A bus multiplexing-based energy storage battery pack fault monitoring system and method

Abstract

The application discloses a kind of energy storage battery pack fault monitoring systems and methods based on bus multiplexing, it is related to big data analysis technical field, the present application is accurately divided and orderly managed by voltage data unit level, and the voltage fluctuation value of battery unit is calculated in combination with bus voltage acquisition device, creates the circuit ringing detection set screening fluctuation data of associated unique identification, while analyzing historical voltage data extracts standardization fluctuation characteristics, determine fluctuation cascade range in combination with energy storage battery pack physical connection topology, break through the limitation of only paying attention to single unit fluctuation of existing technology;Again, fluctuation cascade range and type are grouped to build voltage recovery duration curve, the next recovery duration is predicted by neural network model training, actual and predicted duration are compared in real time to determine line state, overcome the deficiency of single threshold detection in prior art, realize accurate abnormal monitoring of different fluctuation data of bus multiplexing energy storage battery pack, effectively guarantee line reliable operation.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, specifically to a fault monitoring system and method for energy storage battery packs based on bus multiplexing. Background Technology

[0002] Energy storage battery packs are widely used in new energy power generation, energy storage power stations, electric vehicles and other fields. Their operating status directly affects the reliability and safety of related systems. If problems such as cell aging, overcharging and over-discharging, line faults and signs of thermal runaway are not detected in time, it can easily lead to a significant decline in battery performance, system shutdown, or even fire, explosion and other safety accidents, causing serious safety hazards. Therefore, real-time and accurate fault monitoring of energy storage battery packs is the core means to ensure the stable operation of battery packs, extend their service life and avoid safety risks, and is of great significance to promoting the safe and efficient development of the energy storage industry.

[0003] In bus-multiplexed energy storage battery packs, the signals of each battery cell share the same transmission bus, which is prone to superimposed interference. Furthermore, due to differences in production batches, charge-discharge cycles, and environment, the electrochemical characteristics of each cell decay differently, and abnormal signals are easily masked by bus noise. In addition, the interference from voltage ringing during charge-discharge switching makes it difficult for existing detection methods to accurately extract the true state of the cells and identify potential bus faults, thus failing to ensure stable system operation. Therefore, there is an urgent need for a fault monitoring system and method for bus-multiplexed energy storage battery packs. Summary of the Invention

[0004] The purpose of this invention is to provide a fault monitoring system and method for energy storage battery packs based on bus multiplexing, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a fault monitoring method for energy storage battery packs based on bus multiplexing, the fault monitoring method for energy storage battery packs comprising the following steps:

[0006] Step S1: Obtain the operating voltage data of each battery cell in the energy storage battery pack. Divide the obtained operating voltage data into individual battery cells as the basic unit to construct an energy storage battery cell monitoring set. Analyze the data to obtain multiple real-time voltage fluctuation curves. Combine the voltage frequency to calculate the real-time voltage of each battery cell.

[0007] Step S1-1: Deploy voltage acquisition sensors to collect voltage data information of the energy storage battery pack. Specifically, the detection end of each sensor is electrically connected to the positive and negative terminals of a single battery cell in the energy storage battery pack to obtain the voltage data information of a single battery cell in the energy storage battery pack. Specifically, the original voltage dataset is formed by recording the unique battery cell identifier code and acquisition timestamp corresponding to each voltage value.

[0008] Step S1-2: Sort the voltage data set of each battery cell in time sequence, based on the order of the acquisition timestamps, to obtain the voltage time sequence of a single battery cell; extract the key information from the voltage time sequence of each battery cell, including the unique battery cell identifier code, the acquisition timestamp, and the voltage value corresponding to the acquisition timestamp. Select the unique battery cell identifier code as the key, and select the acquisition timestamp and the voltage value corresponding to the acquisition timestamp as the value to construct the energy storage battery cell monitoring set.

[0009] By deploying voltage acquisition sensors to stimulate and record the unique battery cell identifier code and acquisition timestamp corresponding to each voltage value, the voltage data information of a single battery cell can be accurately obtained while avoiding data confusion between cells. The resulting original voltage dataset has cell traceability. Then, the voltage data group of each battery cell is sorted according to the acquisition timestamp to obtain the voltage time series data sequence. After extracting key information, the energy storage battery cell monitoring set is constructed using the unique battery cell identifier code as the key. This enables precise cell-level division and orderly management of voltage data, providing an accurate and well-organized data foundation for subsequent traversal reading of the monitoring set and carrying out battery cell voltage monitoring.

[0010] Steps S1-3: Sequentially read the key-value pair data corresponding to the unique battery unit identifier code in the energy storage battery unit monitoring set; for a single battery unit, select the acquisition timestamp as the horizontal axis coordinate and the voltage value corresponding to the acquisition timestamp as the vertical axis coordinate, and mark all data points in the two-dimensional coordinate system; connect adjacent data points with continuous line segments according to the order of acquisition timestamps to form the real-time voltage fluctuation curve of the battery unit; synchronously process all other battery units in the energy storage battery unit monitoring set to obtain the real-time voltage fluctuation curve of each battery unit;

[0011] Step S1-4: When the bus voltage meets the preset target voltage condition, for the real-time voltage fluctuation curve of each battery cell in the energy storage battery pack, the voltage value is filtered based on the frequency of occurrence. Specifically, the mode of the voltage value in each real-time voltage fluctuation curve is selected as the real-time voltage of the battery cell. The real-time voltage of the battery cell is represented as the reference voltage under the normal and stable operation of the energy storage battery pack, and satisfies the preset target voltage that the real-time voltage of the battery cell is equal to the bus voltage.

[0012] By sequentially reading the key-value pairs corresponding to the unique battery cell identifier codes in the energy storage battery cell monitoring set, marking data points with the acquisition timestamp as the horizontal axis and the corresponding voltage value as the vertical axis, and connecting them in chronological order, a real-time voltage fluctuation curve for each battery cell is generated synchronously. This can intuitively present the dynamic change trend of the voltage of a single battery cell over time, realizing the visualization of cell-level voltage fluctuations. When the bus voltage meets the preset target voltage, the mode of the voltage value in the real-time voltage fluctuation curve is used as the real-time voltage of the battery cell. A stable benchmark can be determined based on the frequently occurring voltage value, ensuring the accuracy of the real-time voltage of the battery cell and providing a reliable cell benchmark voltage basis for subsequent analysis of the fluctuation range in conjunction with the bus voltage.

[0013] Step S2: Obtain the real-time bus voltage of the energy storage battery pack and analyze the real-time voltage of each battery cell to obtain the voltage fluctuation value of each battery cell; and create a circuit ringing detection set to filter and store the voltage fluctuation data of each battery cell according to the voltage fluctuation range.

[0014] Step S2-1: Deploy the bus voltage acquisition device to acquire the real-time bus voltage of the multiplexed bus of the energy storage battery pack and record the acquisition timestamp corresponding to the real-time bus voltage; for each battery cell, extract its real-time battery cell voltage and corresponding acquisition timestamp, compare the voltage values ​​with frequencies less than the real-time voltage at the same acquisition timestamp with the real-time bus voltage, and calculate the absolute difference between the voltage values ​​and the real-time bus voltage as the voltage fluctuation value of the battery cell.

[0015] Step S2-2: Create an independent circuit ringing detection set for each battery cell, and associate the set with the corresponding unique battery cell identifier code; iterate through the voltage data of the battery cell and add data, specifically:

[0016] When the actual voltage collected in a single battery cell does not exceed the real-time voltage minus the voltage fluctuation value, it is judged as an undervoltage fluctuation and added to the circuit ringing detection set; when the voltage in a single battery cell exceeds the real-time voltage plus the voltage fluctuation value, it is judged as an overvoltage fluctuation and added to the circuit ringing detection set.

[0017] Step S2-3: Synchronously process the data addition operation of each battery unit, and complete the creation and data filtering of each circuit ringing detection set;

[0018] The system collects real-time bus voltage from the bus voltage acquisition device and records the corresponding acquisition timestamp. For each battery cell, its real-time voltage and corresponding timestamp are extracted. The voltage fluctuation value is calculated by taking the voltage value with a frequency less than the real-time voltage at the same timestamp as the benchmark and calculating the absolute difference with the real-time bus voltage. This can accurately define the benchmark boundary of cell-level voltage fluctuation and ensure the pertinence of fluctuation value calculation. By creating a dedicated circuit ringing detection set with a unique identifier code for each battery cell, data is added by traversing undervoltage fluctuation and overvoltage fluctuation. Then, all cells are processed simultaneously to complete the set creation and filtering. This enables independent classification and accurate filtering of cell-level fluctuation data, avoiding confusion of data from different cells and providing a regular and reliable data source for subsequent fluctuation monitoring based on the ringing set.

[0019] Step S3: Read and analyze the historical voltage monitoring data corresponding to each battery cell to obtain multiple different voltage fluctuation characteristics corresponding to each battery cell. Perform consistency characterization processing on the multiple voltage fluctuation characteristics obtained from the analysis to obtain the cascade range of voltage fluctuation impact, which is denoted as the fluctuation cascade range.

[0020] Step S3-1: Read the historical voltage monitoring data corresponding to each battery cell. The historical voltage monitoring data is associated with a unique battery cell identification code. The historical voltage monitoring data is classified according to undervoltage fluctuations and overvoltage fluctuations. Voltage fluctuation characteristics under the same fluctuation type are extracted. The voltage fluctuation characteristics include fluctuation duration, fluctuation peak value, deviation value from bus voltage, and fluctuation start time.

[0021] Step S3-2: Standardize the characteristics of the same fluctuation type, specifically: unify the calculation benchmark for the duration of fluctuation, from the start of the fluctuation to the moment when the fluctuation recovers to the real-time voltage of the battery cell; unify the calculation benchmark for the peak deviation of fluctuation, using the bus voltage as the benchmark, and calculate the difference between the peak value of the fluctuation and the bus voltage.

[0022] Step S3-3: Select battery cells with voltage fluctuations from historical voltage monitoring data as target battery cells and associate them with their unique identification codes; based on the physical connection topology of the energy storage battery pack, obtain the directly electrically connected adjacent cells of the target battery cell and their corresponding unique identification codes.

[0023] Step S3-4: Set a fluctuation correlation timing window. The window starts at the time when the fluctuation of the target battery cell begins and ends at the time when the voltage of the target battery cell recovers to the real-time voltage. Within the window, detect the historical voltage data of adjacent cells to find voltage fluctuations of the same type as those of the target battery cell. The same type means that the overvoltage fluctuation or undervoltage fluctuation is consistent and the peak deviation of the fluctuation after standardization belongs to the same range. Record the target battery cell and the adjacent cells that have the same type of fluctuation to obtain the fluctuation cascade range corresponding to the voltage at the time of fluctuation start under different voltage fluctuation types.

[0024] By reading historical voltage data associated with unique identifiers, features are extracted and standardized according to fluctuation type to ensure feature consistency. The target battery unit and adjacent units are determined by combining physical connection topology. A time window is set to detect fluctuations of the same type. The fluctuation cascade range is obtained by recording the target unit and its corresponding adjacent units, accurately defining the range of units affected by fluctuations, and providing clear boundary basis for subsequent monitoring.

[0025] Step S4: Monitor the circuit ringing detection set through the fluctuation cascade range, and analyze the circuit ringing detection set based on voltage fluctuation data. The time required for the voltage to recover to the real-time bus voltage in the fluctuation cascade range under different voltage fluctuation data is recorded as the voltage recovery time.

[0026] Real-time acquisition of voltage data for each battery cell; when undervoltage or overvoltage fluctuations are detected in a single battery cell, the start time and corresponding voltage value of the fluctuation are recorded. Based on the type of the fluctuation and the standardized peak deviation of the fluctuation, the fluctuation cascade range under the same fluctuation type and the same peak deviation range in step S3-4 is matched to determine the fluctuation cascade range corresponding to the voltage fluctuation.

[0027] Call the historical data that matches the current fluctuation type and fluctuation cascade range from the historical voltage monitoring data, and extract the time data of the voltage within the fluctuation cascade range recovering from the fluctuation state to the real-time bus voltage from the historical data; take the start time of the current fluctuation as the timing start point, monitor the voltage data of each battery cell within the fluctuation cascade range in real time until the voltage of all battery cells recovers to the real-time bus voltage, and record the time from the timing start point to the recovery time, which is recorded as the voltage recovery time corresponding to the voltage fluctuation data;

[0028] By collecting real-time voltage data from each battery cell, when undervoltage or overvoltage fluctuations are detected in a single battery cell, the start time and corresponding voltage value of the fluctuation are recorded. The corresponding fluctuation cascade range is matched according to the fluctuation type and the deviation of the standardized fluctuation peak value. Then, the historical records of the same fluctuation type and the same cascade range are called up. At the same time, the voltage of each cell within the fluctuation cascade range is monitored in real time until the real-time voltage of the recovery bus is restored, and the voltage recovery time is recorded. The recovery time under different voltage fluctuation data can be accurately obtained, providing an accurate time basis for subsequent prediction and anomaly monitoring.

[0029] Step S5: Based on the voltage fluctuation data and voltage recovery time of the energy storage battery pack, construct multiple voltage recovery time curves based on the same voltage fluctuation data; use a neural network model to process and predict each voltage recovery time curve to obtain the next voltage recovery time of each voltage recovery time curve, which is recorded as the predicted voltage recovery time. Use the predicted voltage recovery time to monitor the abnormal voltage fluctuation data of the bus-reused energy storage battery lines.

[0030] Step S5-1: Divide the voltage fluctuation data and corresponding voltage recovery time into several groups according to the fluctuation cascade range and fluctuation type. Each group corresponds to a unique fluctuation cascade range identifier and fluctuation type label. Sort the voltage fluctuation data in each group according to the order of occurrence time and determine them as the first historical fluctuation, the second historical fluctuation, ..., the nth historical fluctuation in sequence. Record the voltage recovery time corresponding to each fluctuation to obtain the associated dataset of the relationship between the number of historical fluctuations and the voltage recovery time.

[0031] Step S5-2: For each associated dataset, construct a voltage recovery time curve. Specifically, select the number of historical fluctuations as the horizontal axis and the voltage recovery time corresponding to the number of historical fluctuations as the vertical axis. Connect the data points in the order of the number of fluctuations to obtain the voltage recovery time curve.

[0032] Step S5-3: Input the historical fluctuation number and voltage recovery duration of each voltage recovery duration curve into the neural network model for training to obtain the voltage recovery duration prediction model; input the voltage fluctuation data of the current energy storage battery pack as the dependent variable into the voltage recovery duration prediction model to predict the voltage recovery duration of the next n+1 voltage fluctuation in the voltage recovery duration curve, which is denoted as the predicted voltage recovery duration.

[0033] Step S5-4: Monitor the voltage fluctuation data of the energy storage battery pack in real time. When a voltage fluctuation is detected, match the corresponding voltage recovery time curve according to the fluctuation cascade range and fluctuation type, and retrieve the predicted voltage recovery time of the curve; record the actual voltage recovery time of this fluctuation, and compare the actual voltage recovery time with the predicted voltage recovery time.

[0034] When the actual voltage recovery time exceeds the predicted voltage recovery time, it is determined that there is an abnormality in the monitoring results of the bus-multiplexed energy storage battery pack, and an early warning signal is generated and sent.

[0035] When the actual voltage recovery time does not exceed the predicted voltage recovery time, the monitoring result of the bus-multiplexed energy storage battery pack is determined to be normal, and the energy storage battery pack fault monitoring continues.

[0036] By grouping and sorting voltage fluctuation data and corresponding voltage recovery times according to the fluctuation cascade range and type, a correlation dataset of historical fluctuation counts and recovery times is obtained, and a voltage recovery time curve is constructed based on this. The curve data is input into a neural network model for training and prediction of the next recovery time. When fluctuations occur in real time, the corresponding curve is matched, and the predicted time is compared with the actual recovery time. If the predicted time exceeds the actual recovery time, an anomaly is identified and an alert is issued; if the predicted time does not exceed the actual recovery time, it is considered normal. This enables precise anomaly monitoring of different fluctuation data of bus-reused energy storage battery lines, providing a guarantee for the reliable operation of the lines.

[0037] Furthermore, a bus-based energy storage battery pack fault monitoring system includes a voltage data acquisition module, a voltage curve reference module, a fluctuation value ringing module, a fluctuation characteristic cascade module, a cascade monitoring and recovery module, and a recovery prediction and monitoring module.

[0038] The voltage data acquisition module is used to acquire voltage data information of each battery cell in the energy storage battery pack and construct an energy storage battery cell monitoring set based on this data; the voltage curve benchmark module is used to construct a real-time voltage fluctuation curve based on the energy storage battery cell monitoring set and calculate the real-time voltage of the battery cell; the fluctuation value ringing module is used to acquire the real-time voltage of the bus and analyze the voltage fluctuation value, and at the same time create a circuit ringing detection set to filter fluctuation data; the fluctuation feature cascading module is used to analyze the historical voltage monitoring data of the battery cell to obtain fluctuation characteristics and determine the fluctuation cascading range; the cascading monitoring and recovery module is used to monitor the circuit ringing detection set through the fluctuation cascading range and analyze the voltage recovery time; the recovery prediction monitoring module is used to construct a voltage recovery time curve and predict the recovery time, and at the same time perform abnormal monitoring of the bus-reused energy storage battery lines;

[0039] The output of the voltage data acquisition module is electrically connected to the input of the voltage curve reference module; the output of the voltage curve reference module is electrically connected to the input of the fluctuation value ringing module; the output of the fluctuation value ringing module is electrically connected to the input of the fluctuation characteristic cascade module; the output of the fluctuation characteristic cascade module is electrically connected to the input of the cascade monitoring and recovery module; and the output of the cascade monitoring and recovery module is electrically connected to the input of the recovery prediction monitoring module.

[0040] The voltage data acquisition module includes a voltage sensing unit and a monitoring set construction unit; the voltage sensing unit is used to collect voltage data information of a single battery cell in the energy storage battery pack; the monitoring set construction unit is used to divide the voltage data information according to a single battery cell and construct an energy storage battery cell monitoring set.

[0041] The voltage curve reference module includes a fluctuation curve generation unit and a real-time voltage calculation unit; the fluctuation curve generation unit is used to construct a real-time voltage fluctuation curve based on the voltage data of the battery cell; the real-time voltage calculation unit is used to calculate the voltage frequency of the real-time voltage fluctuation curve to obtain the real-time voltage of the battery cell.

[0042] The fluctuation value ringing module includes a voltage fluctuation value calculation unit and a ringing set filtering unit; the voltage fluctuation value calculation unit is used to calculate the voltage fluctuation value of the battery unit by combining the real-time bus voltage and the real-time battery unit voltage; the ringing set filtering unit is used to create a circuit ringing detection set and filter the voltage fluctuation data of the battery unit.

[0043] The fluctuation feature cascade module includes a historical feature extraction unit and a cascade range determination unit; the historical feature extraction unit is used to read historical voltage monitoring data of the battery unit and extract voltage fluctuation features; the cascade range determination unit is used to process the voltage fluctuation features to obtain the cascade range of voltage fluctuations.

[0044] The cascaded monitoring and recovery module includes a cascaded range monitoring unit and a recovery time calculation unit; the cascaded range monitoring unit is used to monitor the circuit ringing detection set through the fluctuation cascaded range; the recovery time calculation unit is used to obtain the time it takes for the voltage in the fluctuation cascaded range to recover to the real-time bus voltage based on voltage fluctuation data analysis;

[0045] The recovery prediction and monitoring module includes a recovery time prediction unit and a line anomaly determination unit. The recovery time prediction unit is used to construct a voltage recovery time curve and predict the next voltage recovery time through a neural network model. The line anomaly determination unit is used to compare the actual voltage recovery time with the predicted voltage recovery time to determine whether the energy storage battery line is abnormal.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] 1. This invention deploys voltage acquisition sensors and electrically connects them to the positive and negative terminals of individual battery cells. It records the unique battery cell identifier code and acquisition timestamp corresponding to each voltage value to form an original voltage dataset. Then, it sorts the dataset by acquisition timestamp and constructs an energy storage battery cell monitoring set using the unique identifier as the key. This avoids data confusion between cells and achieves precise division and orderly management of voltage data at the cell level. Compared with the problem of insufficient data regularity in existing technologies, this invention provides an accurate data foundation for subsequent monitoring and improves the reliability of data processing.

[0048] 2. This invention acquires real-time bus voltage through a bus voltage acquisition device, calculates battery cell voltage fluctuation values, creates a circuit ringing detection set with a unique identifier to filter fluctuation data, then analyzes historical data to extract fluctuation characteristics and standardizes them, and determines the fluctuation cascade range by combining physical topology. Compared with existing technologies that only focus on the fluctuation of a single cell, this invention can capture the fluctuation correlation between cells, accurately define the scope of influence, provide clear boundaries for the analysis of complex ringing phenomena, and improve the comprehensiveness of fluctuation analysis.

[0049] 3. This invention constructs voltage recovery time curves by grouping them according to the range and type of fluctuations, inputs them into a neural network model for training, and predicts the next recovery time. It compares the actual and predicted times in real time to determine the line status. Compared with existing technologies that rely on a single threshold detection, this invention can achieve accurate anomaly monitoring for different fluctuation data, provide timely warnings of faults, ensure the reliable operation of bus-reused energy storage battery packs, and improve the targeting of monitoring and system safety. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating a fault monitoring method for energy storage battery packs based on bus multiplexing according to the present invention.

[0051] Figure 2 This is a schematic diagram of the structure of a bus-based energy storage battery pack fault monitoring system according to the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Example 1: As Figure 1 As shown, the present invention provides a technical solution: a fault monitoring method for energy storage battery packs based on bus multiplexing. The fault monitoring method for energy storage battery packs includes the following steps:

[0054] Step S1: Obtain the operating voltage data of each battery cell in the energy storage battery pack. Divide the obtained operating voltage data into individual battery cells as the basic unit to construct an energy storage battery cell monitoring set. Analyze the data to obtain multiple real-time voltage fluctuation curves. Combine the voltage frequency to calculate the real-time voltage of each battery cell.

[0055] Step S1-1: Deploy voltage acquisition sensors to collect voltage data information of the energy storage battery pack. Specifically, the detection end of each sensor is electrically connected to the positive and negative terminals of a single battery cell in the energy storage battery pack to obtain the voltage data information of a single battery cell in the energy storage battery pack. Specifically, the original voltage dataset is formed by recording the unique battery cell identifier code and acquisition timestamp corresponding to each voltage value.

[0056] Step S1-2: Sort the voltage data set of each battery cell in time sequence, based on the order of the acquisition timestamps, to obtain the voltage time sequence of a single battery cell; extract the key information from the voltage time sequence of each battery cell, including the unique battery cell identifier code, the acquisition timestamp, and the voltage value corresponding to the acquisition timestamp. Select the unique battery cell identifier code as the key, and select the acquisition timestamp and the voltage value corresponding to the acquisition timestamp as the value to construct the energy storage battery cell monitoring set.

[0057] In practical implementation, it is essential to ensure that the electrical connection between each voltage acquisition sensor and the positive and negative terminals of a single battery cell is stable and reliable to avoid data distortion caused by poor contact. When recording the unique battery cell identification code, it must correspond one-to-one with the physical cell. The time sequence must be strictly based on the order of the acquisition timestamps to ensure the temporal continuity of the voltage time sequence data. When constructing the energy storage battery cell monitoring set, the key-value pair mapping must be accurate to provide a unique index for subsequent data extraction.

[0058] Steps S1-3: Sequentially read the key-value pair data corresponding to the unique battery unit identifier code in the energy storage battery unit monitoring set; for a single battery unit, select the acquisition timestamp as the horizontal axis coordinate and the voltage value corresponding to the acquisition timestamp as the vertical axis coordinate, and mark all data points in the two-dimensional coordinate system; connect adjacent data points with continuous line segments according to the order of acquisition timestamps to form the real-time voltage fluctuation curve of the battery unit; synchronously process all other battery units in the energy storage battery unit monitoring set to obtain the real-time voltage fluctuation curve of each battery unit;

[0059] Step S1-4: When the bus voltage meets the preset target voltage condition, for the real-time voltage fluctuation curve of each battery cell in the energy storage battery pack, the voltage value is filtered based on the frequency of occurrence. Specifically, the mode of the voltage value in each real-time voltage fluctuation curve is selected as the real-time voltage of the battery cell. The real-time voltage of the battery cell is represented as the reference voltage under the normal and stable operation of the energy storage battery pack, and satisfies the preset target voltage that the real-time voltage of the battery cell is equal to the bus voltage.

[0060] In practical implementation, the construction of real-time voltage fluctuation curves should ensure the uniformity of the data collection timestamps to improve the smoothness of the curves. The scale settings of the horizontal and vertical axes should be adapted to the data range to clearly present the fluctuation trend. The selection of the mode of voltage values ​​as the real-time voltage of the battery unit should be carried out when the bus voltage is stable at the preset target voltage to ensure that the voltage data at this time can reflect the normal working state of the battery unit and avoid the introduction of reference deviation due to bus voltage fluctuations.

[0061] Step S2: Obtain the real-time bus voltage of the energy storage battery pack and analyze the real-time voltage of each battery cell to obtain the voltage fluctuation value of each battery cell; and create a circuit ringing detection set to filter and store the voltage fluctuation data of each battery cell according to the voltage fluctuation range.

[0062] Step S2-1: Deploy the bus voltage acquisition device to acquire the real-time bus voltage of the multiplexed bus of the energy storage battery pack and record the acquisition timestamp corresponding to the real-time bus voltage; for each battery cell, extract its real-time battery cell voltage and corresponding acquisition timestamp, compare the voltage values ​​with frequencies less than the real-time voltage at the same acquisition timestamp with the real-time bus voltage, and calculate the absolute difference between the voltage values ​​and the real-time bus voltage as the voltage fluctuation value of the battery cell.

[0063] Step S2-2: Create an independent circuit ringing detection set for each battery cell, and associate the set with the corresponding unique battery cell identifier code; iterate through the voltage data of the battery cell and add data, specifically:

[0064] When the actual voltage collected in a single battery cell does not exceed the real-time voltage minus the voltage fluctuation value, it is judged as an undervoltage fluctuation and added to the circuit ringing detection set; when the voltage in a single battery cell exceeds the real-time voltage plus the voltage fluctuation value, it is judged as an overvoltage fluctuation and added to the circuit ringing detection set.

[0065] Step S2-3: Synchronously process the data addition operation of each battery unit, and complete the creation and data filtering of each circuit ringing detection set;

[0066] In practical implementation, the deployment of bus voltage acquisition devices must be synchronized with the battery unit voltage acquisition sensors to ensure the comparability of data under the same acquisition timestamp. When calculating voltage fluctuation values, the selected voltage values ​​with frequencies lower than the real-time voltage must be verified multiple times to avoid accidental data affecting the accuracy of fluctuation values. When creating circuit ringing detection sets, it is necessary to strictly distinguish the judgment conditions of undervoltage and overvoltage fluctuations to ensure accurate classification of different types of fluctuation data.

[0067] The energy storage battery pack uses a multiplexed bus as the core transmission carrier to construct a power and signal coordinated transmission architecture. The power line carrier communication module integrates data signals into the power line through carrier technology based on the power transmission channel of the multiplexed bus, realizing time-division coordination of power transmission and signal transmission. This process does not require a dedicated signal daisy chain, effectively simplifying system wiring. During the design phase, the compatibility of communication protocols, signal modulation and demodulation efficiency, and data synchronization verification mechanism need to be considered. The power line carrier communication module further ensures the stability and accuracy of data transmission by adopting advanced modulation and demodulation technology and reliable synchronization verification logic, ensuring that power transmission and signal interaction do not interfere with each other.

[0068] The active balancing BMS module installed in the energy storage battery pack is based on a system architecture supported by a multiplexed bus. It performs fine-grained management of specific types of battery packs. It integrates AFE chips, control units (MCU / ASIC), protection switches and multi-parameter detection components. Through transformer synchronous active balancing technology, it dynamically adjusts the energy state of individual battery cells under high-power transmission scenarios to avoid local overcharging and over-discharging. At the same time, it transmits key state data such as battery pack SOC to the intelligent control unit through the communication link associated with the multiplexed bus, forming a closed-loop operation mechanism of "bus transmission - communication interaction - battery management".

[0069] Step S3: Read and analyze the historical voltage monitoring data corresponding to each battery cell to obtain multiple different voltage fluctuation characteristics corresponding to each battery cell. Perform consistency characterization processing on the multiple voltage fluctuation characteristics obtained from the analysis to obtain the cascade range of voltage fluctuation impact, which is denoted as the fluctuation cascade range.

[0070] Step S3-1: Read the historical voltage monitoring data corresponding to each battery cell. The historical voltage monitoring data is associated with a unique battery cell identification code. The historical voltage monitoring data is classified according to undervoltage fluctuations and overvoltage fluctuations. Voltage fluctuation characteristics under the same fluctuation type are extracted. The voltage fluctuation characteristics include fluctuation duration, fluctuation peak value, deviation value from bus voltage, and fluctuation start time.

[0071] Step S3-2: Standardize the characteristics of the same fluctuation type, specifically: unify the calculation benchmark for the duration of fluctuation, from the start of the fluctuation to the moment when the fluctuation recovers to the real-time voltage of the battery cell; unify the calculation benchmark for the peak deviation of fluctuation, using the bus voltage as the benchmark, and calculate the difference between the peak value of the fluctuation and the bus voltage.

[0072] Step S3-3: Select battery cells with voltage fluctuations from historical voltage monitoring data as target battery cells and associate them with their unique identification codes; based on the physical connection topology of the energy storage battery pack, obtain the directly electrically connected adjacent cells of the target battery cell and their corresponding unique identification codes.

[0073] Step S3-4: Set a fluctuation correlation timing window. The window starts at the time when the fluctuation of the target battery cell begins and ends at the time when the voltage of the target battery cell recovers to the real-time voltage. Within the window, detect the historical voltage data of adjacent cells to find voltage fluctuations of the same type as those of the target battery cell. The same type means that the overvoltage fluctuation or undervoltage fluctuation is consistent and the peak deviation of the fluctuation after standardization belongs to the same range. Record the target battery cell and the adjacent cells that have the same type of fluctuation to obtain the fluctuation cascade range corresponding to the voltage at the time of fluctuation start under different voltage fluctuation types.

[0074] In practice, reading historical voltage monitoring data requires complete association with the unique battery cell identifier code to trace the source. Feature standardization requires unifying the calculation benchmarks of various features to ensure that the features of different fluctuation events are comparable. The determination of adjacent cells of the target battery cell must strictly follow the physical connection topology of the energy storage battery pack to avoid including non-electrically connected cells. The setting of the fluctuation association timing window must cover the complete fluctuation cycle to ensure the comprehensive detection of the same type of fluctuation in adjacent cells.

[0075] Step S4: Monitor the circuit ringing detection set through the fluctuation cascade range, and analyze the circuit ringing detection set based on voltage fluctuation data. The time required for the voltage to recover to the real-time bus voltage in the fluctuation cascade range under different voltage fluctuation data is recorded as the voltage recovery time.

[0076] Real-time acquisition of voltage data for each battery cell; when undervoltage or overvoltage fluctuations are detected in a single battery cell, the start time and corresponding voltage value of the fluctuation are recorded. Based on the type of the fluctuation and the standardized peak deviation of the fluctuation, the fluctuation cascade range under the same fluctuation type and the same peak deviation range in step S3-4 is matched to determine the fluctuation cascade range corresponding to the voltage fluctuation.

[0077] Call the historical data that matches the current fluctuation type and fluctuation cascade range from the historical voltage monitoring data, and extract the time data of the voltage within the fluctuation cascade range recovering from the fluctuation state to the real-time bus voltage from the historical data; take the start time of the current fluctuation as the timing start point, monitor the voltage data of each battery cell within the fluctuation cascade range in real time until the voltage of all battery cells recovers to the real-time bus voltage, and record the time from the timing start point to the recovery time, which is recorded as the voltage recovery time corresponding to the voltage fluctuation data;

[0078] In practical implementation, real-time voltage data acquisition must ensure that the sampling frequency can capture fluctuation details. When fluctuations are detected, the start time and voltage value must be recorded immediately to ensure timeliness. When matching the fluctuation cascade range, the fluctuation type and peak deviation range must be strictly checked. When recalling historical records, cases that are completely consistent with the current conditions must be selected. The determination of the timing start point and recovery time must be based on the standard that all unit voltages have recovered to the real-time bus voltage to avoid recording too early or too late, which would cause duration deviation.

[0079] Step S5: Based on the voltage fluctuation data and voltage recovery time of the energy storage battery pack, construct multiple voltage recovery time curves based on the same voltage fluctuation data; use a neural network model to process and predict each voltage recovery time curve to obtain the next voltage recovery time of each voltage recovery time curve, which is recorded as the predicted voltage recovery time. Use the predicted voltage recovery time to monitor the abnormal voltage fluctuation data of the bus-reused energy storage battery lines.

[0080] Step S5-1: Divide the voltage fluctuation data and corresponding voltage recovery time into several groups according to the fluctuation cascade range and fluctuation type. Each group corresponds to a unique fluctuation cascade range identifier and fluctuation type label. Sort the voltage fluctuation data in each group according to the order of occurrence time and determine them as the first historical fluctuation, the second historical fluctuation, ..., the nth historical fluctuation in sequence. Record the voltage recovery time corresponding to each fluctuation to obtain the associated dataset of the relationship between the number of historical fluctuations and the voltage recovery time.

[0081] Step S5-2: For each associated dataset, construct a voltage recovery time curve. Specifically, select the number of historical fluctuations as the horizontal axis and the voltage recovery time corresponding to the number of historical fluctuations as the vertical axis. Connect the data points in the order of the number of fluctuations to obtain the voltage recovery time curve.

[0082] Step S5-3: Input the historical fluctuation number and voltage recovery duration of each voltage recovery duration curve into the neural network model for training to obtain the voltage recovery duration prediction model; input the voltage fluctuation data of the current energy storage battery pack as the dependent variable into the voltage recovery duration prediction model to predict the voltage recovery duration of the next n+1 voltage fluctuation in the voltage recovery duration curve, which is denoted as the predicted voltage recovery duration.

[0083] Step S5-4: Monitor the voltage fluctuation data of the energy storage battery pack in real time. When a voltage fluctuation is detected, match the corresponding voltage recovery time curve according to the fluctuation cascade range and fluctuation type, and retrieve the predicted voltage recovery time of the curve; record the actual voltage recovery time of this fluctuation, and compare the actual voltage recovery time with the predicted voltage recovery time.

[0084] When the actual voltage recovery time exceeds the predicted voltage recovery time, it is determined that there is an abnormality in the monitoring results of the bus-multiplexed energy storage battery pack, and an early warning signal is generated and sent.

[0085] When the actual voltage recovery time does not exceed the predicted voltage recovery time, the monitoring result of the bus-multiplexed energy storage battery pack is determined to be normal, and the energy storage battery pack fault monitoring continues.

[0086] In practical implementation, data grouping must ensure that the fluctuation cascade range and fluctuation type within the same group are completely consistent. The voltage recovery time curve must be constructed by accurately connecting the lines in the order of the number of fluctuations to reflect the trend. The neural network model training must use sufficient historical data to improve prediction accuracy. When comparing the actual and predicted recovery times, a reasonable judgment threshold must be set to ensure that the abnormal warning does not miss potential faults or generate false alarms, thus ensuring the reliability of line detection.

[0087] Example 2, as Figure 2 As shown, the present invention provides a bus-based energy storage battery pack fault monitoring system, which includes a voltage data acquisition module, a voltage curve reference module, a fluctuation value ringing module, a fluctuation characteristic cascade module, a cascade monitoring and recovery module, and a recovery prediction and monitoring module.

[0088] The voltage data acquisition module is used to acquire voltage data information of each battery cell in the energy storage battery pack and construct an energy storage battery cell monitoring set based on this data; the voltage curve benchmark module is used to construct a real-time voltage fluctuation curve based on the energy storage battery cell monitoring set and calculate the real-time voltage of the battery cell; the fluctuation value ringing module is used to acquire the real-time voltage of the bus and analyze the voltage fluctuation value, and at the same time create a circuit ringing detection set to filter fluctuation data; the fluctuation feature cascading module is used to analyze the historical voltage monitoring data of the battery cell to obtain fluctuation characteristics and determine the fluctuation cascading range; the cascading monitoring and recovery module is used to monitor the circuit ringing detection set through the fluctuation cascading range and analyze the voltage recovery time; the recovery prediction monitoring module is used to construct a voltage recovery time curve and predict the recovery time, and at the same time perform abnormal monitoring of the bus-reused energy storage battery lines;

[0089] The output of the voltage data acquisition module is electrically connected to the input of the voltage curve reference module; the output of the voltage curve reference module is electrically connected to the input of the fluctuation value ringing module; the output of the fluctuation value ringing module is electrically connected to the input of the fluctuation characteristic cascade module; the output of the fluctuation characteristic cascade module is electrically connected to the input of the cascade monitoring and recovery module; and the output of the cascade monitoring and recovery module is electrically connected to the input of the recovery prediction monitoring module.

[0090] The voltage data acquisition module includes a voltage sensing unit and a monitoring set construction unit; the voltage sensing unit is used to collect voltage data information of a single battery cell in the energy storage battery pack; the monitoring set construction unit is used to divide the voltage data information according to a single battery cell and construct an energy storage battery cell monitoring set.

[0091] The voltage curve reference module includes a fluctuation curve generation unit and a real-time voltage calculation unit; the fluctuation curve generation unit is used to construct a real-time voltage fluctuation curve based on the voltage data of the battery cell; the real-time voltage calculation unit is used to calculate the voltage frequency of the real-time voltage fluctuation curve to obtain the real-time voltage of the battery cell.

[0092] The fluctuation value ringing module includes a voltage fluctuation value calculation unit and a ringing set filtering unit; the voltage fluctuation value calculation unit is used to calculate the voltage fluctuation value of the battery unit by combining the real-time bus voltage and the real-time battery unit voltage; the ringing set filtering unit is used to create a circuit ringing detection set and filter the voltage fluctuation data of the battery unit.

[0093] The fluctuation feature cascade module includes a historical feature extraction unit and a cascade range determination unit; the historical feature extraction unit is used to read historical voltage monitoring data of the battery unit and extract voltage fluctuation features; the cascade range determination unit is used to process the voltage fluctuation features to obtain the cascade range of voltage fluctuations.

[0094] The cascaded monitoring and recovery module includes a cascaded range monitoring unit and a recovery time calculation unit; the cascaded range monitoring unit is used to monitor the circuit ringing detection set through the fluctuation cascaded range; the recovery time calculation unit is used to obtain the time it takes for the voltage in the fluctuation cascaded range to recover to the real-time bus voltage based on voltage fluctuation data analysis;

[0095] The recovery prediction and monitoring module includes a recovery time prediction unit and a line anomaly determination unit. The recovery time prediction unit is used to construct a voltage recovery time curve and predict the next voltage recovery time through a neural network model. The line anomaly determination unit is used to compare the actual voltage recovery time with the predicted voltage recovery time to determine whether the energy storage battery line is abnormal.

[0096] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

Claims

1. A fault monitoring method for energy storage battery packs based on bus multiplexing, characterized in that: The energy storage battery pack fault monitoring method is based on a bus multiplexing structure for power transmission lines and signal transmission lines, and includes the following steps: Step S1: Obtain the operating voltage data of each battery cell in the energy storage battery pack. Divide the obtained operating voltage data into individual battery cells as the basic unit to construct an energy storage battery cell monitoring set. Analyze the data to obtain multiple real-time voltage fluctuation curves. Combine the voltage frequency to calculate the real-time voltage of each battery cell. Step S2: Obtain the real-time bus voltage of the energy storage battery pack and analyze the real-time voltage of each battery cell to obtain the voltage fluctuation value of each battery cell; and create a circuit ringing detection set to filter and store the voltage fluctuation data of each battery cell according to the voltage fluctuation range. Step S3: Read and analyze the historical voltage monitoring data corresponding to each battery cell to obtain multiple different voltage fluctuation characteristics corresponding to each battery cell. Perform consistency characterization processing on the multiple voltage fluctuation characteristics obtained from the analysis to obtain the cascade range of voltage fluctuation impact, which is denoted as the fluctuation cascade range. Step S4: Monitor the circuit ringing detection set through the fluctuation cascade range, and analyze the circuit ringing detection set based on voltage fluctuation data. The time required for the voltage to recover to the real-time bus voltage in the fluctuation cascade range under different voltage fluctuation data is recorded as the voltage recovery time. Step S5: Based on the voltage fluctuation data and voltage recovery time of the energy storage battery pack, construct multiple voltage recovery time curves based on the same voltage fluctuation data; use a neural network model to process and predict each voltage recovery time curve to obtain the next voltage recovery time of each voltage recovery time curve, which is recorded as the predicted voltage recovery time. Use the predicted voltage recovery time to monitor the abnormal voltage fluctuation data of the bus-reused energy storage battery lines.

2. The method for fault monitoring of energy storage battery packs based on bus multiplexing according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1: Deploy voltage acquisition sensors to collect voltage data information of the energy storage battery pack. Specifically, the detection end of each sensor is electrically connected to the positive and negative terminals of a single battery cell in the energy storage battery pack to obtain the voltage data information of a single battery cell in the energy storage battery pack. Specifically, the original voltage dataset is formed by recording the unique battery cell identifier code and acquisition timestamp corresponding to each voltage value. Step S1-2: Sort the voltage data set for each battery cell in a time sequence, based on the order of the acquisition timestamps, to obtain the voltage time sequence of a single battery cell; extract the key information from the voltage time sequence of each battery cell, including the unique battery cell identifier code, the acquisition timestamp, and the voltage value corresponding to the acquisition timestamp. Select the unique battery cell identifier code as the key, and select the acquisition timestamp and the voltage value corresponding to the acquisition timestamp as the value to construct the energy storage battery cell monitoring set.

3. The method for fault monitoring of energy storage battery packs based on bus multiplexing according to claim 2, characterized in that: Step S1 also includes: Steps S1-3: Sequentially read the key-value pair data corresponding to the unique battery unit identifier code in the energy storage battery unit monitoring set; for a single battery unit, select the acquisition timestamp as the horizontal axis coordinate and the voltage value corresponding to the acquisition timestamp as the vertical axis coordinate, and mark all data points in the two-dimensional coordinate system; connect adjacent data points with continuous line segments according to the order of acquisition timestamps to form the real-time voltage fluctuation curve of the battery unit; synchronously process all other battery units in the energy storage battery unit monitoring set to obtain the real-time voltage fluctuation curve of each battery unit; Step S1-4: When the bus voltage meets the preset target voltage condition, for the real-time voltage fluctuation curve of each battery cell in the energy storage battery pack, the voltage value is filtered based on the frequency of occurrence. Specifically, the mode of the voltage value in each real-time voltage fluctuation curve is selected as the real-time voltage of the battery cell. The real-time voltage of the battery cell is represented as the reference voltage under the normal and stable operation state of the energy storage battery pack, and satisfies the preset target voltage that the real-time voltage of the battery cell is equal to the bus voltage.

4. The method for fault monitoring of energy storage battery packs based on bus multiplexing according to claim 3, characterized in that: The specific steps of step S2 are as follows: Step S2-1: Deploy the bus voltage acquisition device to acquire the real-time bus voltage of the multiplexed bus of the energy storage battery pack and record the acquisition timestamp corresponding to the real-time bus voltage; for each battery cell, extract its real-time battery cell voltage and corresponding acquisition timestamp, compare the voltage values ​​with frequencies less than the real-time voltage at the same acquisition timestamp with the real-time bus voltage, and calculate the absolute difference between the voltage values ​​and the real-time bus voltage as the voltage fluctuation value of the battery cell. Step S2-2: Create an independent circuit ringing detection set for each battery cell, and associate the set with the corresponding unique battery cell identifier code; iterate through the voltage data of the battery cell and add data, specifically as follows: When the actual voltage collected in a single battery cell does not exceed the real-time voltage minus the voltage fluctuation value, it is judged as an undervoltage fluctuation and added to the circuit ringing detection set; when the voltage in a single battery cell exceeds the real-time voltage plus the voltage fluctuation value, it is judged as an overvoltage fluctuation and added to the circuit ringing detection set. Step S2-3: Synchronously process the data addition operations of each battery unit to complete the creation and data filtering of each circuit ringing detection set.

5. The method for fault monitoring of energy storage battery packs based on bus multiplexing according to claim 4, characterized in that: The specific steps of step S3 are as follows: Step S3-1: Read the historical voltage monitoring data corresponding to each battery cell. The historical voltage monitoring data is associated with a unique battery cell identification code. The historical voltage monitoring data is classified according to undervoltage fluctuations and overvoltage fluctuations. Voltage fluctuation characteristics under the same fluctuation type are extracted. The voltage fluctuation characteristics include fluctuation duration, fluctuation peak value, deviation value from bus voltage, and fluctuation start time. Step S3-2: Standardize the characteristics of the same fluctuation type, specifically: unify the calculation benchmark for the duration of fluctuation, from the start of the fluctuation to the moment when the fluctuation recovers to the real-time voltage of the battery cell; A unified benchmark for calculating peak fluctuation deviation is established, using the bus voltage as the reference, to calculate the difference between the peak fluctuation and the bus voltage. Step S3-3: Select battery cells with voltage fluctuations in historical voltage monitoring data as target battery cells and associate them with their unique identification codes; based on the physical connection topology of the energy storage battery pack, obtain the directly electrically connected adjacent cells of the target battery cell and their corresponding unique identification codes. Step S3-4: Set a fluctuation correlation timing window. The window starts at the time when the fluctuation of the target battery cell begins and ends at the time when the voltage of the target battery cell recovers to the real-time voltage. Within the window, detect the historical voltage data of adjacent cells to find voltage fluctuations of the same type as those of the target battery cell. The same type means that the overvoltage fluctuation or undervoltage fluctuation is consistent and the peak deviation of the standardized fluctuation belongs to the same range. Record the target battery cell and adjacent cells that exhibit the same type of fluctuation to obtain the cascade range of fluctuations corresponding to the voltage at the start of the fluctuation under different voltage fluctuation types.

6. The method for fault monitoring of energy storage battery packs based on bus multiplexing according to claim 5, characterized in that: In step S4, the voltage data of each battery cell is collected in real time; when an undervoltage fluctuation or overvoltage fluctuation is detected in a single battery cell, the start time of the fluctuation and the corresponding voltage value are recorded. Based on the type of the fluctuation and the standardized fluctuation peak deviation, the fluctuation cascade range under the same fluctuation type and the same peak deviation range in steps S3-4 is matched to determine the fluctuation cascade range corresponding to the voltage fluctuation. Call the historical records in the historical voltage monitoring data that are consistent with the current fluctuation type and fluctuation cascade range, and extract the time data of the voltage within the fluctuation cascade range recovering from the fluctuation state to the real-time bus voltage. Using the start time of the current fluctuation as the starting point of the timing, the voltage data of each battery cell within the fluctuation cascade range is monitored in real time until the voltage of all battery cells is restored to the real-time bus voltage. The time from the starting point of the timing to the recovery time is recorded as the voltage recovery time corresponding to the voltage fluctuation data.

7. The method for fault monitoring of energy storage battery packs based on bus multiplexing according to claim 6, characterized in that: The specific steps of step S5 are as follows: Step S5-1: Divide the voltage fluctuation data and corresponding voltage recovery time into several groups according to the fluctuation cascade range and fluctuation type. Each group corresponds to a unique fluctuation cascade range identifier and fluctuation type label. Sort the voltage fluctuation data in each group according to the order of occurrence time and determine them as the first historical fluctuation, the second historical fluctuation, ..., the nth historical fluctuation in sequence. Record the voltage recovery time corresponding to each fluctuation to obtain the associated dataset of the relationship between the number of historical fluctuations and the voltage recovery time. Step S5-2: For each associated dataset, construct a voltage recovery time curve. Specifically, select the number of historical fluctuations as the horizontal axis and the voltage recovery time corresponding to the number of historical fluctuations as the vertical axis. Connect the data points in the order of the number of fluctuations to obtain the voltage recovery time curve. Step S5-3: Input the historical fluctuation number and voltage recovery duration of each voltage recovery duration curve into the neural network model for training to obtain the voltage recovery duration prediction model; input the voltage fluctuation data of the current energy storage battery pack as the dependent variable into the voltage recovery duration prediction model to predict the voltage recovery duration of the next n+1 voltage fluctuation in the voltage recovery duration curve, which is denoted as the predicted voltage recovery duration. Step S5-4: Monitor the voltage fluctuation data of the energy storage battery pack in real time. When a voltage fluctuation is detected, match the corresponding voltage recovery time curve according to the fluctuation cascade range and fluctuation type, and retrieve the predicted voltage recovery time of the curve; record the actual voltage recovery time of this fluctuation, and compare the actual voltage recovery time with the predicted voltage recovery time. When the actual voltage recovery time exceeds the predicted voltage recovery time, it is determined that there is an abnormality in the monitoring results of the bus-multiplexed energy storage battery pack, and an early warning signal is generated and sent. When the actual voltage recovery time does not exceed the predicted voltage recovery time, the monitoring result of the bus-multiplexed energy storage battery pack is determined to be normal, and the energy storage battery pack fault monitoring continues.

8. A bus-multiplexed energy storage battery pack fault monitoring system, applied to the bus-multiplexed energy storage battery pack fault monitoring method according to any one of claims 1-7, characterized in that: The energy storage battery pack fault monitoring system includes a voltage data acquisition module, a voltage curve reference module, a fluctuation value ringing module, a fluctuation characteristic cascade module, a cascade monitoring and recovery module, and a recovery prediction and monitoring module. The voltage data acquisition module is used to acquire voltage data information of each battery cell in the energy storage battery pack and construct an energy storage battery cell monitoring set based on the data; the voltage curve benchmark module is used to construct a real-time voltage fluctuation curve based on the energy storage battery cell monitoring set and calculate the real-time voltage of the battery cell; the fluctuation value ringing module is used to acquire the real-time voltage of the bus and analyze the voltage fluctuation value, and at the same time create a circuit ringing detection set to filter fluctuation data; the fluctuation feature cascading module is used to analyze the historical voltage monitoring data of the battery cell to obtain fluctuation characteristics and determine the fluctuation cascading range; the cascading monitoring and recovery module is used to monitor the circuit ringing detection set through the fluctuation cascading range and analyze the voltage recovery time; the recovery prediction monitoring module is used to construct a voltage recovery time curve and predict the recovery time, and at the same time perform abnormal monitoring of the bus-reused energy storage battery lines.

9. A fault monitoring system for energy storage battery packs based on bus multiplexing according to claim 8, characterized in that: The voltage data acquisition module includes a voltage sensing unit and a monitoring set construction unit; the voltage sensing unit is used to collect voltage data information of a single battery cell in the energy storage battery pack; the monitoring set construction unit is used to divide the voltage data information according to a single battery cell and construct an energy storage battery cell monitoring set. The voltage curve reference module includes a fluctuation curve generation unit and a real-time voltage calculation unit; the fluctuation curve generation unit is used to construct a real-time voltage fluctuation curve based on the voltage data of the battery cell. The real-time voltage calculation unit is used to calculate the voltage frequency of the real-time voltage fluctuation curve to obtain the real-time voltage of the battery cell. The fluctuation value ringing module includes a voltage fluctuation value calculation unit and a ringing set filtering unit; the voltage fluctuation value calculation unit is used to calculate the voltage fluctuation value of the battery unit by combining the real-time bus voltage and the real-time battery unit voltage. The ringing set filtering unit is used to create a circuit ringing detection set and filter the voltage fluctuation data of the battery cells.

10. A fault monitoring system for energy storage battery packs based on bus multiplexing according to claim 8, characterized in that: The fluctuation feature cascade module includes a historical feature extraction unit and a cascade range determination unit; the historical feature extraction unit is used to read historical voltage monitoring data of the battery unit and extract voltage fluctuation features; the cascade range determination unit is used to process the voltage fluctuation features to obtain the cascade range of voltage fluctuations. The cascaded monitoring and recovery module includes a cascaded range monitoring unit and a recovery time calculation unit; the cascaded range monitoring unit is used to monitor the circuit ringing detection set through the fluctuation cascaded range; the recovery time calculation unit is used to obtain the time it takes for the voltage in the fluctuation cascaded range to recover to the real-time bus voltage based on voltage fluctuation data analysis; The recovery prediction and monitoring module includes a recovery time prediction unit and a line anomaly determination unit; The recovery time prediction unit is used to construct a voltage recovery time curve and predict the next voltage recovery time through a neural network model; the line anomaly determination unit is used to compare the actual voltage recovery time with the predicted voltage recovery time to determine whether the energy storage battery line is abnormal.

Images