Battery damage intelligent identification monitoring system and method for battery operation and maintenance management

Abstract

The invention discloses a battery damage intelligent identification monitoring system and method for battery operation and maintenance management, and relates to the technical field of data analysis and intelligent monitoring, and the battery damage intelligent identification monitoring method specifically comprises the following steps: arranging an array type electromagnetic sensor and an acoustic sensor; historical sensor data are collected, and key features causing battery damage are extracted; calculating a first judgment threshold value for judging whether the battery is damaged or not according to the key features; performing correlation analysis on the data records and the key features to obtain effective parameters, and calculating a second effective threshold value; whether the battery is damaged or not is judged according to the first judgment threshold value, and real-time monitoring and damage early warning are carried out; and judging whether the damage early warning is effective or not according to the second effective threshold value. Real-time monitoring and effective parameter analysis enable the operation and maintenance management of the battery to be more intelligent, and the method has high practicability and operation convenience.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis and intelligent monitoring, and specifically to a battery damage intelligent identification and monitoring system and method for battery operation and maintenance management. Background Art

[0002] The technical field of data analysis and intelligent monitoring mainly involves using modern computing technologies, data processing methods, and sensor technologies for real-time monitoring, anomaly detection, fault diagnosis, and predictive analysis applications. The development of this field benefits from the rapid progress of big data, artificial intelligence (AI), and sensor technologies, and is widely applied in various industries, including intelligent manufacturing, battery management, etc.

[0003] In the operation and maintenance management of batteries, the monitoring and identification of battery damage are crucial for ensuring battery safety and extending its service life. Currently, battery monitoring technologies mainly rely on traditional battery parameters such as temperature, humidity, voltage, and current to evaluate the health status of batteries. However, traditional methods have certain limitations and are difficult to detect potential risks of battery damage at an early stage, especially when problems such as micro-cracks, leakage, or thermal runaway occur inside the battery. In addition, battery monitoring means are relatively single, lacking intelligent identification technologies that comprehensively utilize electromagnetic and acoustic sensor data, and it is impossible to further evaluate the effectiveness of the identification, resulting in increased maintenance costs. Summary of the Invention

[0004] The purpose of the present invention is to provide a battery damage intelligent identification and monitoring system and method for battery operation and maintenance management to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A battery damage intelligent identification and monitoring method for battery operation and maintenance management, and the battery damage intelligent identification and monitoring method specifically includes the following steps: Arrange an array of electromagnetic sensors and acoustic sensors in the battery module area; Collect sensor data when the battery has different degrees of damage in history, analyze the collected sensor data, and extract key features that cause battery damage; According to the key features, calculate a first judgment threshold for judging whether the battery is damaged; Obtain the data records of battery damage in history, conduct a correlation analysis on the data records and the key features, obtain effective parameters associated with the key features, and calculate a second effective threshold; Real-time obtain the battery sensor data during operation, judge whether the battery is damaged according to the first judgment threshold, and conduct real-time monitoring and damage warning; Calculate the effectiveness of the damage warning, and judge whether the damage warning is effective according to the second effective threshold.

[0006] Arrange array electromagnetic sensors and acoustic sensors in the battery module area, specifically: Select different array electromagnetic sensors according to the size of different battery internal environments, and integrate the battery module with the array electromagnetic sensors; The acoustic sensor includes an accelerometer and a piezoelectric sensor; A permalloy shielding cover is arranged outside the array electromagnetic sensor for electromagnetic screen signal protection; Integrate a vibration baseline calibration device outside the acoustic sensor; Among them, the array electromagnetic sensors include fluxgate sensors, Hall effect sensors, etc.; and ensure that they can cover the key areas of the battery, including battery cells, cooling systems, etc.

[0007] Collect the sensor data when the battery has different degrees of damage in the collection history, analyze the collected sensor data, and extract the key features that cause the battery to be damaged, specifically: Step S3-1: Collect the sensor data when the battery has different degrees of damage in the history, specifically: {a 1 、a 2 、…、a n}; where a 1 、a 2 、…、a n represent the 1st, 2nd,..., nth sensor data values when the battery is damaged in the collected history, n represents the type label of the sensor data, and n is a positive integer; collect the sensor data when the battery is operating normally in the history, specifically: {b 1 、b 2 、…、b n}; where b 1 、b 2 、…、b n represent the 1st, 2nd,..., nth sensor data values when the battery is operating normally in the collected history; perform N collections; Step S3-2: Based on the sensor data when the battery has different degrees of damage and the sensor data when the battery is operating normally in the N collected histories, calculate the average change rate of each sensor data; Specifically: ; where p(i) represents the average change rate of the i-th sensor data; a represents the sensor data value when the battery is damaged in the history; b represents the sensor data value when the battery is operating normally in the history; N represents the number of collections of sensor data; N is a positive integer; Step S3-3: Obtain the average change rate [p(1), p(2), …, p(n)] of each type of sensor data; where p(1), p(2), …, p(n) represent the calculated average change rate of the 1st, 2nd, …, nth type of sensor data. Conduct a comparative analysis on the average change rate of the n types of sensor data, and determine the sensor data with an average change rate greater than the median as key features, and obtain the key features of different batteries when they are damaged, specifically: [Q1, Q2, …, Q k ; where Q1, Q2, …, Q k represent the obtained 1st, 2nd, …, kth key features; k represents the category label of the key features, k is a positive integer, k ∈ (1, n).

[0008] According to the key features, calculate the first judgment threshold for determining whether the battery is damaged, specifically: Step S4-1: Obtain the key feature values {[Q 11 , Q 12 , …, Q 1M , [Q 21 , Q 22 , …, Q 2M , …, [Q k1 , Q k2 , …, Q kM} when the battery is damaged in M histories; where Q kM represents the key feature value of the kth key feature collected for the Mth time when the battery is damaged in history; M represents the number of times of collecting key feature values, M is a positive integer; Step S4-2: Calculate the first judgment threshold through the obtained key feature values when the battery is damaged in the M histories. The first judgment threshold is obtained based on the numerical difference between the key feature values when the battery is damaged and the key feature values when the battery is operating normally; Specifically: ; where s is the first judgment threshold, representing the numerical difference between the key feature values when the battery is damaged and the key feature values when the battery is operating normally; u and v represent the identifiers of the key feature values, u and v are positive integers, u ∈ [1, k], v ∈ [1, M]; Q uv represents the vth key feature value of the uth key feature when the battery is damaged in history; Q u ’ represents the average value of the uth key feature when the battery is operating normally in history.

[0009] Obtain the data records of the battery being damaged in history, conduct a correlation analysis on the data records and the key features, and obtain the effective parameters associated with the key features, specifically: Step S5-1: Obtain the data records of battery breakage in history, collect the influencing parameters when the key characteristic values change by more than three standard deviations during the battery operation process, and establish an association relationship model between different influencing parameters and key characteristics with the influencing parameters as independent variables and the key characteristics as dependent variables; Among them, the number of collected influencing parameter values is the same as that of the key characteristic values; Establish different association relationship models according to different influencing parameters; Among them, establishing the association relationship model specifically is: ; Among them, Q k ’ represents the predicted value of the key characteristic; G represents the influencing parameter value; α represents the slope of the association relationship model; φ represents the intercept of the association relationship model; Step S5-2: Calculate the model parameters according to the association relationship models between different influencing parameters and key characteristics; draw a scatter plot and analyze the scatter plot; Among them, the specific formulas for calculating the slope and intercept of the association relationship model are: ; Among them, α represents the slope of the association relationship model; z represents the number of collected influencing parameter values and key characteristic values; G represents the collected influencing parameter values; represents the average value of the influencing parameters; Q k represents the collected key characteristic values; represents the average value of the key characteristic values; ; Calculate the parameters of the association relationship models between different influencing parameters and key characteristics in turn; Step S5-3: According to the association relationship models of different influencing parameters, obtain several groups of predicted values of key characteristics, calculate the difference values between the several groups of predicted values of key characteristics and the subsequent actual measured values, and determine the influencing parameters greater than the average difference value as effective parameters; The process of obtaining several groups of predicted values of key characteristics according to the association relationship models of different influencing parameters, calculating the difference values between the several groups of predicted values of key characteristics and the subsequent actual measured values, and determining the influencing parameters greater than the average difference value as effective parameters is specifically: ; Among them, f represents the difference value between the predicted value of the key characteristic and the subsequent actual measured value in the association relationship model of different influencing parameters; Among them, it is considered that the effectiveness of the influencing parameters less than or equal to the average difference value and the key characteristics is insufficient; Determine the influencing parameters greater than the average difference value as effective parameters, and the set of effective parameters is denoted as {g1, g2,..., gr}; where, g1, g2, …, g r respectively represent the 1st, 2nd, …, rth effective parameters; r represents the category label of the effective parameters, and r is a positive integer.

[0010] The calculation of the second effective threshold is specifically as follows: Taking the determined effective parameters as a benchmark, obtaining the historical data change rates and standard deviations of different effective parameters, and calculating the second effective threshold; ; where, s’ represents the second effective threshold; μ g represents the change rate of different effective parameters; σ g represents the standard deviation of different effective parameters.

[0011] The real-time acquisition of the battery sensor data during the operation process, and judging whether the battery is damaged according to the first judgment threshold for real-time monitoring and damage warning, specifically as follows: Setting up a warning device; Obtaining key feature values through real-time monitoring. When the numerical difference of the key feature values obtained by real-time monitoring is greater than the first judgment threshold, giving a warning through the warning device; where, specifically: When is satisfied, giving a warning through the warning device; where, through represents the numerical difference of the key feature values, represents the key feature value collected last time of represents the key feature value obtained by the current monitoring.

[0012] Calculating the effectiveness of the damage warning, and judging whether the damage warning is effective according to the second effective threshold, specifically as follows: Step S5-1: After receiving the warning signal of the warning device, establishing a correlation relationship model between different influencing parameters and key features according to the method described in step S5-1, calculating the effective parameters in the actual situation through the key features, and obtaining the effective parameter values through real-time monitoring; Step S5-2: According to the effective parameter values obtained through real-time monitoring, when the change rate of the effective parameter values obtained through real-time monitoring is greater than the second effective threshold, judging that the damage warning is effective, and arranging staff to check and repair the battery; where, specifically: When is satisfied, judging that the damage warning is effective, and arranging staff to check and repair the battery.

[0013] Specifically: Set a timestamp; The timestamp is used to judge the continuity of the collected data; The timestamp is used to mark the time of the obtained data each time the data is obtained, preventing time dislocation and miscalculating the numerical difference of the key feature value and the change rate of the effective parameter value.

[0014] A battery breakage intelligent identification and monitoring system for battery operation and maintenance management, the battery breakage intelligent identification and monitoring system includes a data acquisition and transmission module, a feature extraction module, an analysis and calculation module, a real-time monitoring and warning module, an effective evaluation module, and a user interaction module; The data acquisition and transmission module is used to collect signals from sensors in the battery module in real time, transmit the data to the processing end by wireless and wired methods, and provide data storage and backup functions; The feature extraction module is used to extract key features from the collected data, calculate the change features of the signal by analyzing the change of the battery health state, identify the fault mode, and provide necessary feature information for subsequent analysis; The analysis and calculation module is used to deeply analyze the extracted key features and calculate the first judgment threshold for judging battery breakage; By establishing a correlation relationship model, obtain effective parameters, and calculate the second effective threshold based on historical data; The real-time monitoring and warning module is used to continuously monitor the battery state according to the real-time data and the set threshold, automatically identify whether the battery is damaged, and trigger a warning device to give a breakage warning when the preset first judgment threshold is exceeded; The effective evaluation module is used to evaluate the effectiveness of the breakage warning, and confirm the effectiveness of the breakage warning by comparing and analyzing the real-time monitoring data and the warning signal; The user interaction module is used to provide a user interface to display real-time monitoring data, warning information, and system status, enabling staff to intuitively view the battery state, adjust parameters, and receive warnings, and providing functions for viewing historical data and generating relevant reports.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: By integrating electromagnetic sensors and acoustic sensors, based on the principle that the inevitable internal damage of the battery will lead to changes in the electromagnetic field distribution and mechanical vibration characteristics, a multi-dimensional monitoring method is adopted to comprehensively capture the battery health status from aspects such as the internal electromagnetic field change and acoustic signal of the battery, enhancing the accuracy of damage identification. Through intelligent analysis of historical sensor data, key features are extracted, combined with dynamically adjusted judgment thresholds and optimized prediction models, effectively improving the sensitivity and real-time response ability of damage detection. Especially when problems such as micro-cracks and liquid leakage occur in the battery, early warning and timely intervention can be achieved, reducing potential safety hazards caused by battery failures. In addition, the real-time monitoring and effective parameter analysis of the present invention make the battery operation and maintenance management more intelligent, with high practicability and simplicity of operation, meeting the requirements of different battery types and working environments, and can be applied to actual battery safety monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic structural diagram of the intelligent damage identification and monitoring system and method for battery operation and maintenance management of the present invention; Figure 2 It is a scatter plot of the correlation relationship model in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution, an intelligent damage identification and monitoring method for battery operation and maintenance management. As Figure 1 shown, the intelligent damage identification and monitoring method for the battery specifically includes the following steps: Arrange array-type electromagnetic sensors and acoustic sensors in the battery module area; Collect sensor data when the battery has different degrees of damage in history, analyze the collected sensor data, and extract key features that cause the battery to be damaged; According to the key features, calculate a first judgment threshold for judging whether the battery is damaged; Obtain the data records of the battery damage in history, conduct a correlation analysis on the data records and the key features, obtain effective parameters associated with the key features, and calculate a second effective threshold; Obtain the battery sensor data during operation in real time, and determine whether the battery is damaged according to the first judgment threshold, so as to conduct real-time monitoring and damage warning; Calculate the effectiveness of the damage warning, and determine whether the damage warning is effective according to the second effectiveness threshold.

[0019] Arrange array electromagnetic sensors and acoustic sensors in the battery module area, specifically: Select different array electromagnetic sensors according to the size of different battery internal environments, and integrate the battery module with the array electromagnetic sensors; The acoustic sensor includes an accelerometer and a piezoelectric sensor; A permalloy shielding cover is arranged outside the array electromagnetic sensor for electromagnetic screen signal protection; Integrate a vibration baseline calibration device outside the acoustic sensor; Among them, the array electromagnetic sensor includes a fluxgate sensor, a Hall effect sensor, etc.; and ensure that they can cover the key areas of the battery, including battery cells, cooling systems, etc.

[0020] Collect the sensor data when the battery has different degrees of damage in the collection history, analyze the collected sensor data, and extract the key features that cause the battery to be damaged, specifically: Step S3-1: Collect the sensor data when the battery has different degrees of damage in the history, specifically: {a 1 、a 2 、…、a n}; where a 1 、a 2 、…、a n represent the 1st, 2nd,..., nth sensor data values when the battery is damaged in the collected history, n represents the type label of the sensor data, and n is a positive integer; collect the sensor data when the battery is operating normally in the history, specifically: {b 1 、b 2 、…、b n}; where b 1 、b 2 、…、b n represent the 1st, 2nd,..., nth sensor data values when the battery is operating normally in the collected history; conduct N collections; Step S3-2: Based on the sensor data when the battery has different degrees of damage and the sensor data when the battery is operating normally in the N collected histories, calculate the average change rate of each sensor data; Specifically: ; where p(i) represents the average change rate of the i-th type of sensor data; a represents the sensor data value when the battery was damaged in history; b represents the sensor data value when the battery was operating normally in history; N represents the number of times of sensor data collection; N is a positive integer; Step S3-3: Obtain the average change rate [p(1), p(2), …, p(n)] of each type of sensor data; where p(1), p(2), …, p(n) represent the calculated average change rates of the 1st, 2nd, …, n-th types of sensor data. Conduct a comparative analysis on the average change rates of the n types of sensor data, and determine the sensor data with an average change rate greater than the median as the key features, and obtain the key features of different batteries when they are damaged, specifically: [Q1, Q2, …, Q k ; where Q1, Q2, …, Q k represent the obtained 1st, 2nd, …, k-th key features; k represents the type label of the key features, k is a positive integer, and k ∈ (1, n).

[0021] According to the key features, calculate the first judgment threshold for determining whether the battery is damaged, specifically: Step S4-1: Obtain the key feature values {[Q 11 , Q 12 , …, Q 1M , [Q 21 , Q 22 , …, Q 2M , …, [Q k1 , Q k2 , …, Q kM} of the battery when it was damaged in M times of history; where Q kM represents the key feature value of the k-th key feature collected for the M-th time when the battery was damaged in history; M represents the number of times of key feature value collection, and M is a positive integer; Step S4-2: Calculate the first judgment threshold through the obtained key feature values of the battery when it was damaged in M times of history. The first judgment threshold is obtained based on the numerical difference between the key feature values of the battery when it was damaged and the key feature values of the battery when it was operating normally; Specifically: ; where s is the first judgment threshold, representing the numerical difference between the key feature values of the battery when it was damaged and the key feature values of the battery when it was operating normally; u and v represent the identifiers of the key feature values, u and v are positive integers, u ∈ [1, k], and v ∈ [1, M]; Q uv represents the key feature value of the u-th key feature of the v-th time when the battery was damaged in history; Q u ’ represents the average value of the u-th key feature of the battery when it was operating normally in history.

[0022] Example 1:

[0023] The fluxgate sensor measures the magnetic field intensity (unit: Tesla, T) in the critical area of the battery (near the battery cell), and the Hall effect sensor measures the magnetic field induction voltage (unit: Volt, V) generated by the coolant flow in the battery cooling system: For battery pack A, the array electromagnetic sensor includes a fluxgate sensor (measuring the magnetic field intensity in the critical area of the battery, and the corresponding data is recorded as the first type of sensor data), a Hall effect sensor (measuring the magnetic field induction voltage generated by the coolant flow in the battery cooling system, and the corresponding data is recorded as the second type of sensor data), a magnetoresistive sensor (measuring the magnetic field change rate in the critical area of the battery, and the corresponding data is recorded as the third type of sensor data), and a current sensor (measuring the battery operating current, and the corresponding data is recorded as the fourth type of sensor data), that is, n = 4, and it is ensured that they cover the critical areas such as the battery cell and the cooling system.

[0024] The sensor data when the battery had different degrees of damage in history was collected, and N = 5 collections were made; First damage data: a 1 = 0.05T (magnetic field intensity data value measured by the fluxgate sensor); a 2 = 2.5V (magnetic field induction voltage data value measured by the Hall effect sensor); a 3 = 0.002T / s (magnetic field change rate data value measured by the magnetoresistive sensor); a 4 = 15A (battery operating current data value measured by the current sensor); Second damage data: a 1 = 0.06T, a 2 = 3.0V, a 3 = 0.003T / s, a 4 = 18A; Third damage data: a 1 = 0.055T, a 2 = 2.0V, a 3 = 0.0025T / s, a 4 = 16A; Fourth damage data: a 1 = 0.07T, a 2 = 3.5V, a 3 = 0.004T / s, a 4 = 20A; Fifth damage data: a 1 = 0.045 T, a 2 = 1.5 V, a 3 = 0.0015 T / s, a 4 = 14 A; Collect the sensor data when the battery operates normally in history, and also collect it N = 5 times; Data for the first normal operation: b 1 = 0.02 T (the data value of the magnetic field intensity measured by the fluxgate sensor); b 2 = 1.0 V (the data value of the magnetic induction voltage measured by the Hall effect sensor); b 3 = 0.001 T / s (the data value of the magnetic field change rate measured by the magnetoresistive sensor); b 4 = 10 A (the data value of the battery operating current measured by the current sensor); Data for the second normal operation: b 1 = 0.025 T, b 2 = 1.2 V, b 3 = 0.0012 T / s, b 4 = 11 A; Data for the third normal operation: b 1 = 0.015 T, b 2 = 0.8 V, b 3 = 0.0008 T / s, b 4 = 9 A; Data for the fourth normal operation: b 1 = 0.02 T, b 2 = 1.0 V, b 3 = 0.001 T / s, b 4 = 10 A; Data for the fifth normal operation: b 1 = 0.025 T, b 2 = 1.2 V, b 3 = 0.0012 T / s, b 4 = 11 A; Calculate the average change rate of each sensor data: = 350%; p(2) = 146%; p(3) = 156%; p(4) = 640%; The median is approximately 253%; Key features: Q1 is the magnetic field intensity data measured by the fluxgate sensor, and Q2 is the battery operating current data measured by the current sensor.

[0025] Obtain the key feature values, and take M = 3: [Q 11 = 0.05, Q 12 = 0.06, Q 13 = 0.055]; [Q 21 = 15, Q 22 = 18, Q 23 = 16]; Calculate the average value of the magnetic field intensity data measured by the fluxgate sensor when the battery is operating normally, Q1' = 0.021T; The average value of the battery operating current data measured by the current sensor when the battery is operating normally, Q2' = 10.2A; The first judgment threshold s ≈ 3.084; Obtain the data records of battery breakage in the history, conduct a correlation analysis on the data records and the key features, and obtain the effective parameters related to the key features. Specifically: Step S5-1: Obtain the data records of battery breakage in the history, collect the influence parameters when the key feature values change by more than three standard deviations during the battery operation process, and establish an association relationship model between different influence parameters and key features with the influence parameters as independent variables and the key features as dependent variables; Among them, the collected influence parameter values are consistent with the key feature values in terms of quantity; Establish different association relationship models according to different influence parameters; Among them, establishing the association relationship model specifically is: ; Among them, Q k ' represents the predicted value of the key feature; G represents the influence parameter value; α represents the slope of the association relationship model; φ represents the intercept of the association relationship model; Example 2: Among them, the influence parameters can be the false alarm probability of the sensor, the consistency probability of voltage, current, and internal resistance, the charge and discharge influence, the historical similarity fitting degree, the duration and intensity of abnormal signals, etc., all of which may affect the effectiveness of battery breakage warnings of different types; From the historical battery breakage data records, relevant data on ambient temperature, charge and discharge rate, and battery internal resistance are collected; 10 groups of data are collected, that is, z = 10; Ambient temperature data G1 (°C): [20, 22, 25, 18, 23, 21, 24, 19, 26, 20]; Charge and discharge rate data G2: [0.5, 0.6, 0.8, 0.4, 0.7, 0.5, 0.8, 0.4, 0.9, 0.5]; Corresponding battery internal resistance data Q k (mΩ): [15, 16, 18, 14, 17, 15, 18, 14, 19, 15]; Step S5-2: Calculate the model parameters according to the correlation relationship model between different influencing parameters and key features; draw a scatter plot and analyze the scatter plot; Among them, calculate the slope and intercept of the correlation relationship model. The specific formula is: ; Among them, α represents the slope of the correlation relationship model; z represents the number of collected influencing parameter values and key feature values; G represents the collected influencing parameter values; represents the average value of the influencing parameters; Q k represents the collected key feature values; represents the average value of the key feature values; ; Calculate the correlation relationship model parameters between different influencing parameters and key features in turn; Establish a model with ambient temperature as the influencing parameter: ; Calculate the average ambient temperature: ≈21.8; Calculate the average battery internal resistance: ≈15.9; Calculate and obtain , ; Establish a model with the charge and discharge rate as the influencing parameter: ; Calculate the average charge and discharge rate: ; Calculate and obtain ,

[0026] As Figure 2 shown, it is the scatter plot of the correlation relationship model established with ambient temperature as the influencing parameter and the scatter plot of the correlation relationship model established with the charge and discharge rate as the influencing parameter; Step S5-3: According to the correlation relationship model of different influencing parameters, obtain several groups of predicted values of key features, calculate the difference values between the several groups of predicted values of key features and the subsequent actual measured values, and determine the influencing parameters greater than the average difference value as effective parameters; The process of obtaining several groups of predicted values of key features according to the correlation relationship model of different influencing parameters, calculating the difference values between the several groups of predicted values of key features and the subsequent actual measured values, and determining the influencing parameters greater than the average difference value as effective parameters is specifically as follows: ; where f represents the difference value between the predicted value of the key feature and the subsequent actual measured value in the correlation relationship model of different influencing parameters; Among them, it is considered that the influencing parameters less than or equal to the average difference value have insufficient effectiveness with respect to the key features; Determine the influencing parameters greater than the average difference value as effective parameters, and denote the set of effective parameters as {g1, g2,..., g r}; where g1, g2,..., g r respectively represent the 1st, 2nd,..., rth effective parameters; r represents the type label of the effective parameters, and r is a positive integer.

[0027] The process of calculating the second effective threshold is specifically as follows: Taking the determined effective parameters as the benchmark, obtain the historical data change rate and standard deviation of different effective parameters, and calculate the second effective threshold; ; where s’ represents the second effective threshold; μ g represents the change rate of different effective parameters; σ g represents the standard deviation of different effective parameters.

[0028] The process of obtaining the battery sensor data during the operation in real time, and judging whether the battery is damaged according to the first judgment threshold for real-time monitoring and damage warning is specifically as follows: Set up a warning device; Obtain the key feature values through real-time monitoring. When the numerical difference of the key feature values obtained by real-time monitoring is greater than the first judgment threshold, give a warning through the warning device; Among them, specifically: When give a warning through the warning device; Among them, represents the numerical difference of the key feature values, represents the key feature value collected last time, represents the key feature value obtained by the current monitoring.

[0029] Calculate the effectiveness of the breakage warning, and determine whether the breakage warning is effective according to the second effectiveness threshold. Specifically: Step S5-1: After receiving the warning signal from the warning device, establish a correlation model between different influencing parameters and key features according to the method described in step S5-1, calculate the effective parameters in the actual situation through the key features, and obtain the effective parameter values by real-time monitoring; Step S5-2: According to the effective parameter values obtained by real-time monitoring, when the change rate of the effective parameter values obtained by real-time monitoring is greater than the second effectiveness threshold, determine that the breakage warning is effective, and arrange for staff to inspect and repair the battery; Among them, specifically: When judge that the breakage warning is effective, and arrange for staff to inspect and repair the battery.

[0030] Specifically: Set a timestamp; The timestamp is used to judge the continuity of the collected data; The timestamp is used to mark the time of the obtained data each time the data is obtained, prevent time dislocation, and wrongly calculate the numerical difference of the key feature value and the change rate of the effective parameter value.

[0031] A battery breakage intelligent identification and monitoring system for battery operation and maintenance management, the battery breakage intelligent identification and monitoring system includes a data collection and transmission module, a feature extraction module, an analysis and calculation module, a real-time monitoring and warning module, an effectiveness evaluation module, and a user interaction module; The data collection and transmission module is used to collect signals from sensors in the battery module in real time, transmit the data to the processing end by wireless and wired means, and provide data storage and backup functions; The feature extraction module is used to extract key features from the collected data, calculate the change features of the signals by analyzing the changes in the battery health status, identify the fault modes, and provide necessary feature information for subsequent analysis; The analysis and calculation module is used to deeply analyze the extracted key features and calculate the first judgment threshold for judging battery breakage; By establishing a correlation model, obtain effective parameters, and calculate the second effectiveness threshold based on historical data; The real-time monitoring and warning module is used to continuously monitor the battery status according to the real-time data and the set threshold, automatically identify whether the battery is damaged, and trigger the warning device to give a breakage warning when the preset first judgment threshold is exceeded; The effectiveness evaluation module is used to evaluate the effectiveness of the breakage warning, and confirm the effectiveness of the breakage warning through comparative analysis of the real-time monitoring data and the warning signal; The user interaction module is used to provide a user interface for displaying real-time monitoring data, warning information, and system status, enabling the staff to intuitively view the battery status, adjust parameters, and receive warnings, and providing functions for viewing historical data and generating relevant reports.

[0032] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims concerned.

Claims

1. A battery damage intelligent identification and monitoring method for battery operation and maintenance management, characterized by: The battery damage intelligent identification and monitoring method specifically comprises the following steps: Arrange array electromagnetic sensors and acoustic sensors in the battery module area; Collecting sensor data of batteries that have been damaged to varying degrees in history, analyzing the collected sensor data, and extracting key features that cause the batteries to be damaged; Calculating, based on the key feature, a first judgment threshold for judging whether the battery is damaged; Acquire historical data records of battery damage, perform correlation analysis on the data records and the key features, acquire valid parameters associated with the key features, and calculate a second valid threshold; Acquire battery sensor data during operation in real time, determine whether the battery is damaged according to the first judgment threshold, and perform real-time monitoring and damage warning; The effectiveness of the damage warning is calculated, and whether the damage warning is effective is determined according to the second effectiveness threshold.

2. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 1 is characterized in that: The arrangement of array electromagnetic sensors and acoustic sensors in the battery module area is specifically as follows: Select different array-type electromagnetic sensors according to the size of different internal environments of the battery, and integrate the battery module with the array-type electromagnetic sensors; The acoustic sensor includes an accelerometer and a piezoelectric sensor; A Permalloy shielding cover is provided outside the array-type electromagnetic sensor for electromagnetic screen signal protection; A vibration baseline calibration device is integrated outside the acoustic sensor.

3. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 2 is characterized in that: The sensor data when the battery is damaged to different degrees in the history is collected, and the collected sensor data is analyzed to extract the key features that cause the battery to be damaged, specifically: Step S3-1, collecting sensor data when the battery is damaged to different degrees in history, specifically: 1 、a 2 , …, a n }; Among them, a 1 、a 2 , …, a n Indicates the 1st, 2nd, ..., nth sensor data values ​​when the battery is damaged in the collected history, where n represents the type label of the sensor data and is a positive integer; the sensor data when the battery is operating normally in the collected history is specifically: {b 1 , b 2 , …, b n }; where b 1 , b 2 , …, b n Indicates the 1st, 2nd, ..., nth sensor data values ​​when the battery is operating normally in the collected history; the collection is performed N times; Step S3-2, using the sensor data collected in N historical records when the battery is damaged to varying degrees and the sensor data when the battery is operating normally as a benchmark, calculate the average change rate of each sensor data; Step S3-3, obtaining the average change rate of each sensor data [p(1), p(2), ..., p(n)]; wherein p(1), p(2), ..., p(n) represent the calculated average change rates of the first, second, ..., nth sensor data, performing comparative analysis on the average change rates of the n sensor data, determining the sensor data with an average change rate greater than the median as a key feature, and obtaining the key features of different batteries when they are damaged, specifically: [Q1, Q2, ..., Q k ]; among them, Q1, Q2, …, Q k represents the first, second, ..., kth key features obtained; k represents the type label of the key feature, k is a positive integer, k∈(1,n).

4. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 3 is characterized in that: According to the key features, a first judgment threshold for judging whether the battery is damaged is calculated, specifically: Step S4-1, obtain the key feature value {[Q 11 , Q 12 , …, Q 1M ]、[Q 21 , Q 22 , …, Q 2M ], …, [Q k1 , Q k2 , …, Q kM ]}; where Q kM It represents the key characteristic value of the Mth time collected of the kth key characteristic when the battery is damaged in the history; M represents the number of times the key characteristic value is collected, and M is a positive integer; Step S4-2, calculating a first judgment threshold by using the key characteristic values ​​when the battery is damaged in the M times of history, wherein the first judgment threshold is obtained according to the numerical difference between the key characteristic values ​​when the battery is damaged and the key characteristic values ​​when the battery is operating normally.

5. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 4 is characterized in that: The data record of battery damage in the acquisition history is analyzed for correlation between the data record and the key feature to obtain valid parameters associated with the key feature, specifically: Step S5-1, obtaining data records of battery damage in history, collecting influencing parameters when the key characteristic values ​​change by more than three standard deviations during the battery operation process, and establishing a correlation model between different influencing parameters and key characteristics by taking the influencing parameters as independent variables and the key characteristics as dependent variables; Step S5-2, calculating model parameters according to the correlation relationship model between different influencing parameters and key features; drawing a scatter plot, and analyzing the scatter plot; Step S5-3: obtain several groups of key feature prediction values ​​according to the association relationship model of different influencing parameters, calculate the difference between the several groups of key feature prediction values ​​and subsequent actual measured values, and determine the influencing parameters greater than the difference average as effective parameters.

6. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 5 is characterized in that: The calculation of the second effective threshold is specifically as follows: Based on the determined effective parameters, the historical data change rates and standard deviations of different effective parameters are obtained to calculate the second effective threshold.

7. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 6 is characterized in that: The real-time acquisition of battery sensor data during operation, judging whether the battery is damaged according to the first judgment threshold, and performing real-time monitoring and damage warning are specifically as follows: Set up early warning devices; The key characteristic value is obtained through real-time monitoring, and when the numerical difference of the key characteristic value obtained through real-time monitoring is greater than the first judgment threshold, an early warning is issued through the early warning device.

8. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 7 is characterized in that: The effectiveness of the damage warning is calculated, and whether the damage warning is effective is determined according to the second effective threshold, specifically: Step S5-1, after receiving the warning signal from the warning device, establish a correlation relationship model between different influencing parameters and key features according to the method described in step S5-1, calculate the effective parameters in the actual situation through the key features, and obtain the effective parameter values ​​through real-time monitoring; Step S5-2: according to the effective parameter value obtained by real-time monitoring, when the rate of change of the effective parameter value obtained by real-time monitoring is greater than the second effective threshold, it is determined that the damage warning is effective, and staff are arranged to inspect and repair the battery.

9. The battery damage intelligent identification and monitoring method for battery operation and maintenance management according to claim 8 is characterized in that: Specifically: Set timestamp; The timestamp is used to determine the continuity of the collected data.

10. A battery damage intelligent identification and monitoring system for battery operation and maintenance management, using the battery damage intelligent identification and monitoring method for battery operation and maintenance management as claimed in any one of claims 1 to 9, characterized in that: The battery damage intelligent identification and monitoring system includes a data acquisition and transmission module, a feature extraction module, an analysis and calculation module, a real-time monitoring and early warning module, an effective evaluation module and a user interaction module; The data acquisition and transmission module is used to collect signals from sensors in the battery module in real time, transmit the data to the processing end by wireless and wired means, and provide data storage and backup functions; The feature extraction module is used to extract key features from the collected data, calculate the change characteristics of the signal by analyzing the changes in the battery health status, identify the failure mode, and provide necessary feature information for subsequent analysis; The analysis and calculation module is used to perform in-depth analysis on the extracted key features and calculate a first judgment threshold for judging battery damage; By establishing an association relationship model, effective parameters are obtained, and the second effective threshold is calculated based on historical data; The real-time monitoring and early warning module is used to continuously monitor the battery status according to the real-time data and the set threshold value, automatically identify whether the battery is damaged, and trigger the early warning device to issue a damage warning when the preset first judgment threshold is exceeded; The effectiveness evaluation module is used to evaluate the effectiveness of the damage warning, and confirm the effectiveness of the damage warning by comparing and analyzing the real-time monitoring data and the warning signal; The user interaction module is used to provide a user interface to display real-time monitoring data, warning information and system status, so that staff can intuitively view battery status, adjust parameters and receive warnings, and provide functions for viewing historical data and generating related reports.

Images