A charging cabinet charging and discharging control method and system

Abstract

The present invention belongs to the technical field of charge and discharge control, and specifically relates to a charge and discharge control method and system for a charging cabinet, including: obtaining a detection curve and capacity of each charging cycle of a battery sample, clustering all charging cycles according to capacity and sequence number, screening out multiple reference categories according to the sequence number of the most recently charged battery to be charged, weighting the detection curve according to the number of charging cycles with the same sequence number in the reference category, calculating the uncertainty of each reference category and the membership and anti-membership of the battery to be charged to each reference category according to the distance of the weighted average curve of each reference category obtained and the distance between the battery to be charged and the weighted average curve of each reference category, and then calculating the probability that the battery to be charged belongs to each reference category, obtaining the estimated capacity value of the battery to be charged through the reference category with the highest probability, and performing charging control when the charging cabinet charges the battery to be charged. The present invention improves the accuracy of charging control.

Description

Technical Field

[0001] The present invention relates to the technical field of charge and discharge control, and more specifically, to a charge and discharge control method and system for a charging cabinet. Background Art

[0002] For batteries to be charged placed in the charging cabinet, by accurately evaluating the capacity of the batteries to be charged before charging and then implementing effective charging control on the charging cabinet, it is not only possible to protect the batteries from damage, but also to improve the efficiency and safety of battery use, thereby extending the battery life and reducing replacement costs.

[0003] In the related technology, for example, the Chinese patent application document with application publication number CN114172234A discloses a charging method, device, electronic device and storage medium for a battery swap cabinet. The charging method for the battery swap cabinet includes: obtaining battery type and battery status data; determining a battery charging strategy based on the battery type and the battery status data; and sending a charging instruction to the charger based on the battery charging strategy to control the charger to charge the battery. The charging method, device, electronic device and storage medium for the battery swap cabinet provided in the patent application document are used to solve the defects of the prior art that the charger cannot adapt to different types of batteries, has low charging efficiency, and reduces the battery life, so as to realize the charger adapting to different types of batteries, improve charging efficiency, and extend battery life.

[0004] Existing methods for evaluating and detecting the capacity of batteries to be charged have limitations in accuracy, which affects the accuracy of charging control when charging the batteries to be charged in a charging cabinet according to the capacity, and may cause overcharging of the batteries, thereby accelerating battery aging and reducing their lifespan. Summary of the invention

[0005] In order to solve the technical problem that the above-mentioned existing methods for evaluating and detecting battery capacity have accuracy limitations, which may cause overcharging of the battery, thereby accelerating battery aging and reducing its life, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a charging and discharging control method for a charging cabinet, comprising: for a battery to be charged placed in the charging cabinet, obtaining a detection curve of the most recent charge of the battery to be charged from historical records, and screening multiple reference categories from all categories according to the most recent charge sequence number; the method for obtaining all categories is: performing multiple charging cycles on a battery sample; in each charging cycle, collecting a detection curve of the battery sample, and performing an electrochemical impedance spectroscopy test on the battery sample to obtain impedance spectrum data, and calculating the capacity of the battery sample according to the impedance spectrum data; clustering all charging cycles according to the capacity and sequence number of the battery sample in each charging cycle to obtain multiple categories; for a target category in all reference categories, calculating the weight of each charging cycle according to the number of charging cycles with the same sequence number in the target category; weighting the detection curves of all charging cycles by the weights to obtain a weighted average curve of the target category; calculating the uncertainty of the target category according to the distance between the target category and the weighted average curve of other reference categories. ; Calculate the membership of the battery to be charged to the target category based on the distance between the most recently charged detection curve and the weighted average curve of the target category ; Calculate the inverse membership of the battery to be charged and the target category based on the distance between the most recently charged detection curve and the weighted average curve of other reference categories ; Calculate the probability that the battery to be charged belongs to the target category , The average value of the capacities of all charging cycles in the reference category with the highest probability is used as the estimated capacity of the battery to be charged, which is used to control the charging when the charging cabinet charges the battery to be charged.

[0007] The present invention clusters all charging cycles by capacity and serial number. The performance change of battery samples can be more deeply understood through cluster analysis. For batteries to be charged, the clustering result is an important reference for evaluating their capacity. The uncertainty of the target category and the membership and anti-membership of the battery to be charged and the target category are calculated based on the clustering result. The membership and anti-membership of the battery to be charged and the target category are two complementary indicators. When the characteristics of the battery to be tested in the most recent charge show strong similarity with the characteristics of the target category and at the same time show strong differences with the characteristics of other reference categories except the target category, it can be more confirmed that the battery to be tested has been charged recently. The category of the charging characteristics, this double comparison method helps to make more accurate classification judgments; the uncertainty of the target category obtained reflects the credibility of the target category as a prediction result, and the membership and inverse membership of the battery to be charged and the target category are gamma transformed through the uncertainty of the target category, and then the probability that the battery to be charged belongs to the target category is calculated. The obtained probability can more accurately reflect the relationship between the battery to be charged and the target category, and then accurately obtain the capacity estimation value of the battery to be charged. When the charging cabinet charges the battery to be charged, the accuracy of charging control is improved, thereby extending the battery life.

[0008] Preferably, the reference category contains a sequence number equal to Any integer number of charge cycles within The most recent charging sequence number. is the preset value.

[0009] Preferably, the detection curve is composed of detection data at each time point, and the detection data includes voltage, current and temperature.

[0010] The voltage, current and temperature of the battery are important data for understanding the changes in battery performance. Therefore, the present invention collects the voltage, current and temperature of the battery sample in each charging cycle as sample data for evaluating the capacity of the battery to be charged. At the same time, the detection curve of each charging of the battery to be charged is collected and stored as a historical record of the battery to be charged, so as to evaluate the capacity of the battery to be charged.

[0011] Preferably, the calculation of the capacity of the battery sample includes: the impedance spectrum data includes impedance values ​​at different frequencies, the impedance value includes a real part and an imaginary part, the real part is resistance, and the imaginary part is capacitance property; using fitting software to analyze the impedance spectrum data to obtain an equivalent circuit model, the equivalent circuit model includes charge transfer resistance ; Calculate the charge transfer coefficient based on the charge transfer resistance ; Estimate the battery capacity of the battery sample based on the charge transfer coefficient, the active surface area of ​​the electrode of the battery sample, the number of electron transfers of the electrode reaction of the battery sample, and the Faraday constant. The battery capacity of the battery sample is equal to the product of the charge transfer coefficient, the active surface area of ​​the electrode of the battery sample, the number of electron transfers of the electrode reaction of the battery sample, and the Faraday constant.

[0012] Preferably, the step of calculating the weight of each charging cycle according to the number of charging cycles with the same serial number in the target category comprises: The weight of the charging cycle satisfies the expression: ; In the formula, is the target category with a sequence number equal to The weight of the charging cycle, is the target category with a sequence number equal to The combined number of charging cycles, is the target category with a sequence number equal to The combined number of charging cycles, The total number of charge cycles performed when testing the battery sample. The value range is .

[0013] Since the number of charging cycles with the same serial number in the reference category is different, when finally obtaining the weighted average curve that can represent the reference category, the weight of each charging cycle is calculated according to the number of charging cycles with the same serial number in the target category, and the weighted average curve of the reference category is obtained by the weighted average curve method, which can better represent the distribution of parameters in the reference category.

[0014] Preferably, the target category has a sequence number equal to The combined number of charging cycles Satisfies the expression: ; In the formula, is the target category with a sequence number equal to The combined number of charging cycles, is the target category with a sequence number equal to The number of charge cycles, is the preset value, for An integer in the range.

[0015] Preferably, the uncertainty of the calculated target category is , including: the other reference categories refer to the reference categories other than the target category in all reference categories, calculating the mean of the DTW distances between the weighted average curve of the target category and the weighted average curves of other reference categories, and normalizing the mean as the uncertainty of the target category , and the normalized result of the mean is inversely proportional to the mean.

[0016] In the present invention, the DTW distance between the weighted average curve of the target category and the weighted average curves of other reference categories reflects the similarity between the weighted average curve of the target category and the weighted average curves of other reference categories. When the similarity is high, the boundary between the target category and the reference category is more blurred, that is, the target category and multiple reference categories overlap in the feature space. At this time, the classification model becomes uncertain, resulting in a decrease in the credibility of using the target category as a prediction result. Therefore, the uncertainty of the target category obtained can reflect the credibility of using the target category as a prediction result.

[0017] Preferably, the calculation of the membership degree between the battery to be charged and the target category , including: calculating the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of the target category, recorded as ; then the membership degree of the battery to be charged and the target category Satisfies the expression: , is the normalization function.

[0018] Preferably, the calculation of the inverse membership between the battery to be charged and the target category , including: calculating the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of other reference categories, recorded as ; then the reverse membership of the battery to be charged and the target category Satisfies the expression: ; In the formula, The detection curve of the battery to be charged in the last charge and the The DTW distance of the weighted average curve of other reference categories, is the number of all reference categories, where the other reference categories refer to the reference categories other than the target category in all reference categories, then the number of all other reference categories is equal to , is the normalization function.

[0019] In the present invention, the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of other reference categories is used to represent the similarity in features between the battery to be tested and other reference categories except the target category. When the similarity is high, the features of the battery to be tested in the most recent charge cannot be distinguished from other reference categories except the target category. Therefore, the inverse membership of the battery to be charged and the target category can be obtained.

[0020] In a second aspect, the present invention provides a charging cabinet charge and discharge control system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned charging cabinet charge and discharge control method is implemented.

[0021] By adopting the above technical solution, the above-mentioned charging cabinet charging and discharging control method is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is made according to the memory and the processor for easy use.

[0022] The beneficial effects of the present invention are:

[0023] The present invention clusters all charging cycles by capacity and serial number. The performance change of battery samples can be more deeply understood through cluster analysis. For batteries to be charged, the clustering result is an important reference for evaluating their capacity. The uncertainty of the target category and the membership and anti-membership of the battery to be charged and the target category are calculated based on the clustering result. The membership and anti-membership of the battery to be charged and the target category are two complementary indicators. When the characteristics of the battery to be tested in the most recent charge show strong similarity with the characteristics of the target category and at the same time show strong differences with the characteristics of other reference categories except the target category, it can be more confirmed that the battery to be tested has been charged recently. The category of the charging characteristics, this double comparison method helps to make more accurate classification judgments; the uncertainty of the target category obtained reflects the credibility of the target category as a prediction result, and the membership and inverse membership of the battery to be charged and the target category are gamma transformed through the uncertainty of the target category, and then the probability that the battery to be charged belongs to the target category is calculated. The obtained probability can more accurately reflect the relationship between the battery to be charged and the target category, and then accurately obtain the capacity estimation value of the battery to be charged. When the charging cabinet charges the battery to be charged, the accuracy of charging control is improved, thereby extending the battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below through the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0025] Figure 1 is a flow chart schematically illustrating a charging and discharging control method of a charging cabinet in the present invention;

[0026] Figure 2 is a flow chart schematically showing step S1 in the present invention;

[0027] Figure 3is a flow chart schematically illustrating step S5 in the present invention;

[0028] Figure 4 is a flow chart schematically showing step S6 in the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0030] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0031] The embodiment of the present invention discloses a charging and discharging control method of a charging cabinet, referring to Figure 1 , comprising steps S1 to S7:

[0032] S1. Perform multiple charging cycles on the battery sample, and calculate the capacity of the battery sample in each charging cycle according to the impedance spectrum data in each charging cycle.

[0033] The flowchart in step S1 refers to Figure 2 , including step S101 to step S102, specifically:

[0034] S101, performing multiple charging cycles on the battery sample.

[0035] It should be noted that when testing battery performance, performing multiple charge and discharge cycles is a common method for evaluating key parameters such as battery capacity, life, and internal resistance.

[0036] Specifically, multiple batteries are selected as battery samples from unused batteries of the same model and specification to ensure the consistency of the test results; the battery samples are charged in cycles, wherein the charging cycle is stopped when the battery samples are fully charged; after the charging cycle, the batteries are allowed to stand at room temperature for a period of time to eliminate the effects of heat generated during the charging cycle.

[0037] S102, performing an electrochemical impedance spectroscopy test on the battery sample in each charging cycle, and calculating the capacity of the battery sample in each charging cycle according to the impedance spectrum data.

[0038] It should be noted that electrochemical impedance spectroscopy (EIS) is an electrochemical technology used to study battery performance. It can provide dynamic information on the internal processes of the battery, including charge transfer reactions, diffusion processes, and battery interface characteristics, and thus evaluate the battery capacity.

[0039] Specifically, after the charging cycle is completed, an electrochemical impedance spectrometer is used to perform an electrochemical impedance spectroscopy test on the battery sample to obtain impedance spectrum data, which includes impedance values ​​at different frequencies, and the impedance values ​​include a real part (resistance) and an imaginary part (capacitance property); the impedance spectrum data is analyzed using fitting software to obtain an equivalent circuit model, which includes a charge transfer resistance , Warburg impedance W and other parameters; Calculate the charge transfer coefficient based on the charge transfer resistance , and the charge transfer coefficient ; Estimate the battery capacity of the battery sample based on the charge transfer coefficient, the active surface area of ​​the electrode of the battery sample, the number of electron transfers of the electrode reaction of the battery sample, and the Faraday constant. The battery capacity of the battery sample is equal to the product of the charge transfer coefficient, the active surface area of ​​the electrode of the battery sample, the number of electron transfers of the electrode reaction of the battery sample, and the Faraday constant.

[0040] S2. Collect the detection curve of the battery sample in each charging cycle, and collect the detection curve of each charging of the battery to be charged.

[0041] It should be noted that electrochemical impedance spectroscopy testing (EIS) usually takes a relatively long time to complete a full frequency scan, and the interpretation of impedance spectrum data is usually more complicated and requires professional knowledge and experimental skills. Therefore, although EIS is a powerful electrochemical analysis tool that can provide detailed information about the internal charging process of the battery, it is generally not suitable for directly evaluating the capacity of the battery in the charging cabinet.

[0042] Specifically, the voltage, current and temperature of the battery are important data for understanding the change in battery performance; therefore, the present invention collects the voltage, current and temperature of the battery sample in each charging cycle as sample data for evaluating the capacity of the battery to be charged; for the battery to be charged placed in the charging cabinet, the detection curve of each charging of the battery to be charged is collected and stored as a historical record of the battery to be charged, so as to evaluate the capacity of the battery to be charged.

[0043] Specifically, the detection curve is composed of detection data at each time point, and the detection data includes voltage, current and temperature; wherein the current of the battery sample and the battery to be charged at each moment is measured by a built-in Hall effect sensor, the current of the battery sample and the battery to be charged at each moment is measured by a built-in voltage sensor, and the temperature of the battery sample and the battery to be charged at each moment is measured by a built-in thermocouple or thermistor or a temperature sensor.

[0044] S3. Clustering all charging cycles of the battery samples according to the capacity of the battery samples in each charging cycle and the sequence number of each charging cycle to obtain multiple categories.

[0045] Specifically, according to the capacity of the battery samples in each charging cycle and the sequence number of each charging cycle, all charging cycles of the battery samples are clustered by a clustering algorithm to obtain multiple categories; wherein, before clustering, the capacity of the battery samples in each charging cycle and the sequence number of each charging cycle are normalized to unify the dimensions, and the normalization method includes but is not limited to maximum and minimum value normalization, and the clustering algorithm includes but is not limited to the DBSCAN clustering algorithm and the OPTICS clustering algorithm, and the DBSCAN clustering algorithm and the OPTICS clustering algorithm are both well-known technologies and will not be described in detail here.

[0046] It should be noted that both the DBSCAN (Density-based spatial clustering of applications with noise) clustering algorithm and the OPTICS (Ordering points to identify the clustering structure) clustering algorithm are density-based clustering algorithms, and do not require the number of clusters to be specified in advance. They can automatically identify the number of clusters. Therefore, they are very suitable for classifying all charging cycles of battery samples.

[0047] S4. Filter out a plurality of reference categories from all categories according to the most recently charged sequence number of the battery to be charged.

[0048] It should be noted that cluster analysis can provide a deeper understanding of the performance changes of battery samples, and the obtained clustering results can serve as an important reference for evaluating the capacity of batteries to be charged.

[0049] It should be noted that the present invention hopes to estimate the capacity of the battery to be charged by comparing the detection curve of the battery to be charged at the most recent charge with the detection curves of all charging cycles in each category. Prior to this, multiple reference categories are screened out from all categories according to the sequence number of the battery to be charged at the most recent charge. By screening out reference categories close to the sequence number of the most recent charge, the number of categories that need to be compared can be reduced, thereby reducing the calculation complexity. At the same time, since the performance change of the battery is often closely related to the usage cycle, selecting the reference category related to the sequence number of the most recent charge can ensure the relevance of the data and provide a more accurate capacity estimation.

[0050] Specifically, according to the most recently charged sequence number of the battery to be charged Filter out multiple reference categories from all categories, where the reference categories contain Any integer number of charge cycles within The most recent charging sequence number. is the preset value.

[0051] The preset value The specific value of can be set according to the actual application scenario and requirements, and the value range of the preset value is [5,10]. The present invention sets the preset value to 5.

[0052] S5. Weighting the detection curves of each charging cycle according to the weight of each charging cycle in each reference category to obtain a weighted average curve of each reference category.

[0053] It should be noted that, since the number of charging cycles with the same serial number in the reference category is different, when finally obtaining a weighted average curve that can represent the reference category, the weighted average curve of the reference category is obtained by the weighted average curve method.

[0054] The flowchart of step S5 is shown in FIG. Figure 3 The process includes steps S501 to S502, which are specifically:

[0055] S501: Calculate the weight of each charging cycle in the target category.

[0056] Specifically, any reference category among all reference categories is used as the target category, and the weight of each charging cycle in the target category is calculated according to the number of charging cycles with the same serial number in the target category. The weight of the charging cycle satisfies the expression:

[0057] ;

[0058] In the formula, is the target category with a sequence number equal to The weight of the charging cycle, is the target category with a sequence number equal to The combined number of charging cycles, is the target category with a sequence number equal to The combined number of charging cycles, The total number of charge cycles performed when testing the battery sample. The value range is .

[0059] Among them, the target category has a sequence number equal to The combined number of charging cycles Satisfies the expression:

[0060] ;

[0061] In the formula, is the target category with a sequence number equal to The number of charge cycles, is the preset value, for An integer in the range.

[0062] It should be noted that when Out of range hour, If it does not exist, it will not be included in the calculation , for example, when , hour, , , as well as , , , , None exist, at this time .

[0063] S502 : Weighting the detection curves of all charging cycles in the target category by the weight of each charging cycle to obtain a weighted average curve of the target category.

[0064] Specifically, the detection data of the detection curves of all charging cycles in the target category at each time point are weighted and averaged through the weight of each charging cycle in the target category to obtain a weighted average value at each time point. The weighted average values ​​of all time points are connected to form a new curve, which is used as the weighted average curve of the target category.

[0065] It should be noted that since the number of charging cycles with the same serial number in the reference category is different, when the weighted average curve that can represent the reference category is finally obtained, the weight of each charging cycle is calculated according to the number of charging cycles with the same serial number in the target category, and the weighted average curve of the reference category is obtained by the weighted average curve method, which can better represent the distribution of parameters in the reference category.

[0066] S6. Calculate the uncertainty of each reference category and the membership and anti-membership of the battery to be charged to each reference category according to the distance between the weighted average curves of each reference category and the distance between the detection curve of the battery to be charged during the most recent charge and the weighted average curve of each reference category.

[0067] It should be noted that the membership and anti-membership of the battery to be charged and the target category are two complementary indicators. When the battery to be tested shows strong similarity in the comparison between the most recently charged features and the target category, and at the same time shows strong difference in the comparison with other reference categories except the target category, it can be more confirmed that the battery to be tested belongs to the category of the most recently charged features. This double comparison method helps to make more accurate classification judgments.

[0068] The flowchart of step S6 is shown in FIG. Figure 4 , including steps S601 to S603, specifically:

[0069] S601 , calculating the uncertainty of the target category according to the DTW distance between the weighted average curves of the target category and other reference categories.

[0070] Specifically, any one of all reference categories is used as the target category, and the other reference categories refer to the reference categories other than the target category in all reference categories. The mean of the DTW distance between the weighted average curve of the target category and the weighted average curve of other reference categories is calculated, and the normalized result of the mean is used as the uncertainty of the target category. , and the normalized result of the mean is inversely proportional to the mean. For example, the normalization function The mean is normalized to represents the natural exponential function, x represents the input, and y represents the output.

[0071] It should be noted that the smaller the DTW distance between the weighted average curve of the target category and the weighted average curves of other reference categories, the more similar the weighted average curve of the target category is to the weighted average curves of other reference categories. When the similarity is high, the boundary between the target category and the reference category is more blurred. That is to say, when the target category overlaps with multiple reference categories in the feature space, the classification model becomes uncertain, resulting in a decrease in the credibility of using the target category as a prediction result.

[0072] The DTW distance is obtained by a dynamic time warping (DTW) algorithm. The DTW algorithm is a method for calculating the similarity of two sequences. The DTW algorithm searches for a path with the minimum sum of matrix elements from the first data to the last data in the two sequences, and uses the number of data in the path as the DTW distance of the two sequences to represent the similarity of the two sequences.

[0073] It should be noted that the DTW distance between the weighted average curve of the target category and the weighted average curves of other reference categories in the present invention reflects the similarity between the weighted average curve of the target category and the weighted average curves of other reference categories. When the similarity is higher, the boundary between the target category and the reference category is more blurred, that is, the target category and multiple reference categories overlap in the feature space. At this time, the classification model becomes uncertain, resulting in a decrease in the credibility of using the target category as a prediction result. Therefore, the uncertainty of the target category obtained can reflect the credibility of using the target category as a prediction result.

[0074] S602: Calculate the membership degree of the battery to be charged and the target category according to the DTW distance between the detection curve of the battery to be charged in the most recent charging and the weighted average curve of the target category.

[0075] Specifically, the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of the target category is calculated, which is recorded as ; then the membership degree of the battery to be charged and the target category Satisfies the expression:

[0076] ;

[0077] In the formula, is the degree of membership of the battery to be charged and the target category, is the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of the target category, is the normalization function.

[0078] It should be noted that the smaller the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of the target category, the smaller the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of the target category, which indicates that the two are highly consistent in characteristics. This consistency indicates that the battery to be tested exhibits similar characteristics to the target category in the most recent charge, and thus the membership of the battery to be charged to the target category is The larger it is, the more likely it is that the characteristics of the battery under test in the most recent charge belong to the target category.

[0079] S603 , calculating the inverse membership of the battery to be charged and the target category according to the DTW distance between the detection curve of the battery to be charged in the most recent charging and the weighted average curves of other reference categories.

[0080] Specifically, the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of other reference categories is calculated, which is recorded as ; then the reverse membership of the battery to be charged and the target category Satisfies the expression:

[0081] ;

[0082] In the formula, is the inverse membership between the battery to be charged and the target category, The detection curve of the battery to be charged in the last charge and the The DTW distance of the weighted average curve of other reference categories, is the number of all reference categories, where the other reference categories refer to the reference categories other than the target category in all reference categories, then the number of all other reference categories is equal to , is the normalization function.

[0083] It should be noted that the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of other reference categories is small, which means that the characteristics of the battery to be tested and other reference categories except the target category are similar, so that the characteristics of the battery to be tested in the most recent charge cannot be distinguished from other reference categories except the target category. Therefore, the anti-membership degree of the battery to be charged and the target category is The bigger.

[0084] It should be noted that in the present invention, the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of other reference categories is used to represent the similarity in features between the battery to be tested and other reference categories except the target category. When the similarity is high, the features of the battery to be tested in the most recent charge cannot be distinguished from those of other reference categories except the target category. Therefore, the inverse membership of the battery to be charged and the target category can be obtained.

[0085] S7. Calculate the probability that the battery to be charged belongs to each reference category according to the uncertainty, the degree of membership and the degree of inverse membership, and then obtain the estimated capacity of the battery to be charged, which is used for charging control when the charging cabinet charges the battery to be charged.

[0086] Specifically, taking any one of all reference categories as the target category, the probability that the battery to be charged belongs to the target category is Satisfies the expression:

[0087] ;

[0088] In the formula, is the degree of membership of the battery to be charged and the target category, is the inverse membership between the battery to be charged and the target category, is the uncertainty of the target category.

[0089] It should be noted that the uncertainty of the target category As the gamma coefficient, treat the membership of the rechargeable battery to the target category Gamma transformation is performed, the uncertainty of the target category The larger the value, the less reliable the target category is as the prediction result. Therefore, the transformed membership The smaller the value, the higher the degree of membership of the battery to be charged and the target category. The larger the value is, the more likely the characteristics of the battery under test in the most recent charge belong to the target category. Therefore, the transformed membership degree The larger the value, the probability that the battery to be charged belongs to the target category. The bigger.

[0090] It is further noted that the uncertainty of the target category is The difference As the gamma coefficient, treat the inverse membership of the rechargeable battery to the target category Gamma transformation is performed, the uncertainty of the target category The smaller it is, the greater the credibility of the target category as the prediction result. Therefore, the transformed anti-membership The smaller the value, the lower the anti-membership degree of the battery to be charged and the target category. The larger the value is, the less likely it is that the characteristics of the battery under test in the most recent charge belong to the target category. Therefore, the transformed inverse membership The larger the value, the probability that the battery to be charged belongs to the target category. The smaller.

[0091] It should be noted that the membership and anti-membership of the battery to be charged and the target category are two complementary indicators. When the battery to be tested shows strong similarity in the comparison between the most recently charged features and the target category, and at the same time shows strong difference in the comparison with other reference categories except the target category, it can be more confirmed that the battery to be tested belongs to the category of the most recently charged features. This double comparison method helps to make more accurate classification judgments.

[0092] Furthermore, the probability that the battery to be charged belongs to each reference category is calculated, and the average value of the capacity of all charging cycles in the reference category with the largest probability is used as the estimated capacity of the battery to be charged, which is used to control the charging when the battery to be charged is charged in the charging cabinet; the charging control method includes but is not limited to increasing the rest time in the charging cycle, lowering the charging cut-off voltage, and lowering the charging rate.

[0093] Among them, by adding rest time during the charging cycle, the battery can have enough time to cool down and reduce heat accumulation, thereby extending the battery life; as the battery ages, the maximum voltage that the battery can withstand will decrease. By lowering the charge cut-off voltage, the battery can be avoided from overcharging, thereby extending the battery life; by reducing the charging rate, the heat generated by the battery during the charging process can be reduced, thereby extending the battery life.

[0094] It should be noted that the present invention clusters all charging cycles by capacity and serial number. The performance changes of battery samples can be more deeply understood through cluster analysis. For batteries to be charged, the clustering results are an important reference for evaluating their capacity. The uncertainty of the target category and the membership and anti-membership of the battery to be charged and the target category are calculated based on the clustering results. The membership and anti-membership of the battery to be charged and the target category are two complementary indicators. When the characteristics of the battery to be tested in the most recent charge show strong similarities with the characteristics of the target category, and at the same time show strong differences with the characteristics of other reference categories except the target category, the battery to be tested can be more confirmed. The double comparison method helps to make more accurate classification judgments. The uncertainty of the target category obtained reflects the credibility of the target category as a prediction result. The membership and inverse membership of the battery to be charged and the target category are gamma transformed through the uncertainty of the target category, and then the probability that the battery to be charged belongs to the target category is calculated. The obtained probability can more accurately reflect the relationship between the battery to be charged and the target category, and then accurately obtain the capacity estimation value of the battery to be charged. When the charging cabinet charges the battery to be charged, the accuracy of charging control is improved, thereby extending the service life of the battery.

[0095] An embodiment of the present invention further discloses a charging cabinet charge and discharge control system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a charging cabinet charge and discharge control method according to the present invention is implemented.

[0096] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0097] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0098] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A charging cabinet charging and discharging control method, characterized in that: include: For the battery to be charged placed in the charging cabinet, the detection curve of the battery to be charged recently is obtained from the historical records, and multiple reference categories are selected from all categories according to the sequence number of the most recent charge; the reference category includes the battery with the sequence number equal to Any integer number of charge cycles within The most recent charging sequence number. is the preset value; All categories were obtained by: subjecting battery samples to multiple charging cycles; In each charging cycle, the detection curve of the battery sample is collected, and the impedance spectrum data obtained by performing an electrochemical impedance spectroscopy test on the battery sample is calculated based on the impedance spectrum data; Cluster all charging cycles according to the capacity and sequence number of the battery samples in each charging cycle to obtain multiple categories; For the target categories in all reference categories, the weight of each charging cycle is calculated according to the number of charging cycles with the same serial number in the target category; the detection curves of all charging cycles are weighted by the weight to obtain the weighted average curve of the target category; Calculate the mean of the DTW distances between the weighted average curve of the target category and the weighted average curves of other reference categories, and use the normalized result of the mean as the uncertainty of the target category. , and the normalized result of the mean is inversely proportional to the mean; the other reference categories refer to the reference categories other than the target category in all reference categories; The degree of membership of the battery to be charged to the target category is calculated based on the distance between the most recently charged detection curve and the weighted average curve of the target category. ; Calculate the inverse membership of the battery to be charged and the target category based on the distance between the most recently charged detection curve and the weighted average curve of other reference categories ; Calculate the probability that the battery to be charged belongs to the target class , ; The average value of the capacities of all charging cycles in the reference category with the highest probability is used as the estimated capacity of the battery to be charged, which is used to control the charging when the charging cabinet charges the battery to be charged.

2. A charging cabinet charging and discharging control method according to claim 1, characterized in that: The detection curve is composed of detection data at each time point, and the detection data includes voltage, current and temperature.

3. A charging cabinet charging and discharging control method according to claim 1, characterized in that: The calculating the capacity of the battery sample comprises: The impedance spectrum data includes impedance values ​​at different frequencies, and the impedance values ​​include real and imaginary parts, the real part is resistance, and the imaginary part is capacitance. The impedance spectrum data is analyzed using fitting software to obtain an equivalent circuit model, and the equivalent circuit model includes charge transfer resistance ; Calculate the charge transfer coefficient based on the charge transfer resistance ; The battery capacity of the battery sample is estimated based on the charge transfer coefficient, the active surface area of ​​the electrode of the battery sample, the electron transfer number of the electrode reaction of the battery sample, and the Faraday constant. The battery capacity of the battery sample is equal to the product of the charge transfer coefficient, the active surface area of ​​the electrode of the battery sample, the electron transfer number of the electrode reaction of the battery sample, and the Faraday constant.

4. A charging cabinet charging and discharging control method according to claim 1, characterized in that: The step of calculating the weight of each charging cycle according to the number of charging cycles with the same sequence number in the target category includes: The target category has a sequence number equal to The weight of the charging cycle satisfies the expression: ； In the formula, is the target category with a sequence number equal to The weight of the charging cycle, is the target category with a sequence number equal to The combined number of charging cycles, is the target category with a sequence number equal to The combined number of charging cycles, The total number of charge cycles performed when testing the battery sample. The value range is .

5. A charging cabinet charging and discharging control method according to claim 4, characterized in that: The target category has a sequence number equal to The combined number of charging cycles Satisfies the expression: ； In the formula, is the target category with a sequence number equal to The combined number of charging cycles, is the target category with a sequence number equal to The number of charge cycles, is the preset value, for An integer in the range.

6. A charging cabinet charging and discharging control method according to claim 1, characterized in that: The calculation of the membership degree of the battery to be charged and the target category ,include: Calculate the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of the target category, denoted as ; then the membership degree of the battery to be charged and the target category Satisfies the expression: , is the normalization function.

7. A charging cabinet charge and discharge control method according to claim 1, characterized in that: The calculation of the reverse membership between the battery to be charged and the target category ,include: Calculate the DTW distance between the detection curve of the battery to be charged in the most recent charge and the weighted average curve of other reference categories, denoted as ; then the reverse membership of the battery to be charged and the target category Satisfies the expression: ； In the formula, The detection curve of the battery to be charged in the last charge and the The DTW distance of the weighted average curve of other reference categories, is the number of all reference categories, where the other reference categories refer to the reference categories other than the target category in all reference categories, then the number of all other reference categories is equal to , is the normalization function.

8. A charging and discharging control system for a charging cabinet, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a charging cabinet charging and discharging control method according to any one of claims 1 to 7 is implemented.

Images