to between 0 and 1 by calculating (parameter 2 - A2) / (B2 - A2).

[0073] When dealing with large sample sizes, in order to shorten the computation time and facilitate computer calculations, the normalization range can be set to a large integer, such as between 0 and 255, or between 0 and 1023, so that the normalized electrochemical parameters take integer values.

[0074] If the values ​​of the samples in the first sample set are normalized, then when determining the final identification result of the battery parameters based on the maximum density point in the first sample set, it is necessary to first perform an equivalent inverse normalization on the values ​​of the maximum density point, and then use the corresponding actual parameter values ​​as the final identification result of the battery parameters.

[0075] If the ranges of values ​​for each electrochemical parameter to be identified are not significantly different and are within an acceptable range, normalization can be omitted, and the actual values ​​of the electrochemical parameters can be used directly.

[0076] Another embodiment of the present invention, such as Figure 2 As shown, a method for determining battery parameter identification results includes:

[0077] Step S200: Obtain the M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of one of the K electrochemical parameters to be identified in the battery.

[0078] Step S210 takes each set of parameter identification results as a sample point and forms the first sample set based on all the obtained sample points.

[0079] The above steps can be referred to the description in the previous embodiment.

[0080] Step S220 divides the first sample set into N subsets, with each subset containing the data of the previous subset, and the data of the last subset being the same as the data of the first sample set.

[0081] Step S230: Obtain the minimum distance and points of the top L positions in each subsample set;

[0082] Step S240 forms a second sample set by the minimum distance and points of the first L positions of the N sub-sample sets.

[0083] Specifically, for example, the first sample set contains 100,000 data points, which can be divided into 10 sub-sample sets with 10,000, 20,000, 30,000, ..., 100,000 data points. Each subsequent sub-sample set always contains the data from the previous sub-sample set, and the last sub-sample set is the same as the first sample set.

[0084] Other grouping methods can also be used, such as 8,000, 20,000, 40,000, 50,000, 55,000, ..., 100,000, etc.; the difference in the amount of data between two adjacent subsample sets can be equal, and the sample size can be expanded in an arithmetic progression or in a variable progression, such as a large difference in the amount of data in the first part and a smaller difference in the second part.

[0085] To quickly find a stable maximum density point, it is preferable to group the data into sets with a relatively large initial sample size (i.e., a large sample size in the first subset) and a relatively large difference in the data size between adjacent subsets.

[0086] The following method can be used to obtain the minimum distance and sum of the top L positions in each subsample set (L>=1):

[0087] Step S231 calculates the sum of distances for each sample point in the subsample set, where the sum of distances for each sample point is the sum of the distances between that sample point and all other sample points in the subsample set.

[0088] Step S232: Based on the sum of distances among all sample points in the subsample set, obtain the minimum distance sum of the L positions in the sorted subsample set.

[0089] For step S232, we can first find the minimum distance and point in the subsample set, remove this point, then find the minimum distance and point in the remaining points, remove this point again, and repeat this process until we find the minimum distance and point in the first L positions of the sort.

[0090] Alternatively, the sum of distances between all sample points in the subsample set can be sorted. Bubble sort can be used to sort in ascending order, and then the first L points in the sorted set can be selected. Or, the sorting can be done in descending order, and then the last L points in the sorted set can be selected.

[0091] The minimum distances and points of each subset are used to form a second subset. For example, the subset is divided into 10 groups (N=10), and the first 10 minimum distances and points of each group are selected (L=10). The minimum distances and points of each group are then used to form a second subset according to the group order. For example, of these 100 points, points 0-9 are in the first group, points 10-19 are in the second group, and so on. This subset records the K-dimensional coordinates of each minimum distance and point in each group.

[0092] Step S250: Determine the maximum density point of the first sample set based on the second sample set;

[0093] If the maximum density point of the first sample set can be determined based on the second sample set in step S260, then the final identification result of the electrochemical parameter to be identified can be obtained based on the values ​​of the K electrochemical parameters of the maximum density point.

[0094] If the maximum density point of the first sample set cannot be determined based on the second sample set in step S270, then obtain more samples and jump to step S210 to update the first sample set. Obtain the second sample set based on the updated first sample set, and try to obtain the maximum density point of the first sample set based on the second sample set. Repeat the above process until the maximum density point of the first sample set is determined based on the second sample set.

[0095] Specifically, the larger the sample size, the longer the program execution time. How can a reasonable sample size be determined within an acceptable time cost? To address this, this embodiment first attempts to find the maximum density point of the first sample set with a preset sample size (initial value M). If the maximum density point cannot be determined based on the preset sample size, the sample size is increased by a certain increment, the first sample set is updated, and the search for the maximum density point is repeated until the maximum density point is found.

[0096] This embodiment can automatically increase the sample size to find a stable maximum density point.

[0097] In one embodiment, determining the maximum density point of the first sample set based on the second sample set in step S250 includes:

[0098] Step S251: Count the duplicate points and their frequency in the second sample set;

[0099] Step S252 If there exists a duplicate point that is repeated N times, and the ranking of the minimum distance sum of the duplicate point in the N sub-sample sets sorted in ascending order of sample size gradually moves forward, and stabilizes at the 1st position as the sample size increases, then the duplicate point is the maximum density point of the first sample set.

[0100] Step S253: If there exists a point that is repeated less than N times but more than N / 2 times, and the point gradually moves up in the order of the minimum distance sum among the N sub-sample sets sorted by sample size in ascending order, and stabilizes at the first position as the sample size increases, then the point is the point with the highest density in the first sample set.

[0101] Specifically, the second sample set consists of the minimum distance sums of the top L points in each of the N subsample sets, and the number of minimum distance sums in the second sample set is A = N*L. Statistical analysis is performed on the duplicate points and their frequency of repetition among the A minimum distance sums. The A points are then sorted in descending order according to their frequency of repetition. Points with a frequency of repetition greater than N / 2 are examined, and their ranking within each group is compared.

[0102] Ideally, there exists a point that is repeated N times, and its ranking in the minimum distance sum of N subsets sorted by sample size gradually increases, stabilizing at number 1 as the sample size increases. This means the sample size is sufficiently rich that the K-dimensional electrochemical space corresponding to this battery condition is statistically "filled," ensuring that further increasing the sample size will not change the minimum distance sum point; in other words, a stable clustering of minimum distance sum point clusters has been achieved. This point is the maximum density point we are looking for. The phrase "gradually increasing ranking" means that starting from a certain subset, the point's ranking gradually improves in subsequent subsets.

[0103] In the suboptimal case, there exists a point whose repetition count is slightly less than N and greater than N / 2. There may be multiple such points. In this case, we need to examine the ranking of each repetition point in each group. If a point's ranking gradually moves up and stabilizes at number 1 as the sample size increases, this "slightly" means that the sample size is rich enough to statistically "fill" the K-dimensional electrochemical space corresponding to the battery condition. This means that further increasing the sample size will not change the minimum distance and the point, which is to say, the stable clustering of the minimum distance and the point cluster has been achieved.

[0104] To verify this finding, we can return to step S200 and repeat the subsequent steps to obtain a more sufficient sample size and obtain a second sample set. If the distance sum of the point ranks first among the minimum distance sum point sets in each group with more than N / 2 occurrences, then the point is determined as the maximum density point to be searched.

[0105] Taking a sample set of 100,000 points as an example, it is divided into 10 sub-sample sets (i.e., N=10), with the number of samples in each group being 10,000, 20,000, 30,000, ..., 100,000 points. The next sub-sample set always contains the data of the previous sub-sample set.

[0106] Obtain the minimum distance and point sum of the top 10 items in each subsample set (i.e., L = 10).

[0107] A second sample set of 100 points was obtained. Statistical analysis of the second sample set yielded the following results:

[0108] Of these 100 points, 18 points are independent, and the remaining 82 points are repeated. The number of repetitions are as follows:

[0109] There is one point that is repeated 10 times: the point is sorted in each group as 0, 1, 4, 1, 0, 0, 0, 0, 0, 0; each group is sorted in ascending order by the minimum distance sum, where 0 represents the first point, 1 represents the second point, and so on.

[0110] There is 1 point that is repeated 6 times; there are 6 points that are repeated 5 times; there are 3 points that are repeated 4 times; there are 4 points that are repeated 3 times; and there are 6 points that are repeated 2 times.

[0111] The point that is repeated 10 times is ranked first starting from the 5th group. It satisfies the condition that its ranking in each group gradually moves forward and gradually stabilizes at the 1st position. Therefore, the coordinates of the point that is repeated 10 times are the maximum density point corresponding to the working condition data.

[0112] In one embodiment, the first sample set is divided into N sub-sample sets with equal arithmetic progression, and the number of N sub-sample sets can form an arithmetic progression sequence.

[0113] This arithmetic progression grouping method is easy to implement through program control, and the increase in sample points is relatively uniform.

[0114] In one embodiment, step S210, forming a first sample set based on all acquired sample points, includes: normalizing all acquired sample points according to the upper and lower limits of the parameter identification space of each electrochemical parameter, and forming a first sample set based on all normalized sample points.

[0115] In step S250, the final identification result of the electrochemical parameter to be identified is obtained based on the values ​​of the K electrochemical parameters at the maximum density point. This includes: inversely normalizing the values ​​of the K electrochemical parameters at the maximum density point to the corresponding actual electrochemical values, and using the actual electrochemical values ​​as the final identification result of the electrochemical parameter to be identified.

[0116] Specifically, there are many electrochemical parameters to be identified, and these parameters may have different value spaces. If normalization is not performed, when calculating the distance between points formed by parameters of different magnitudes, parameters of lower magnitudes will be overwhelmed by parameters of higher magnitudes. To ensure that electrochemical parameters of different magnitudes or with significant differences in magnitude have an equal impact in the distance calculation process, it is necessary to normalize the values ​​of each electrochemical parameter in the sample points, unifying the values ​​of each electrochemical parameter to the same range.

[0117] If the values ​​of the samples in the first sample set are normalized, then when determining the final identification result of the battery parameters based on the maximum density point in the first sample set, it is necessary to first perform inverse normalization on the values ​​of the maximum density point, and then use the corresponding actual parameter values ​​as the final identification result of the battery parameters.

[0118] If the ranges of values ​​for each electrochemical parameter to be identified are not significantly different and are within an acceptable range, normalization can be omitted, and the actual values ​​of the electrochemical parameters can be used directly.

[0119] One embodiment of the present invention, such as Figure 3 As shown, a device for determining battery parameter identification results includes:

[0120] The sample acquisition module 100 is used to acquire the M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery.

[0121] The first sample set formation module 200 is used to take each group of parameter identification results as a sample point and form a first sample set based on all the acquired sample points.

[0122] The search module 300 is used to search for the sample point with the smallest sum of distances to all other sample points in the first sample set, and record it as the minimum distance sum point in the first sample set.

[0123] The identification result determination module 400 is used to obtain the maximum density point of the first sample set based on the minimum distance and point of the first sample set, and to obtain the final identification result of the electrochemical parameter to be identified based on the value of K electrochemical parameters of the maximum density point.

[0124] In this embodiment, when the sample size is sufficient, a stable and reliable maximum density point is obtained, and the value of the maximum density point in the sample space is used as the battery parameter identification result, thereby improving the reliability and stability of the identification result.

[0125] In one embodiment, the first sample set forming module 200 is further configured to normalize all acquired sample points according to the upper and lower limits of the parameter identification space of each electrochemical parameter, and form a first sample set based on all normalized sample points. The identification result determining module 400 is further configured to inversely normalize the values ​​of the K electrochemical parameters at the maximum density point to the corresponding actual electrochemical values, and use the obtained actual electrochemical values ​​as the final identification result of the electrochemical parameter to be identified.

[0126] Another embodiment of the present invention, such as Figure 4 As shown, a device for determining battery parameter identification results includes:

[0127] The sample acquisition module 100 is used to acquire the M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of one of the K electrochemical parameters to be identified in the battery.

[0128] The first sample set formation module 200 is used to take each set of parameter identification results as a sample point and form a first sample set based on all the acquired sample points.

[0129] Grouping module 210 is used to divide the first sample set into N sub-sample sets, where each subsequent sub-sample set contains the data of the previous sub-sample set, and the data of the last sub-sample set is the same as the data of the first sample set.

[0130] The search module 310 is used to obtain the minimum distance and point of the top L positions in each subsample set; and to form a second sample set by combining the minimum distance and point of the top L positions in the N subsample sets.

[0131] The identification result determination module 410 is used to determine the maximum density point of the first sample set based on the second sample set, and to obtain the final identification result of the electrochemical parameter to be identified based on the values ​​of the K electrochemical parameters of the maximum density point.

[0132] The sample acquisition module 100 is also used to acquire more samples when the maximum density point of the first sample set cannot be determined based on the second sample set.

[0133] The first sample set formation module 200 is also used to update the first sample set.

[0134] This embodiment can automatically increase the sample size to find a stable maximum density point.

[0135] In one embodiment, the module 410 for determining the identification result further includes:

[0136] Statistical unit, used to count the number of repetitions and their frequency in the second sample set;

[0137] The analysis unit is used to determine the maximum density point of the first sample set if a point is repeated N times, and the ranking of the minimum distance sum of the repeated point in the N sub-sample sets gradually increases and stabilizes at the first position as the sample size increases; if a point is repeated less than N times but more than N / 2 times, and the ranking of the minimum distance sum of the repeated point in the N sub-sample sets gradually increases and stabilizes at the first position as the sample size increases, then the repeated point is the maximum density point of the first sample set.

[0138] In one embodiment, the grouping module is further configured to divide the first sample set into N sub-sample sets with equal arithmetic progression, the number of which constitutes an arithmetic progression sequence.

[0139] In one embodiment, the first sample set forming module 200 is further configured to normalize all acquired sample points according to the upper and lower limits of the parameter identification space of each electrochemical parameter, and form a first sample set based on all normalized sample points. The identification result determining module 410 is further configured to inversely normalize the values ​​of the K electrochemical parameters at the maximum density point to the corresponding actual electrochemical values, and use the actual electrochemical values ​​as the final identification result of the electrochemical parameter to be identified.

[0140] It should be noted that the embodiments of the battery parameter identification result determination device provided by the present invention and the embodiments of the battery parameter identification result determination method provided above are all based on the same inventive concept and can achieve the same technical effect. Therefore, other specific details of the embodiments of the battery parameter identification result determination device can be referred to the description of the embodiments of the aforementioned computing device.

[0141] In another embodiment of the invention, such as Figure 5 As shown, a device for determining battery parameter identification results includes a memory 21 and a processor 22.

[0142] Memory 21 is used to store computer program 23;

[0143] The processor 22 is used to implement, when running the computer program 23, a method for determining the battery parameter identification result as described above.

[0144] The memory 10 can be any internal storage unit and / or external storage device capable of storing data and programs. For example, the memory 10 can be a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD card), or a flash memory card.

[0145] Depending on the requirements, the processor 10 may be a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a general-purpose processor, or other logic devices.

[0146] In one embodiment of the present invention, a computer-readable storage medium stores a computer program thereon. When executed by a processor, the computer program can implement the method for determining battery parameter identification results as described in the foregoing embodiments. That is, when part or all of the technical solutions contributing to the prior art in the foregoing embodiments of the present invention are embodied in the form of a computer software product, the aforementioned computer software product is stored in a computer-readable storage medium. The computer-readable storage medium can be any entity device or apparatus capable of carrying computer program code. For example, the computer-readable storage medium can be a USB flash drive, a portable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, etc.

[0147] It should be noted that the above embodiments can be freely combined as needed. The above are merely preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining battery parameter identification results, characterized in that, include: Obtain M sets of parameter identification results for the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery. Each set of parameter identification results is taken as a sample point, and the first sample set is formed based on all the obtained sample points. The first sample set is divided into N sub-sample sets, and each subsequent sub-sample set contains the data of the previous sub-sample set. The last sub-sample set is the first sample set. Obtain the minimum distance and point sum of the top L positions in each subsample set; The minimum distance and points of the first L positions of the N sub-sample sets are used to form a second sample set; The maximum density point of the first sample set is determined based on the second sample set, and the final identification result of the electrochemical parameter to be identified is obtained based on the values ​​of K electrochemical parameters of the maximum density point.

2. The method for determining the battery parameter identification result according to claim 1, characterized in that, Determining the maximum density point of the first sample set based on the second sample set includes: Count the number of repetitions in the second sample set and their frequency. If there exists a point that is repeated N times, and the ranking of the minimum distance sum of the repeated point in the N sub-sample sets gradually moves forward, and stabilizes at the first position as the sample size increases, then the repeated point is the point with the highest density in the first sample set.

3. The method for determining the battery parameter identification result according to claim 2, characterized in that, include: If there exists a point that is repeated less than N times but more than N / 2 times, and the ranking of the minimum distance sum of the repeated point in the N sub-sample sets gradually moves forward and stabilizes at the first position as the sample size increases, then the repeated point is the point with the highest density in the first sample set.

4. The method for determining the battery parameter identification result according to claim 1, characterized in that, include: When the maximum density point of the first sample set cannot be determined based on the second sample set, more samples are obtained, the first sample set is updated, and the second sample set is obtained based on the updated first sample set, until the maximum density point of the first sample set is determined based on the second sample set.

5. The method for determining the battery parameter identification result according to claim 1, characterized in that, The first sample set is divided into N sub-sample sets with equal arithmetic progression, and the number of samples in each of the N sub-sample sets forms an arithmetic progression sequence.

6. The method for determining the battery parameter identification result according to claim 1, characterized in that, Obtaining the minimum distance and sum of the top L positions in a subsample set includes: Calculate the sum of distances for each sample point in the subsample set, where the sum of distances for each sample point is the sum of distances between that sample point and all other sample points in the subsample set. Based on the sum of distances to all sample points in the subsample set, the minimum distance sum of the L positions in the sorted subsample set is obtained.

7. The method for determining the battery parameter identification result according to any one of claims 1-6, characterized in that, The step of forming the first sample set based on all acquired sample points includes: All the obtained sample points are normalized according to the upper and lower limits of the parameter identification space of each electrochemical parameter, and the first sample set is formed based on all the normalized sample points. The method of obtaining the final identification result of the electrochemical parameter to be identified based on the values ​​of K electrochemical parameters at the maximum density point includes: The values ​​of the K electrochemical parameters at the maximum density point are inversely normalized to the corresponding actual electrochemical values, and the actual electrochemical values ​​are used as the final identification results of the electrochemical parameters to be identified.

8. A device for determining battery parameter identification results, characterized in that, include: The sample acquisition module is used to acquire M sets of parameter identification results of the battery under the same operating conditions. Each set of parameter identification results corresponds to a value of K electrochemical parameters to be identified in the battery. The first sample set formation module is used to take each set of parameter identification results as a sample point and form the first sample set based on all the acquired sample points. The grouping module is used to divide the first sample set into N sub-sample sets, where each sub-sample set contains the data of the previous sub-sample set, and the data of the last sub-sample set is the same as the data of the first sample set. The search module is used to obtain the minimum distance and point of the top L positions in each subsample set; and to form a second sample set by combining the minimum distance and point of the top L positions in the N subsample sets. The identification result determination module is used to determine the maximum density point of the first sample set based on the second sample set, and to obtain the final identification result of the electrochemical parameter to be identified based on the values ​​of K electrochemical parameters of the maximum density point.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the method for determining the battery parameter identification result as described in any one of claims 1 to 7.

Images