Copper bar looseness identification method, device, equipment and medium

Abstract

The invention relates to the technical field of battery management, and discloses a copper bar looseness identification method and device, equipment and a medium, and the method comprises the steps: obtaining multi-modal data of a battery system; the multi-modal data comprises at least one of voltage, current and temperature and equivalent direct current resistance; determining a first looseness prediction result of the battery system based on the change rates of the at least two types of multi-modal data; calculating the outlier degree of each copper bar point location indicated by the first looseness prediction result; and according to the outlier degree of each copper bar point location, generating positioning information and a fault result of the copper bar point location. According to the scheme, the limitation of single signal diagnosis in the prior art is broken through, and the copper bar looseness is analyzed by voltage-impedance combination based on multi-modal characteristics, so that the accuracy and robustness of copper bar looseness detection are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery management, and in particular to a copper bar loosening identification method, device, equipment and medium. BACKGROUND

[0002] With the increase of the capacity of the energy storage system, the capacity and number of battery monomers increase, which leads to the increase of the size of the inter-battery connecting piece and the complication of the spatial arrangement. This change makes the stability of the connecting piece and the copper bar face greater challenges, especially the increase of the length of the copper bar connection area and the increase of the number of copper bars, thereby reducing the overall reliability of the system. Copper bar loosening is one of the common problems in the battery system.

[0003] Copper bar loosening can cause the increase of local internal resistance (due to the reduction of the contact surface between the copper bar and the pole), thereby causing a series of problems such as temperature rise, power loss and efficiency decline. If the copper bar loosening is not detected in time, it may have a great impact on the stability and safety of the energy storage system. The prior art can monitor the abnormal condition of the connecting piece by increasing the measurement circuit to measure the voltage difference between the batteries. However, this scheme is easily disturbed by external interference and cannot accurately locate the specific position of the copper bar loosening. SUMMARY

[0004] Therefore, the present application provides a copper bar loosening identification method, device, equipment and medium, which can accurately obtain the specific position of the copper bar loosening while ensuring the identification accuracy. The technical scheme is as follows.

[0005] In a first aspect, a copper bar loosening identification method is provided, the method comprising:

[0006] Obtaining multi-modal data of a battery system; the multi-modal data comprising at least one of voltage, current, temperature and equivalent direct current resistance;

[0007] Determining a first loosening prediction result of the battery system based on the change rate of at least two kinds of multi-modal data;

[0008] Calculating the outlying degree of each copper bar point indicated by the first loosening prediction result;

[0009] Generating positioning information and fault results of the copper bar point according to the outlying degree of each copper bar point.

[0010] In a possible implementation, the method further comprises:

[0011] Obtaining the voltage-temperature ratio of the copper bar connection point and the change rate of the equivalent direct current resistance of the copper bar connection point;

[0012] If at least one of the voltage-temperature ratio of the copper busbar connection point and the rate of change of the equivalent DC resistance of the copper busbar connection point exceeds a preset range, it is determined that the copper busbar connection point is loose.

[0013] In a possible implementation, determining a first looseness prediction result of the battery system based on the change rates of at least two multi-modal states further includes:

[0014] Calculating the voltage change rate and temperature change rate of the copper busbar connection point;

[0015] If the change rate of the equivalent DC resistance of the copper busbar connection point is greater than a first threshold, the voltage change rate is greater than a second threshold, and the temperature change rate increases, it is determined that the copper busbar connection point is loose.

[0016] In a possible implementation, the method further includes:

[0017] If the change rate of the equivalent DC resistance of the copper busbar connection point is less than a third threshold, the voltage of the copper busbar connection point decreases, and the temperature change rate increases, it is determined that the copper busbar connection point is short-circuited.

[0018] In a possible implementation, obtaining the equivalent DC resistance of the battery system includes:

[0019] Under a sudden current change condition of the battery system, the ratio of the voltage to the current at the copper busbar connection point is obtained as the equivalent DC resistance.

[0020] The method further comprises:

[0021] If the equivalent DC resistance exceeds a healthy threshold range under specific load conditions and is in a nonlinear growth state, it is determined that the copper busbar connection point is loose.

[0022] In a possible implementation, the method further includes:

[0023] Obtaining the historical copper busbar impedance data;

[0024] Clustering the historical copper bus impedance data according to the copper bus connection type to obtain data features corresponding to the copper bus connection type;

[0025] The connection type of the copper busbar connection point is determined according to the data feature corresponding to the copper busbar connection type.

[0026] In a possible implementation, the data features include a mean value of equivalent DC resistance and a normal distribution feature;

[0027] The calculating the outlier degree of each copper busbar point indicated by the first looseness prediction result includes:

[0028] Obtaining a first error between the equivalent DC resistance of the copper busbar point and the average value of the equivalent DC resistance;

[0029] The degree of outlier of the copper busbar position is determined according to the ratio of the first error to the normal distribution characteristic.

[0030] In a second aspect, a device for identifying a loose copper busbar is provided, the device comprising:

[0031] A data acquisition module, configured to acquire multimodal data of the battery system; the multimodal data including at least one of voltage, current, temperature, and equivalent DC resistance;

[0032] a looseness prediction module, configured to determine a first looseness prediction result of the battery system based on a rate of change of at least two multimodal data;

[0033] an outlier calculation module, configured to calculate the outlier degree of each copper busbar point indicated by the first looseness prediction result;

[0034] The fault determination module is used to generate the positioning information and fault results of the copper busbar points according to the degree of outlier of the copper busbar points.

[0035] In a third aspect, an electronic device is provided, comprising a processor and a storage medium, wherein the storage medium stores program instructions executable by the processor, and the processor executes the program instructions to perform the above-mentioned method for identifying loose copper busbars.

[0036] In a fourth aspect, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded by a processor to execute the above-mentioned method for identifying loose copper busbars.

[0037] In a fourth aspect, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the above-mentioned method for identifying loose copper busbars.

[0038] The technical solution provided by this application may have the following beneficial effects:

[0039] The present application obtains multimodal data of a battery system; the multimodal data includes at least one of voltage, current, and temperature, as well as equivalent DC resistance; based on the rate of change of at least two types of multimodal data, determines a first loosening prediction result of the battery system; calculates the degree of outliers of each copper busbar point indicated by the first loosening prediction result; and generates positioning information and fault results of the copper busbar point based on the degree of outliers of each copper busbar point. The above scheme breaks through the limitations of single signal diagnosis in the existing technology and proposes a voltage-impedance combination based on multimodal characteristics to analyze copper busbar loosening, thereby improving the accuracy and robustness of copper busbar loosening detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the specific implementation methods or the description of the prior art. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 The present invention is a flowchart of a method for identifying a loose copper busbar according to an exemplary embodiment.

[0042] Figure 2 The present invention is a flowchart of a method for identifying a loose copper busbar according to an exemplary embodiment.

[0043] Figure 3 This is a diagram of multiple charge and discharge capacity and Coulomb efficiency of the battery cluster.

[0044] Figure 4 It is the equivalent circuit diagram of internal short circuit and copper bar resistance.

[0045] Figure 5 This is the single cell voltage graph of the most recent charge.

[0046] Figure 6 It is the single cell voltage diagram of the most recent discharge.

[0047] Figure 7 It is the equivalent DC resistance of 48 cells.

[0048] Figure 8 This is a schematic diagram of the battery copper busbar connection method.

[0049] Figure 9 This is a schematic structural diagram of a device for identifying loose copper busbars provided in an embodiment of the present application.

[0050] Figure 10 It is a structural diagram of an electronic device provided by an optional embodiment of the present invention. DETAILED DESCRIPTION

[0051] As the capacity of the energy storage system increases, the capacity and number of battery cells increase, leading to an increase in the size of the connectors between batteries and a more complicated spatial arrangement. This change poses greater challenges to the stability of the connectors and copper bars, especially the increase in the length of the copper bar connection area and the increase in the number of copper bars, thereby reducing the overall reliability of the system. Loose copper bars are one of the common problems in battery systems. Loose copper bars can lead to an increase in local internal resistance (due to the reduction in the contact area between the copper bar and the pole), which in turn causes a series of problems such as temperature rise, power loss and efficiency reduction. If the loose copper bar is not detected in time, it may have a significant impact on the stability and safety of the energy storage system. At present, there are a variety of technical methods for detecting the problem of loose copper bars in battery systems, mainly including the following categories:

[0052] 1. Add a measurement circuit: For example, the "Battery System Cross-Connect Copper Busbar Voltage Calculation Method" monitors connector anomalies by measuring voltage differences between cells. While this method can detect voltage differences, it is susceptible to external interference and cannot accurately locate the specific location of a loose copper busbar.

[0053] 2. Temperature monitoring: For example, the "Method and System for Identifying Loose Copper Busbars in Battery Plug-in Boxes for Energy Storage Systems" uses temperature changes as an indicator of looseness. Temperature changes are slow and easily affected by environmental factors, resulting in delayed response times and inability to detect faults in a timely manner.

[0054] 3. Voltage-based statistical analysis: For example, the "Lithium Battery Copper Busbar Loosening Fault Diagnosis and Prediction Method" diagnoses loose copper busbar problems by analyzing battery voltage trends. However, relying solely on voltage variation statistics may not accurately distinguish between loose copper busbars and other types of faults (such as short circuits), resulting in a high misdiagnosis rate.

[0055] 4. Based on adjacent cell voltage differences: For example, the "Method, Device, and System for Monitoring Copper Busbar Connection Stability in a Power Battery System" determines connection stability by monitoring the voltage differences between adjacent battery cells. However, in some cases, the voltage differences between adjacent cells may not accurately reflect the actual condition of copper busbar loosening, especially when the battery operating conditions are complex.

[0056] 5. Based on charging voltage and static voltage: For example, the "Loose Copper Busbar Detection Method in a Battery Management System" method determines whether a copper busbar is loose by comparing the battery's charging voltage and static voltage changes. However, the changes in charging and static voltages can vary significantly across different batteries and operating conditions, potentially leading to inaccurate identification.

[0057] Most of the prior art relies on a single voltage, temperature or current data for fault diagnosis, which can detect some problems, but cannot comprehensively analyze multiple parameters (such as the comprehensive change of voltage and internal resistance), so the ability to accurately diagnose in complex working conditions is weak. In addition, these methods cannot accurately locate the specific position of the copper bar loosening, can only provide rough warning of the fault, and are difficult to implement fast and effective maintenance and processing. Especially in complex battery system topology, the existing technology cannot fully consider the heterogeneity of the connection between the batteries, resulting in obvious deficiencies in the accuracy and reliability of fault location.

[0058] To solve the above problems, the embodiments of the present application provide a copper bar loosening identification method. Figure 1 It is a method flow chart of a copper bar loosening identification method according to an exemplary embodiment. The method is applied to an electronic device, and the method comprises:

[0059] Step 101, obtaining multi-modal data of the battery system; the multi-modal data includes at least one of voltage, current, temperature and equivalent direct current resistance.

[0060] In the embodiments of the present application, in order to comprehensively and real-time capture the running state information of the battery system, it is necessary to first collect the key multi-modal data of the battery system in real time. The multi-modal data is usually obtained through the sensors built-in the battery management system (BMS) or the special sensors deployed externally, for example, the real-time voltage value of each battery monomer or module can be collected, and the voltage difference before and after the copper bar connection point (in the embodiments of the present application, the copper bar point includes the copper bar connection point, or the copper bar point is the copper bar connection point) can be collected; the current flowing through the main loop and each branch of the battery system is monitored, for example, the current information related to the charging and discharging process; the temperature of the copper bar connection point and its surrounding area is obtained.

[0061] In the embodiments of the present application, the equivalent direct current resistance (DCR) is not directly measured, but is derived by calculation under a specific working condition. Optionally, in the embodiments of the present application, the system selects the working condition where the current changes significantly (for example, when the battery system performs charging and discharging transient), and the ratio of the transient voltage step (Vstep) to the current step (Istep) of the copper bar at this time, that is, R DCR = Vstep / Istep, is measured, so as to derive the equivalent direct current resistance.

[0062] Step 102, determining a first loosening prediction result of the battery system based on the change rate of at least two multi-modal data.

[0063] In the embodiments of the present application, the dynamic change trend of the multi-modal data can realize early warning and preliminary type differentiation of the potential copper bar loosening fault.

[0064] Specifically, after obtaining the multi-modal data, the system analyzes the change rate of at least two kinds of data:

[0065] For example, the resistance change rate and the voltage-temperature ratio (VT-Ratio) can be calculated: the system continuously calculates the change rate of the copper bar connection resistance in a unit of time (Δt): dR / dt=(R DCR,t -R DCR,t-1 ) / Δt. At the same time, for the copper bar connection point, the ratio of the voltage change (ΔV) to the temperature change (ΔT) is calculated, i.e. VT-Ratio=ΔV / ΔT. This VT-Ratio plays a key role in distinguishing copper bar loosening caused anomalies from simple environmental temperature fluctuations. If the impedance increase rate or VT-Ratio exceeds the preset threshold, the system will issue a copper bar loosening warning signal.

[0066] Alternatively, in the embodiments of the present application, a sliding window analysis method can also be used to continuously monitor the copper bar resistance, voltage gradient and temperature rise trend. On this basis, the system calculates the voltage change rate (dV / dt=(V c,t -V c,t-1 ) / Δt) and the temperature change rate (dT / dt=(T c,t -T c,t-1 ) / Δt) of the copper bar connection point in a unit of time (Δt).

[0067] Step 103, calculating the out-of-range degree of each copper bar point indicated by the first loosening prediction result.

[0068] In the embodiments of the present application, the copper bar points preliminarily predicted as abnormal can be quantitatively evaluated to determine the severity of their abnormalities, providing a basis for subsequent accurate positioning and classification. Specifically, after determining the first loosening prediction result of the battery system, the system calculates the out-of-range degree of the connection impedance of these potential abnormal copper bar points. For example, the out-of-range degree of the copper bar connection impedance can be calculated by Z-Score standardization method and K-Nearest Neighbor (KNN) analysis.

[0069] Step 104, generating the positioning information and fault result of the copper bar point according to the out-of-range degree of the copper bar point.

[0070] Specifically, after calculating the out-of-range degree of each copper bar point, the system generates detailed fault results and accurate positioning information according to the out-of-range degree. In the embodiments of the present application, according to the out-of-range degree of the copper bar connection impedance and the voltage, temperature and impedance change trend distinguished in step 102, the copper bar loosening fault is subdivided into different levels:

[0071] Mild looseness: manifested as a slight increase in impedance, but no significant temperature rise. This is usually an early stage of failure, which can be arranged for regular maintenance.

[0072] Severe looseness: manifested as a sharp rise in impedance, accompanied by voltage fluctuations and temperature abnormalities. This indicates that the fault has been relatively serious, which may involve safety hazards, and needs immediate attention and treatment.

[0073] False alarm elimination: if the temperature fluctuation is significant, but the voltage or impedance does not change significantly, it is determined that it may be a false alarm caused by environmental factors, rather than copper bar looseness. This helps to reduce unnecessary dispatch and maintenance costs.

[0074] Therefore, the above scheme can accurately map the abstract resistance data to the specific physical location by analyzing the copper bar connection topology inside the battery cluster. At the same time, the system also provides a prediction result of the fault, such as predicting the trend of possible deterioration in the future, for the maintenance personnel to review and develop a preventive maintenance plan.

[0075] In summary, the present application obtains multi-modal data of a battery system; the multi-modal data includes at least one of voltage, current, temperature and equivalent direct current resistance; based on the change rate of at least two multi-modal data, a first looseness prediction result of the battery system is determined; the out-of-range degree of each copper bar point indicated by the first looseness prediction result is calculated; and the positioning information of the copper bar point and the fault result are generated according to the out-of-range degree of each copper bar point. The above scheme breaks through the limitation of single signal diagnosis in the prior art, and proposes a voltage-impedance joint analysis of copper bar looseness based on multi-modal features, thereby improving the accuracy and robustness of copper bar looseness detection.

[0076] Figure 2 A method flowchart of a copper bar looseness identification method according to an exemplary embodiment is shown. The method is applied to an electronic device, and the method includes:

[0077] Step 201, obtaining multi-modal data of a battery system; the multi-modal data includes at least one of voltage, current, temperature and equivalent direct current resistance.

[0078] In the embodiment of the present application, when the multi-modal data collected by the sensor is obtained, a sensor adaptive calibration algorithm can be used to compensate for the drift of the temperature and voltage data. Taking temperature as an example:

[0079] T adjust = T raw - ΔT sensor_drift

[0080] Where T adjust is the adjusted temperature value of the sensor; T raw is the measured temperature value of the sensor; ΔTsensor_drift The drift temperature value of the sensor. Optionally, the system continuously monitors the historical data of the sensor, and combines the environmental conditions, sensor model characteristics, and cross-validation with other related sensors (e.g., multiple temperature sensors in the same area), to intelligently evaluate and determine the drift amount (AT sensor_drift ) of the sensor.

[0081] Step 202, determining the first looseness prediction result of the battery system based on the change rate of at least two kinds of multi-modal data.

[0082] Optionally, step 202 includes: obtaining the voltage-temperature ratio of the copper bar connection point and the change rate of the equivalent DC resistance of the copper bar connection point; if at least one of the voltage-temperature ratio of the copper bar connection point and the change rate of the equivalent DC resistance of the copper bar connection point exceeds the preset range, it is determined that the copper bar connection point is loose.

[0083] Specifically, the system calculates the ratio of the voltage change amount (AV) and the temperature change amount (AT) of the copper bar connection point, i.e. the voltage-temperature ratio (VT-Ratio = AV / AT). The VT-Ratio plays a key role in distinguishing between copper bar looseness-induced abnormalities and simple environmental temperature fluctuations, because the increase in contact resistance caused by copper bar looseness will simultaneously cause an increase in local voltage drop and an increase in temperature.

[0084] At the same time, the system continuously calculates the change rate of the copper bar connection resistance in unit time (At): dR / dt = (RDCR,t-RDCR,t-1) / At. If the impedance increase rate or VT-Ratio exceeds the preset threshold, the system will issue a copper bar looseness warning signal, indicating that the copper bar connection point may be loose. The above warning mechanism can enable the system to intervene before the failure escalates, thereby avoiding safety hazards and large-scale shutdowns.

[0085] Optionally, step 202 includes: calculating the voltage change rate and the temperature change rate of the copper bar connection point; if the change rate of the equivalent DC resistance of the copper bar connection point is greater than a first threshold, the voltage change rate is greater than a second threshold, and the temperature change rate is increasing, it is determined that the copper bar connection point is loose.

[0086] Specifically, the system uses a sliding window analysis method to continuously monitor the copper bar resistance, voltage gradient, and temperature rise trend. On this basis, the voltage change rate (dV / dt = (V c,t -V c,t-1 ) / At) and the temperature change rate (dT / dt = (T c,t -T c,t-1 ) / At) of the copper bar connection point in unit time (At) are calculated.

[0087] If the monitored rate of change of the equivalent DC resistance (dR / dt) of the copper busbar connection point is greater than the preset first threshold (indicating that the impedance continues to increase), and its voltage change rate (dV / dt) is greater than the preset second threshold (indicating significant voltage fluctuations), and the temperature change rate (dT / dt) shows an increasing trend, then it is comprehensively judged that the copper busbar connection point is loose.

[0088] Optionally, step 202 includes: if the change rate of the equivalent DC resistance of the copper busbar connection point is less than a third threshold, the voltage of the copper busbar connection point decreases, and the temperature change rate increases, determining that the copper busbar connection point is short-circuited.

[0089] Specifically, in dynamic trend analysis, the system can not only identify loose copper busbars, but also accurately distinguish other fault types. If the rate of change of the equivalent DC resistance (dR / dt) at the copper busbar connection point is less than the preset third threshold (indicating no significant increase in impedance), but the voltage at the busbar connection point is abnormally low and the rate of temperature change (dT / dt) is increasing, the system will comprehensively identify the anomaly as a short circuit fault.

[0090] Optionally, under the current mutation condition of the battery system, the voltage to current ratio of the copper busbar connection point is obtained as the equivalent DC resistance; if the equivalent DC resistance exceeds the healthy threshold range under specific load conditions and is in a nonlinear growth, it is determined that the copper busbar connection point is loose.

[0091] Specifically, the system will identify the moment when the current of the battery system undergoes a significant step change during the charging and discharging process (transient operating condition), and under this condition, the transient voltage step (V step ) and current step (I step ) as the equivalent DC resistance of the copper busbar connection point (R DCR =V step / I step ).

[0092] The system sets a healthy threshold range for the copper busbar's DCR and uses a dynamic adjustment strategy to update this threshold range based on historical DCR data to improve adaptability. If the acquired equivalent DC resistance (DCR) under specific load conditions exceeds its healthy threshold range and shows an accelerating, nonlinear growth trend, the copper busbar connection may be loose.

[0093] Step 203: Obtain a first error between the equivalent DC resistance of the copper busbar point and the average value of the equivalent DC resistance.

[0094] Specifically, after determining the potential loosening prediction results in step 202, the system obtains the current equivalent DC resistance values ​​for the indicated copper busbar points. Simultaneously, based on the specific type of copper busbar connection (e.g., inter-module jumper, intra-module jumper, or standard inter-unit connection), the system obtains the average equivalent DC resistance (μR) of that type of copper busbar connection from a pre-established historical data model. The system then calculates the difference between the equivalent DC resistance of that copper busbar point and the average equivalent DC resistance (μR) of the corresponding type, i.e., the first error (RDCR - μR). This error value directly reflects the absolute amount by which the current measurement value deviates from its normal average value.

[0095] Step 204 : determining the outlier degree of the copper busbar point according to the ratio of the first error to the normal distribution characteristic.

[0096] Specifically, after obtaining the first error (RDCR-μR) in step 203, the system will calculate the outlier degree of the copper busbar point in combination with the normal distribution characteristics corresponding to the copper busbar connection type, especially its standard deviation (σR).

[0097] Z-Score standardization: The system will use the Z-Score standardization method to divide the first error by the standard deviation of the equivalent DC resistance of the copper busbar connection type (σ R ), calculate the Z-Score value: Z=(R DCR -μ R ) / σ R The Z-Score value quantifies how many standard deviations the current DCR value deviates from its category mean, providing a measure of statistical anomaly relative to the overall distribution.

[0098] K-nearest neighbor (KNN) analysis: Optionally, in this embodiment of the present application, the system also incorporates K-nearest neighbor (KNN) analysis to further assess the degree of outliers at the busbar locations. KNN adds a layer of context by evaluating the distance or similarity between the data point and its K nearest neighbors, making outlier detection more robust. A point may have a high Z-score, but if its neighbors also exhibit similar (albeit high) values, it may still be considered "normal," and vice versa.

[0099] In an embodiment of the present application, the computer device can also obtain the historical copper bus impedance data; cluster the historical copper bus impedance data according to the copper bus connection type to obtain data features corresponding to the copper bus connection type; and determine the connection type of the copper bus connection point based on the data features corresponding to the copper bus connection type.

[0100] Specifically, the computer device will acquire and store the historical impedance data of all copper busbar connection points in the battery cluster for a long time. Since the copper busbar connections within the battery cluster are not all homogeneous (for example, the copper busbars connecting different battery modules and the copper busbars connecting the battery cells within a single module may be different in design), the system will calculate the impedance data based on the known physical connection topology of the battery cluster (for example, Figure 6 The copper busbar connection mode shown in FIG2 is used to classify and cluster the historical copper busbar impedance data.

[0101] The system can obtain data features corresponding to each copper busbar connection type, such as the mean (μR) and standard deviation (σR) of its equivalent DC resistance. For example, in this embodiment, all steady-state resistance data is clustered into three categories: inter-module jumper resistance (mean μ = 0.0011Ω), intra-module jumper resistance (mean μ = 0.0009Ω), and inter-cell common resistance (mean μ = 0.0008Ω).

[0102] When determining the outlier level of a copper busbar connection point, the system first determines the specific connection type based on its physical location and topology information. It then uses the data features corresponding to that connection type (i.e., its specific μR and σR) to calculate the Z-Score and outlier level.

[0103] Step 205: Generate positioning information and fault results of each copper busbar point according to the degree of outlier of each copper busbar point.

[0104] This solution uses combined voltage-impedance analysis to identify loose busbar faults earlier, improving the accuracy and real-time nature of fault detection. Sliding window trend analysis provides early warning of loose busbars, enabling maintenance personnel to intervene proactively and avoid potential safety hazards, energy loss, and equipment overheating.

[0105] This solution also combines equivalent DC resistance (DCR) calculation and voltage-temperature ratio (VT-Ratio) analysis to quantitatively assess the copper busbar connection status. Using Z-Score normalization and KNN nearest neighbor analysis, the looseness of the copper busbar is classified as mild, severe, or false positive, significantly reducing the complexity and misdiagnosis rate of manual inspections. Steady-state resistance comparison accurately distinguishes between inter-module copper busbar connection anomalies and loose individual copper busbars, enabling more precise fault location.

[0106] Furthermore, the use of an adaptive sensor calibration algorithm effectively reduces false alarms caused by temperature drift, signal noise, and other factors, improving system stability. Combined with battery system topology analysis, the diagnostic strategy for copper busbar connections between and within modules is optimized to ensure the monitoring solution is suitable for different energy storage system types. Through intelligent trend analysis and data adaptive learning, the generalization capability of copper busbar loosening detection is improved, enabling it to adapt to diverse operating conditions (high and low temperature environments, different discharge rates, etc.).

[0107] This method also reduces reliance on single sensor signals through multimodal fusion (voltage + current + temperature + impedance), lowering hardware deployment costs. Through a data-driven dynamic optimization model, it reduces unnecessary manual inspections and maintenance costs caused by false alarms, improving the operational efficiency and cost-effectiveness of the energy storage system.

[0108] The above steps are described below with an example of a specific implementation method:

[0109] A certain energy storage battery cluster is observed to have a serious output shortage, such as Figure 3 As shown in the figure, the charge and discharge capacity of the last 8 times show that the CE-HI change trend is: the decrease of 1% for more than 3 consecutive times, and the number of times below 96% is greater than 4. At this time, the equivalent circuit diagram of the battery system can be drawn, as shown in the figure. Figure 4 , it can be seen that the copper row resistance value is the series resistance of the external circuit, and the internal short circuit is the resistance in parallel with the battery cell. Extract the recent voltage data of the abnormal module as follows Figure 5 and Figure 6 As shown in the figure, the voltage of cell 4 is 82.2% higher during charging and 100% lower during discharging. Based on the above analysis, it can be preliminarily determined that there is a problem with the overall structure of cell 4 and its connectors. To further determine the location of the copper busbar and quantify the extent of the fault, the steady-state resistance of the four adjacent modules was calculated. The results are shown in the figure. Figure 7 As shown. It can be seen that there is an obvious arrangement pattern, that is, every 12 monomers form a module, and the connection resistance between modules is obviously the largest. In a single module, the fourth monomer has a jumper, and the resistance is slightly larger than the remaining resistance. Compared with the actual topology of the cluster ( Figure 8 )Comparison shows that the calculation method is accurate. Extracting the normal operation data of the past year, clustering all the steady-state resistance data, the results are divided into three categories:

[0110] Category 1: Inter-module jumper resistance, showing a normal distribution with a mean value of μ = 0.0011Ω.

[0111] Category 2: The jumper resistance within the module shows a normal distribution with a mean value of μ = 0.0009Ω.

[0112] Category 3: Common resistance between cells, showing a normal distribution with a mean value of μ = 0.0008Ω.

[0113] However, the data of No. 4 monomer is out of range, and the value is 0.001052Ω. The Z-Score of No. 4 monomer is calculated as-0.33, indicating that the resistance value is about 0.33 standard deviations lower than the average. According to all the steady-state resistance data, the value is still in the normal range, but it is seriously out of range for the category it belongs to. After manual review, it is determined that the copper bar at this position is loose, and the battery cluster operates normally after repair.

[0114] In summary, the multi-modal data of the battery system is obtained in the application; the multi-modal data includes at least one of voltage, current, temperature and equivalent DC resistance; the first looseness prediction result of the battery system is determined based on the change rate of at least two kinds of multi-modal data; the out-of-range degree of each copper bar point indicated by the first looseness prediction result is calculated; and the positioning information and the fault result of the copper bar point are generated according to the out-of-range degree of each copper bar point. The above-mentioned scheme breaks through the limitation of single signal diagnosis in the prior art, and proposes to analyze the copper bar looseness based on the voltage-impedance combination of multi-modal features, thereby improving the accuracy and robustness of the copper bar looseness detection.

[0115] In the embodiments of the application, a copper bar looseness identification device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be described again. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware or a combination of software and hardware is also possible and contemplated.

[0116] The embodiments of the application provide a copper bar looseness identification device, Figure 9 is a structural schematic diagram of a copper bar looseness identification device provided by the embodiments of the application, which comprises:

[0117] The data acquisition module 901 is configured to acquire multi-modal data of a battery system; the multi-modal data includes at least one of voltage, current, temperature and equivalent DC resistance;

[0118] The looseness prediction module 902 is configured to determine a first looseness prediction result of the battery system based on the change rate of at least two kinds of multi-modal data;

[0119] The out-of-range calculation module 903 is configured to calculate the out-of-range degree of each copper bar point indicated by the first looseness prediction result;

[0120] The fault determination module 904 is configured to generate the positioning information and the fault result of the copper bar point according to the out-of-range degree of each copper bar point.

[0121] In summary, the present application obtains multimodal data of the battery system; the multimodal data includes at least one of voltage, current, temperature and equivalent DC resistance; based on the rate of change of at least two types of multimodal data, determines the first loosening prediction result of the battery system; calculates the degree of outliers of each copper busbar point indicated by the first loosening prediction result; and generates the positioning information of the copper busbar point and the fault result according to the degree of outliers of each copper busbar point. The above scheme breaks through the limitations of the single signal diagnosis of the existing technology, and proposes a voltage-impedance combination based on multimodal characteristics to analyze the looseness of the copper busbar, thereby improving the accuracy and robustness of the copper busbar loosening detection.

[0122] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0123] The above-mentioned apparatus is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0124] See also Figure 10 , Figure 10 is a structural diagram of an electronic device provided by an optional embodiment of the present invention, such as Figure 10 As shown, the electronic device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface).

[0125] The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0126] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0127] The memory 20 can include a program storage area that can store an operating system, application programs required for at least one function, and a data storage area that can store data created according to the use of the electronic device according to the presentation of a small program landing page, and the like. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory such as at least one of a magnetic disk storage device, a flash memory device, or other non-transitory solid state memory device. The memory 20 can include a volatile memory such as a random access memory, and the memory can also include a non-volatile memory such as a flash memory, a hard disk, or a solid state disk, and the memory 20 can further include a combination of the above-mentioned kinds of memories.

[0128] The electronic device further includes a communication interface 30 for communication of the electronic device with other devices or communication networks.

[0129] The embodiments of the present application also provide a computer readable storage medium, and the above-mentioned method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or stored in a remote storage medium or a non-transitory machine readable storage medium by downloading through a network and stored in a local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special purpose hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc.; further, the storage medium can also include a combination of the above-mentioned kinds of memories. It can be understood that the computer, processor, microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, processor, or hardware, the method shown in the above embodiments is implemented.

[0130] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A method for identifying loose copper busbars, characterized in that: The method comprises: Acquire multimodal data of the battery system; the multimodal data includes at least one of voltage, current, temperature, and equivalent DC resistance; determining a first looseness prediction result of the battery system based on a rate of change of at least two multimodal data; Calculating the degree of outlier of each copper busbar point indicated by the first loosening prediction result; According to the degree of outlier of each copper busbar point, the positioning information and fault result of the copper busbar point are generated.

2. The method according to claim 1, characterized in that The determining a first looseness prediction result of the battery system based on the change rates of at least two multi-modal states includes: Obtain the voltage-temperature ratio of the copper busbar connection point and the rate of change of the equivalent DC resistance of the copper busbar connection point; If at least one of the voltage-temperature ratio of the copper busbar connection point and the rate of change of the equivalent DC resistance of the copper busbar connection point exceeds a preset range, it is determined that the copper busbar connection point is loose.

3. The method according to claim 2, characterized in that The determining a first looseness prediction result of the battery system based on the change rates of at least two multi-modal conditions further includes: Calculating the voltage change rate and temperature change rate of the copper busbar connection point; If the change rate of the equivalent DC resistance of the copper busbar connection point is greater than a first threshold, the voltage change rate is greater than a second threshold, and the temperature change rate increases, it is determined that the copper busbar connection point is loose.

4. The method according to claim 3, characterized in that The method further comprises: If the change rate of the equivalent DC resistance of the copper busbar connection point is less than a third threshold, the voltage of the copper busbar connection point decreases, and the temperature change rate increases, it is determined that the copper busbar connection point is short-circuited.

5. The method according to any one of claims 1 to 4, characterized in that: Obtaining the equivalent DC resistance of the battery system, including: Under a current mutation condition of the battery system, the ratio of the voltage to the current at the copper busbar connection point is obtained as the equivalent DC resistance; The method further comprises: If the equivalent DC resistance exceeds a healthy threshold range under specific load conditions and is in a nonlinear growth state, it is determined that the copper busbar connection point is loose.

6. The method according to claim 5, characterized in that The method further comprises: Obtain historical copper bus impedance data; Clustering the historical copper bus impedance data according to the copper bus connection type to obtain data features corresponding to the copper bus connection type; The connection type of the copper busbar connection point is determined according to the data feature corresponding to the copper busbar connection type.

7. The method according to claim 6, characterized in that The data characteristics include the mean value of equivalent DC resistance and normal distribution characteristics; The calculating the outlier degree of each copper busbar point indicated by the first looseness prediction result includes: Obtaining a first error between the equivalent DC resistance of the copper busbar point and the average value of the equivalent DC resistance; The degree of outlier of the copper busbar position is determined according to the ratio of the first error to the normal distribution characteristic.

8. A device for identifying loose copper busbars, characterized in that: The device comprises: A data acquisition module, configured to acquire multimodal data of the battery system; the multimodal data including at least one of voltage, current, temperature, and equivalent DC resistance; a looseness prediction module, configured to determine a first looseness prediction result of the battery system based on a rate of change of at least two multimodal data; an outlier calculation module, configured to calculate the outlier degree of each copper busbar point indicated by the first looseness prediction result; The fault determination module is used to generate the positioning information and fault results of the copper busbar points according to the degree of outlier of the copper busbar points.

9. An electronic device, characterized in that: The electronic device includes a processor and a storage medium, wherein the storage medium stores program instructions executable by the processor, and the processor executes the program instructions to perform the method for identifying loose copper busbars according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The storage medium stores at least one instruction, and the at least one instruction is loaded by the processor to execute the method for identifying a loose copper busbar according to any one of claims 1 to 7.

Images