A method for calculating and evaluating the internal resistance of all battery cells for cloud-based energy storage

Abstract

The present invention discloses a method for calculating and evaluating the internal resistance of all cells for cloud-based energy storage, comprising obtaining full cell-level data and dividing and storing the obtained full cell-level data; inputting the stored full cell-level data into a preset evaluation model to obtain a cell internal resistance database; establishing a cell internal resistance data matrix through the cell internal resistance database, and calculating the eigenvectors of the cell internal resistance data matrix; and analyzing the correlation between each of the eigenvectors to evaluate the battery state. The present invention utilizes historical operating data from a big data platform, and calculates changes in cell internal resistance in a data-driven manner by fitting an equivalent mathematical model of cells. This method has the advantages of full coverage and high fitting accuracy, overcoming the problem that traditional methods only involve inaccurate calculation of internal resistance at the cluster and pack levels, and are unable to calculate and evaluate all cells in a cluster.

Description

Technical Field

[0001] The present invention relates to the field of battery management, and in particular to a method for calculating and evaluating the internal resistance of full-capacity battery cells for cloud-based energy storage. Background Art

[0002] With advances in energy storage technology, more and more energy storage systems are being connected to the grid, providing flexible and dispatchable energy reserves for building robust and smart grids. The safe operation of batteries is a primary consideration for both centralized and distributed energy storage systems. Batteries primarily consist of cells and protective circuit boards.

[0003] There are inevitably differences in the grouping, structure, assembly, and environment of battery clusters and packs between different energy storage systems, which will affect the operating characteristics of the system. At the same time, due to the high nonlinearity and degradation characteristics of lithium-ion batteries, their capacity decay is coupled with multiple physical and chemical processes. The complex decay mechanism makes the parameter changes of lithium batteries complex and multi-dimensional. Therefore, for cells with a "short board effect", it is impossible to accurately grasp the changes and trends of their internal cell-level parameters by simply conducting statistics and analysis on the batteries at the pack level, because the individual capacity decay, increase in internal resistance, internal short circuit, and increase in battery self-discharge at the cell level will greatly affect the overall operating characteristics of the system.

[0004] Energy storage systems consist of batteries, a BMS (battery management system), a PCS (power storage converter), and external auxiliary equipment. To achieve optimal operating conditions, their operational data requires proper monitoring, analysis, and management. Battery internal resistance significantly impacts the system's available power and is also a key indicator of the battery's State of Health (SoH). By analyzing the decay trend of battery internal resistance, it's even possible to predictively assess the battery's remaining service life. Therefore, determining internal resistance is crucial for addressing these issues.

[0005] Currently, the more common methods for obtaining battery internal resistance are divided into AC measurement and DC measurement methods: 1) AC measurement method (also known as AC injection method): When measuring the battery internal resistance, a small AC current is applied to the positive and negative polarities of the battery, and the internal resistance value is obtained by measuring the voltage response. Author Chen Baoming introduced a relatively common AC injection method in his article "Design and Fabrication of an Online Battery Internal Resistance Measurement Experimental System": When a signal source injects an AC current signal into the battery, the AC voltage signal generated at both ends of the battery and the input current are measured to calculate the battery's internal resistance. 2) DC measurement method (also known as DC discharge method): According to the formula R = U / I, the test equipment forces a large constant DC current through the battery for a short period of time (generally 2 to 3 seconds), measures the voltage at both ends of the battery at this time, and calculates the current battery internal resistance according to the formula. The HPPC (hybrid pulse power characteristic) test method uses the principle of DC measurement for calculation. Patent CN111257774A and patent CN111722134A both utilize the principle of the DC method, generating step power pulses within a working cycle and utilizing the changes in DC voltage and current to calculate the internal resistance.

[0006] Patent CN109856557A "A method for online monitoring of electrochemical impedance testing of lithium-ion batteries" combines the DC method with the AC method, selecting the frequency time at the inflection point of the electrochemical impedance region in the AC impedance spectrum as the pulse time of the DC impedance test, and the pulse current selects a small rate current within 1C. Through this DC impedance test of a small rate current pulse, the purpose of online monitoring of the battery electrochemical impedance changes by the DC method is achieved.

[0007] Disadvantages of AC measurement: The circuit is susceptible to external interference, and the accuracy is not as high as that of DC measurement. Because the internal resistance of the battery is in the milliohm range, similar to the resistance of the test leads, the obtained response voltage is easily affected by interference. Therefore, filtering out related interference and obtaining a valid response voltage is crucial. The measurement accuracy error of this method is generally between 1% and 2%, and it is generally used for small-capacity batteries (commonly used in electronic products).

[0008] DC measurement is generally used for testing large-capacity batteries (large-capacity lead-acid batteries or lithium batteries). The shortcomings of this method are: 1. A large current needs to be generated, otherwise the voltage difference and current difference will change very little. Using this small difference for calculation will amplify the acquisition and calculation errors; 2. The BMS working cycle is very short, and the response speed of the voltage sensor and the current sensor may be inconsistent. The delay in the voltage to current response will cause deviations in the internal resistance calculation; 3. There are deviations in a single calculation, and the internal resistance obtained is very unstable, which violates the principle of gradual change of internal resistance and requires multiple tests; 4. Online detection cannot be performed during normal system operation, and HPPC (hybrid pulse) pulse charging and discharging is required for detection, which is not convenient for continuous operation of the system.

[0009] Therefore, it is hoped that a method for calculating the internal resistance of a battery can overcome or at least alleviate the above-mentioned defects of the prior art. Summary of the Invention

[0010] In view of the above problems, the present invention is proposed to provide a method for calculating and evaluating the internal resistance of all cells of cloud-based energy storage that overcomes the above problems or at least partially solves the above problems.

[0011] According to one aspect of the present invention, a method for calculating and evaluating the internal resistance of all cells in a cloud-based energy storage system is provided, comprising:

[0012] Acquire full cell-level data, and divide and store the acquired full cell-level data;

[0013] Inputting the stored full cell-level data into a preset evaluation model to obtain a cell internal resistance database, wherein the preset evaluation model is an equivalent mathematical model established based on the Thevenin theorem;

[0014] Establishing a cell internal resistance data matrix through the cell internal resistance database, and calculating the eigenvector of the cell internal resistance data matrix;

[0015] The correlation between each of the feature vectors is analyzed to evaluate the battery state.

[0016] Optionally, the method of obtaining full cell-level data and dividing and storing the obtained full cell-level data includes obtaining full cell-level data in a battery cluster for any period of time on a big data platform or a local terminal, and dividing the data into equal time intervals.

[0017] Optionally, the cell-level full data includes current, voltage, and SoC (state of charge) curves of the battery cluster / cell.

[0018] Optionally, the equivalent mathematical model includes a first-order, second-order or multi-order RC model.

[0019] Optionally, the cell internal resistance includes ohmic internal resistance and polarization internal resistance. In a preset evaluation model, the ohmic internal resistance is simulated by a resistor value, and the polarization internal resistance is characterized by a resistor-capacitor network.

[0020] Optionally, the preset evaluation model includes:

[0021] Construct the functional relationship between the OCV opening voltage circuit and the SoC charge state and the least squares vector expression of the functional relationship;

[0022] Based on the least squares vector expression of the functional relationship, the parameter fitting values of the battery cell are obtained according to the data fragments of the SoC within the preset range;

[0023] After analyzing the ohmic internal resistance according to the parameter fitting values, a visualization result of the ohmic internal resistance of the battery cell is obtained and abnormal values are screened out.

[0024] Optionally, the screening includes simulating the distribution of battery cells from the perspectives of internal resistance difference, voltage external characteristics, and spatial consistency, wherein the internal resistance difference utilizes the data of all battery cells in the same cluster, the voltage external characteristics utilize the obtained equivalent open-circuit voltage to simulate the performance of the battery cells, and the spatial consistency is the data of battery cells in the same position between different packs.

[0025] Optionally, when processing the data, the data is first standardized, and the standardization includes scaling the data to a specification with a mean of 0 and a variance of 1.

[0026] Optionally, the method further includes normalizing the standardized data.

[0027] Optionally, the method further includes performing correlation analysis on the data.

[0028] Optionally, the method further includes quantifying the output of the cell screening evaluation to guide operation and maintenance.

[0029] Optionally, analyzing the correlation between each of the feature vectors to evaluate the battery state further includes:

[0030] Euclidean distance is used to classify the cell screening evaluation in the battery status and quantify the output results.

[0031] According to another aspect of the present invention, a cloud-based energy storage full-cell internal resistance evaluation system is provided, comprising:

[0032] A collection unit, used for collecting battery data in real time and storing the collected battery data;

[0033] a processing unit, configured to process the battery data using a model normalization method;

[0034] an analyzing unit, configured to set different time thresholds and analyze the battery data according to the different time thresholds;

[0035] An evaluation unit is used to evaluate the battery status based on different analysis results.

[0036] According to another aspect of the present invention, there is provided an electronic device, comprising: a memory and a processor;

[0037] The memory is used to store program instructions;

[0038] The processor is used to call the program instructions in the memory to execute the above-mentioned cloud-based energy storage full-cell internal resistance calculation and evaluation method.

[0039] According to another aspect of the present invention, a computer-readable storage medium is provided, in which computer program instructions are stored. When the computer program instructions are executed, the above-mentioned method for calculating and evaluating the internal resistance of the full battery cells of cloud-based energy storage is implemented.

[0040] As can be seen from the above, the technical solution of the present invention provides a method for calculating and evaluating the internal resistance of all cells in cloud-based energy storage. It uses the historical operating data of the big data platform, fits the equivalent mathematical model of the cell, and calculates the change in the internal resistance of the cell in a data-driven manner. It has the advantages of full coverage and high fitting accuracy, and overcomes the problem that the traditional method only involves inaccurate calculation of the internal resistance at the cluster and pack level, and cannot calculate and evaluate the full amount of cells in the cluster. At the same time, it also avoids the disadvantages of the traditional method of applying step pulse power to the energy storage during operation, and even requiring shutdown for testing. And based on the obtained full amount of cell internal resistance data, its parameters are evaluated and classified, and the outlier cell numbers can be quickly located and screened out, thereby assisting in guiding the operation and maintenance strategy of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0042] Figure 1 One of the voltage curves of any cluster of full-capacity cells is shown;

[0043] Figure 2 The second voltage curve of any cluster of full-capacity cells is shown;

[0044] Figure 3 The charge and discharge characteristic curve of any cluster in one day is shown;

[0045] Figure 4 Shows a schematic diagram of the battery second-order RC equivalent model;

[0046] Figure 5 Shows the R0 resistance of the cells in the battery cluster;

[0047] Figure 6a One of the three-dimensional visualization diagrams showing the R0 resistance of the battery cells in the battery cluster;

[0048] Figure 6b The second 3D visualization diagram shows the R0 resistance of the battery cells in the battery cluster;

[0049] Figure 7aA schematic diagram describing the distribution of battery cells based on internal resistance differences is shown;

[0050] Figure 7b A schematic diagram describing the distribution of battery cells based on voltage characteristics is shown;

[0051] Figure 7c A schematic diagram describing the distribution of battery cells from the perspective of spatial consistency is shown;

[0052] Figure 8a A schematic diagram showing the correlation between the internal resistance difference and the voltage external characteristic is shown;

[0053] Figure 8b A schematic diagram showing the correlation between voltage external characteristics and spatial consistency is shown;

[0054] Figure 8c A schematic diagram showing the correlation between internal resistance difference and spatial consistency is shown;

[0055] Figure 9 A schematic diagram of cell data screening and evaluation is shown. DETAILED DESCRIPTION

[0056] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0057] The technical solutions provided by various embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0058] Example 1

[0059] This embodiment provides a method for calculating and evaluating the internal resistance of all battery cells in a cloud-based energy storage system, including:

[0060] S1. Obtaining full cell-level data, and dividing and storing the obtained full cell-level data;

[0061] Those skilled in the art should know that a battery is mainly composed of two parts: a battery cell and a protection board.

[0062] This step can usually be carried out in the following manner:

[0063] On the big data platform or local end, obtain the full amount of cell-level data, which includes the voltage, current, and SoC curve of the battery cluster / cell. For example, ① obtain the voltage information of all cells in the battery cluster during a certain period of time, and divide the data into equal time intervals (seconds or minutes). Figure 1The data obtained here shows the voltage curve of all cells in energy storage battery cluster 1 on July 22, 2021, divided into one-minute time intervals. Therefore, there are 1440 time intervals in a day. The battery cluster consists of 216 cells connected in series. Therefore, the data format is 1440 x 216 voltage data. The x-axis is the time unit, the y-axis is the cell number, and the z-axis represents the voltage value of a specific cell at a specific moment.

[0064] Figure 1 and Figure 2 The voltage curve of any full battery cell is shown. Figure 2 From the curve in Figure 2, we can see that there is an obvious deviation in the voltage of cell No. 45.

[0065] ② Obtaining the current operating data of the battery cells. Since the cells in the cluster are connected in series, the current flowing through them is uniform, so only the total current curve of the cluster is required. This method is also applicable to parallel or series-parallel structures, requiring only the current data of each cell to be calculated separately.

[0066] Similarly, on the big data platform or local end Figure 3 The corresponding cluster total current, total voltage, and SoC curves are divided into one-minute time intervals, with a day divided into 1440 time intervals. The energy storage operation mode is: charging at night, discharging during peak electricity prices during the day, and replenishing electricity during flat electricity prices.

[0067] Therefore, analyzing internal resistance solely by the difference between total voltage and total current yields a total external internal resistance value, but fails to accurately and precisely describe the internal resistance of all cells. In particular, the "short board effect" of a cell can significantly impact overall operating characteristics, necessitating a precise description and quantification of the internal resistance of all cells.

[0068] Figure 3 The charge and discharge characteristic curve of an arbitrary cluster in one day is shown.

[0069] S2. Based on the Thevenin theorem, an equivalent mathematical model is established, and the stored full cell-level data is substituted into the equivalent mathematical model to obtain a cell internal resistance database;

[0070] This step can usually be carried out in the following manner:

[0071] (1) Online identification of model parameters. For example, for lithium iron phosphate batteries, an equivalent mathematical model is established using mathematical formulas. The cell models usually include RC models, second-order and multi-order RC models, etc. The internal resistance of lithium iron phosphate batteries is divided into ohmic internal resistance and polarization internal resistance. The ohmic internal resistance is composed of the electrode material, electrolyte, diaphragm resistance, and the contact resistance of each part. The polarization internal resistance refers to the resistance caused by polarization in the electrochemical reaction, including the resistance caused by electrochemical polarization and concentration polarization. It is used to simulate the dynamic characteristics of the battery during the generation and elimination of polarization.

[0072] To accurately describe the changing characteristics of the battery cell, this paper uses a second-order RC equivalent circuit model to accurately estimate nonlinear characteristics through parameter identification. In this model, the ohmic internal resistance is described by a resistor, and the polarization internal resistance is described by two RC networks. E is the battery's open-circuit voltage, which has a fixed functional relationship with the SoC at the same temperature; R0 is the battery's ohmic internal resistance; i is the current flowing through the battery. R1C1 and R2C2 form two RC circuits to simulate the dynamic characteristics of the battery during the generation and elimination of polarization.

[0073] According to Thevenin's theorem, we get the following relationship:

[0074]

[0075]

[0076] V(t)=E(s,t)-i(t)R0-i(t)R1(1-exp(-t / (R1C1)))-i(t)R2(1-exp(-t / (R2C2)))(2)

[0077] Where t is the sampling time, and E(s,t) represents the open circuit voltage of the battery. Under specific temperature conditions, it has a specific OCV and SoC function correspondence curve.

[0078] The variables to be solved are set first, namely R0, R1, C1, R2, C2, independent variable i(t), and dependent variable V(t). Construct the least squares vector expression of formula (2):

[0079] y=α*x+β (3)

[0080] Where, α = [R0, R1(1-exp(-t / (R1C1)), R2(1-exp(-t / (R2C2))], x = [i(t)], y = [V(t)-E(s,t)],

[0081] β is the residual vector. Using the least squares method to minimize the sum of squares of the residuals, we get:

[0082]

[0083] Furthermore, since V(t) and i(t) are the voltage and current at time t, they can be directly obtained on the big data platform or locally. E(s,t) is the function corresponding to OCV-SoC and has a corresponding value at time t. The quadratic function of Equation (4) is further differentiated, and the minimum value is obtained when the derivative is 0:

[0084]

[0085] and then:

[0086] V(t1)+...+V(t m )-[E(s,t1)+...+E(s,t m )]=α[i(t1)+...+i(t m )] (6)

[0087] In the formula, V(t) and i(t) are both known, while E(s, t) is unknown. However, it is clear that at any time t, E(s, t) is a value with specific practical significance. For lithium batteries whose open circuit voltage (OCV) is in a plateau period, changes in the state of charge (SoC) hardly cause changes in the open circuit voltage. Therefore, in formula (6), a data segment with SoC in the range of [0.2, 0.8] is cut out. At this time, the open circuit voltage changes less and can be equivalent to a constant value. At the same time, the calculated value of the internal resistance also changes with the change in capacity, showing a characteristic of slow changes in the middle and rapid changes at both ends. In the SoC range of [0.2, 0.8], the internal resistance can maintain good consistency, which is used to solve the equivalent internal resistance within the variable α.

[0088] The solution proposed in the present invention is to extract the data segment of SoC in the interval [0.2, 0.8] based on the above analysis, and add the equivalent open circuit voltage to be solved, which is set as the variable E, so that:

[0089] V(t1)+...+V(t n )-nE=α[i(t1)+...+i(t n )] (7)

[0090] Therefore, the variables to be solved are E, R0, R1, C1, R2, and C2 after the changes. At the same time, after the changes in formula (4), we get:

[0091]

[0092] Substitute the filtered V(t) and i(t) data into the equation (8) to solve it and get the parameter fitting values of all cells. Some of the fitting data of the cells are selected as shown in the following table:

[0093] Table 1 Battery cell data list

[0094]

[0095]

[0096] S3. Establishing a cell internal resistance data matrix through the cell internal resistance database, and calculating the eigenvector of the cell internal resistance data matrix;

[0097] This step can usually be carried out in the following manner:

[0098] First, analyze and select the ohmic internal resistance R0, and the rest of the parameters can be obtained by the same logic. Figure 5 As shown in the figure, there are obvious outliers, among which the top three are cells 45, 102, and 170. Calculating the data by 3σ yields

[0099] Mean: 0.0005192491977233525, Standard Deviation: 9.345373196741397e-05,

[0100] Lower limit: 0.00023888800182111057, upper limit: 0.0007996103936255944.

[0101] Therefore, the abnormal cell numbers and over-limit values screened out are:

[0102] [[0.0008063652519535399,18],[0.0009459146854555062,45],[0.0008162551101448272,66],[0.0009243223475185519,102],[0.000807619158532184,114],[0.0008282107912748434,162],[0.0008863232903658249,170],[0.000818269577033306,186]], a total of 8 abnormal battery cells.

[0103] A single dimension alone is not enough to identify and select key cells, so a three-dimensional visualization is constructed based on the data in Table 1. The x-axis is the pack number, the y-axis is the cell number within the pack, and the z-axis is the ohmic internal resistance value. There are a total of 18 packs*12c=216 data points. Figure 6a 、 Figure 6b shown.

[0104] from Figure 6b We discovered that the internal resistance at the sixth position in each pack was abnormally high. Considering that ohmic internal resistance is composed of the electrode material, electrolyte, diaphragm resistance, and contact resistance of various components, this is partly due to the influence of spatial structure and contact. Simply classifying these cells as abnormal is unreasonable, so we constructed a feature vector for cell screening.

[0105] like Figure 7a 、 Figure 7b 、 Figure 7c As shown in the figure, the distribution of battery cells is described from the perspectives of internal resistance difference, voltage external characteristics, and spatial consistency. The internal resistance difference uses the data of all battery cells in the same cluster, the voltage external characteristics use the equivalent open-circuit voltage obtained above to describe the performance of the battery cells, and the spatial consistency is analyzed by analyzing the data of battery cells at the same position between different packs.

[0106] The above data are first standardized and scaled to a mean of 0 and a variance of 1. Because different dimensions affect the weights used in subsequent machine learning, further normalization is performed to scale the different feature dimensions to ensure that the weights of each feature dimension on the objective function are consistent.

[0107] standardization:

[0108] Normalization:

[0109] S4. Analyze the correlation between each of the feature vectors to evaluate the battery status.

[0110] Figure 8a 、 8b ,8c describes the correlation analysis between eigenvectors, Figure 8a The correlation between the internal resistance difference and the voltage external characteristic can be seen. The spatial dispersion of the battery cell can be seen. The x-axis screens the degree of deviation in the continuous time series, and the y-axis describes the degree of voltage deviation. It can clearly obtain the deviation points where the internal resistance and voltage deviate significantly at the same time. Figure 8b The correlation between the voltage external characteristics and spatial consistency can filter out abnormal data caused by structure and contacts in terms of spatial consistency; Figure 8c The figure shows the correlation between internal resistance difference and spatial consistency. The internal resistance deviation of the battery cell at the 6th position on each pack is large, but its consistency performance is good within the same spatial distribution dimension.

[0111] Based on the above analysis, it is necessary to quantify the output of the cell screening evaluation to guide operation and maintenance. In this solution, Euclidean distance is used for classification. The Euclidean distance between two n-dimensional vectors a(x11,x12,…,x1n) and b(x11,x12,…,x1n) is:

[0112]

[0113] This solution uses Euclidean distance for clustering. Unlike the k-means algorithm, which iterates cluster centers and results in non-unique results, this solution uses data from three dimensions. Because the data has been normalized, its abnormal extreme points are fixed. A binary classification method is constructed based on the mean point of all current battery cells for screening.

[0114] The steps of the binary classification method are as follows: 1) Determine two cluster centers, which are the extreme point [1, 0, 1] and the mean point of all cells [xavg, yavg, zavg]; 2) Calculate the Euclidean distance of each cell data to the two cluster centers according to the formula; 3) Compare the distances to the two cluster centers to determine the cluster to which the cell belongs. Through the above method, the screening classification results are obtained, such as Figure 9 As shown in the figure, red circles indicate abnormal cells or cells requiring special attention, and stars indicate normal cell clusters.

[0115] Conclusion: After evaluation, there are two battery cells that require special attention, namely: battery cell No. 45 and battery cell No. 170.

[0116] In summary, the technical solution of the present invention adopts a big data-driven approach to calculate the internal resistance of the battery, and performs online parameter identification on the constructed battery cell equivalent model to fit the relationship between voltage, current and internal resistance. Unlike the DC internal resistance method, which requires the acquisition of change signals at the second or even millisecond level, and there is a problem of deviation in single measurements, the present invention records the continuous time, voltage and current values during the normal operation of the battery, and reduces the requirements for data granularity and accuracy. In process data with large current change steps, parameter fitting is more biased towards it, which can well characterize the characteristics of voltage change with current; at the same time, continuous voltage and current operation data can be obtained, and the equivalent internal resistance obtained has better adaptability and stability. At the same time, there is no need to deliberately apply power change signals using methods similar to HPPC. The present invention avoids interference with the normal operation process, and even the disadvantages of needing to shut down and stand still for open-circuit voltage measurement. Furthermore, for the internal resistance value obtained, a multi-dimensional feature vector is constructed, and the battery cell is comprehensively evaluated from the internal resistance difference, voltage external characteristics, and spatial consistency, avoiding the evaluation deviation due to a single dimension. This method is not limited to implementation in the cloud, and can be deployed locally for edge computing.

[0117] It should be noted that:

[0118] The algorithm and display provided herein are not inherently related to any particular computer, virtual device or other equipment. Various general-purpose devices can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing this type of device. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages can be utilized to implement the content of the present invention described herein, and the above description of specific languages is for the purpose of disclosing the best mode of the present invention.

[0119] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0120] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Accordingly, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the invention.

[0121] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition may be divided into multiple submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed herein may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0122] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination.

[0123] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the cloud-based energy storage full-cell internal resistance calculation and evaluation system according to an embodiment of the present invention. The present invention may also be implemented as a device or apparatus program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0124] It should be noted that the above embodiments illustrate rather than limit the invention, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between brackets should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising several different elements and by means of appropriately programmed computers. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

Claims

1. A method for calculating and evaluating the internal resistance of all cells in a cloud-based energy storage system, characterized in that: include: Acquire full cell-level data including voltage, current, and SoC curves within the battery cluster, and divide and store the acquired full cell-level data at equal time intervals; Inputting the stored full cell-level data into a preset evaluation model to obtain a cell internal resistance database including cell numbers, equivalent open-circuit voltages, and internal resistances, wherein the preset evaluation model is an equivalent mathematical model established based on the Thevenin theorem; Establishing a cell internal resistance data matrix through the cell internal resistance database, and calculating the eigenvector of the cell internal resistance data matrix; Analyzing the correlation between each of the feature vectors to evaluate the battery state; The cell internal resistance data matrix is a three-dimensional matrix, wherein the x-axis of the three-dimensional matrix is the pack number, the y-axis is the cell number inside the pack, and the z-axis is the value of the ohmic internal resistance; The characteristic vectors of the cell internal resistance data matrix include internal resistance difference, voltage external characteristics, and spatial consistency. Analyzing the correlation between each of the characteristic vectors to evaluate the battery state includes: The correlation between each pair of feature vectors is analyzed to evaluate the battery status.

2. The method according to claim 1, wherein: The cell internal resistance includes ohmic internal resistance and polarization internal resistance. In the preset evaluation model, the ohmic internal resistance is simulated by the resistance value, and the polarization internal resistance is characterized by the resistance-capacitance network.

3. The method according to claim 1, characterized in that The preset evaluation model includes: Construct the functional relationship between the OCV opening voltage circuit and the SoC charge state and the least squares vector expression of the functional relationship; Based on the least squares vector expression of the functional relationship, the parameter fitting values of the battery cell are obtained according to the data fragments of the SoC within the preset range.

4. The method according to claim 3, wherein: The method further includes: analyzing the ohmic internal resistance according to the parameter fitting value, obtaining a visualization result of the ohmic internal resistance of the battery cell and screening out abnormal values; Among them, the screening includes simulating the distribution of battery cells from the perspectives of internal resistance difference, voltage external characteristics, and spatial consistency. The internal resistance difference uses the data of all battery cells in the same cluster, the voltage external characteristics use the obtained equivalent open-circuit voltage to simulate the performance of the battery cells, and the spatial consistency is the data of battery cells in the same position between different packs.

5. The method according to claim 4, characterized in that: When processing the data, the data is first normalized, and the normalization includes scaling the data to a specification with a mean of 0 and a variance of 1; And / or, further comprising performing normalization processing on the standardized data; and / or, also includes performing correlation analysis on the data; And / or, it also includes quantitative output of battery cell screening evaluation to guide operation and maintenance.

6. The method according to claim 1, characterized in that Analyzing the correlation between each of the feature vectors to evaluate the battery state further includes: Euclidean distance is used to classify the cell screening evaluation in the battery status and quantify the output results.

7. A cloud-based energy storage full-cell internal resistance evaluation system, characterized in that: include: The acquisition unit acquires the full amount of cell-level data in the battery cluster, including voltage, current, and SoC curve, and divides and stores the acquired full amount of cell-level data at equal time intervals; a processing unit, inputting the stored full cell-level data into a preset evaluation model to obtain a cell internal resistance database including cell numbers, equivalent open-circuit voltages, and internal resistances, wherein the preset evaluation model is an equivalent mathematical model established based on the Thevenin theorem; An analysis unit, which establishes a cell internal resistance data matrix through the cell internal resistance database and calculates a eigenvector of the cell internal resistance data matrix; an evaluation unit, configured to analyze the correlation between each of the feature vectors to evaluate the battery state; The cell internal resistance data matrix is a three-dimensional matrix, wherein the x-axis of the three-dimensional matrix is the pack number, the y-axis is the cell number inside the pack, and the z-axis is the value of the ohmic internal resistance; The characteristic vectors of the battery cell internal resistance data matrix include internal resistance difference, voltage external characteristics and spatial consistency. The evaluation unit analyzes the correlation between each pair of characteristic vectors to evaluate the battery state.

8. An electronic device, characterized in that: include: memory and processor; The memory is used to store program instructions; The processor is used to call the program instructions in the memory to execute the cloud-based energy storage full-cell internal resistance calculation and evaluation method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, which, when executed, implement the method for calculating and evaluating the internal resistance of the full battery cells of cloud-based energy storage as described in any one of claims 1 to 6.

Images