Automatic monitoring system and method for lead storage battery

Abstract

The invention discloses an automatic monitoring system and method of a lead storage battery, relates to the technical field of battery monitoring, and solves the technical problem that the accuracy and efficiency of an automatic monitoring system of a battery are relatively low due to the fact that frequent charging and discharging and the influence of corresponding interval duration on battery aging are not fully considered in the prior art. A plurality of time clusters are generated according to charging and discharging data; calculating a plurality of battery aging coefficients according to the plurality of time clusters and calculating and monitoring a battery aging coefficient; calculating a monitored battery pack aging coefficient according to the monitored battery aging coefficient and then generating a battery pack estimated service life; an alarm signal is generated according to the estimated service life of a battery pack, charging and discharging behaviors of a battery are converged, the influence of frequent charging, the charging and discharging process and other factors on battery aging is considered, the state of each stage of a battery life cycle is calculated, the condition of the battery life cycle is effectively mastered, and the battery life cycle is accurately and accurately determined. And the accuracy and the efficiency of the automatic battery monitoring system are improved.

Description

Technical Field

[0001] This application belongs to the technical field of battery monitoring, and specifically relates to an automatic monitoring system and method for lead-acid batteries. Background Art

[0002] With the progress of technology and the improvement of environmental awareness, batteries, as important components for energy storage and conversion, have been widely used in various fields. However, a series of problems also accompany the use of batteries, such as battery aging and resource waste. During the use of batteries, it is necessary to strengthen the supervision of battery use to ensure the safety and compliance of batteries during use and improve the service life and efficiency of batteries. Therefore, strengthening battery supervision to ensure the safe, environmentally friendly and efficient use of batteries is of great significance for promoting sustainable development.

[0003] When the existing technology conducts automatic battery monitoring, it often monitors conventional parameters such as current, voltage, and temperature. There are some technologies that affect the degree of battery aging, but when monitoring battery aging, the impact of frequent charge and discharge and the corresponding interval duration on battery aging is not fully considered, resulting in low accuracy and efficiency of the battery automatic monitoring system; therefore, the automatic monitoring system for batteries still needs further improvement. Summary of the Invention

[0004] This application aims to solve at least one of the technical problems existing in the prior art; for this purpose, this application proposes an automatic monitoring system and method for lead-acid batteries to solve the technical problem that the existing technology does not fully consider the impact of frequent charge and discharge and the corresponding interval duration on battery aging, resulting in low accuracy and efficiency of the battery automatic monitoring system.

[0005] To achieve the above object, the first aspect of this application provides an automatic monitoring system for lead-acid batteries, including: a data acquisition module, a data analysis module, an early warning module, and a database; the data acquisition module is electrically and / or communicatively connected to the data analysis module; the data analysis module is electrically and / or communicatively connected to the early warning module; the database is electrically and / or communicatively connected to the data acquisition module, the data analysis module, and the early warning module respectively;

[0006] The data acquisition module: obtains battery pack data and environmental data through data acquisition devices; the battery pack data includes a number of battery IDs and battery parameters; the battery parameters include current, voltage, internal resistance, and charge and discharge data; the charge and discharge data includes charge time points;

[0007] The data analysis module: generates a number of time clusters according to the charge and discharge data; calculates a number of battery aging coefficients according to the number of time clusters; obtains the monitoring time in real time, and calculates the monitored battery aging coefficient according to the monitoring time, the battery aging coefficient, and the most recent charging time point; calculates the monitored battery pack aging coefficient according to a number of monitored battery aging coefficients, and then generates the estimated service life of the battery pack; generates an alarm signal according to the estimated service life of the battery pack and a number of monitored battery aging coefficients;

[0008] The warning module: makes a prompt according to the alarm signal and contacts the management personnel;

[0009] The database is used to store the data of each module and the historical data required for training the model.

[0010] Through the above steps, this application comprehensively analyzes the charge and discharge activities of the battery, and deeply considers the effects of various factors, including frequent charging behavior and charge and discharge cycle process, on the battery aging effect; on this basis, calculates the state characteristics of each stage of the battery life cycle, so as to comprehensively and effectively grasp the overall life cycle status of the battery, and improve the accuracy and operation efficiency of the battery automatic monitoring system.

[0011] Further, the generation of a number of time clusters according to the charge and discharge data includes:

[0012] Obtain the charge and discharge data; the charge and discharge data includes the number of charge and discharge times, the charging time point, and the charging duration;

[0013] Obtain the standby duration DS of a number of charging time points by performing a difference operation on adjacent charging time points i ;

[0014] Create a time cluster list SCL;

[0015] Select DS in turn i ;

[0016] When the SCL is an empty list, put DS i into the SCL;

[0017] Otherwise, judge whether DS i is less than the duration threshold; if yes, put DS i into the SCL; if not, create a new SCL.

[0018] Further, the calculation of a number of battery aging coefficients according to a number of time clusters includes:

[0019] Obtain a number of time clusters and their corresponding battery IDs, battery parameters, standby durations, and environmental data; the battery parameters include charge and discharge data; the charge and discharge data includes the charge and discharge rate C and the depth of discharge DoD; the environmental data includes the temperature T;

[0020] Calculate the aging coefficient LX of several batteries through the formula ; where j represents the number corresponding to the battery ID, n represents the number of the time cluster, m represents the number of the remaining power duration within the time cluster; k represents the proportionality coefficient, k ∈ (0, 1), a and b represent the exponential coefficients, a and b ∈ (0, 1); E represents the activation energy; R represents the gas constant; f j,n ; where j represents the number corresponding to the battery ID, n represents the number of the time cluster, m represents the number of the remaining power duration within the time cluster; k represents the proportionality coefficient, k ∈ (0, 1), a and b represent the exponential coefficients, a and b ∈ (0, 1); E represents the activation energy; R represents the gas constant; f j,n represents the time cluster correction function, specifically: where α is the proportionality coefficient, α ∈ (0, 1), β is the exponential coefficient, β ∈ (0, 1); ST j,n represents the total duration within the nth time cluster of the jth battery ID; DT represents the unit duration.

[0021] Furthermore, calculating the monitored battery aging coefficient according to the monitoring time, battery aging coefficient, and the most recent charging time point includes:

[0022] Obtain the monitoring time, the most recent charging time point, several historical remaining power durations in the current time cluster, and the aging coefficient LX corresponding to the most recent historical time cluster in real time j ;

[0023] Calculate the monitored aging increment of the current time cluster according to the monitoring time, the most recent charging time point, and several historical remaining power durations in the current time cluster;

[0024] Calculate the monitored battery aging coefficient JLX through the formula JLX j = LX j + ΔJLX j ; where ΔJLX j represents the monitored aging increment of the jth battery ID in the time cluster corresponding to the monitoring time. j represents the monitored aging increment of the jth battery ID in the time cluster corresponding to the monitoring time.

[0025] Furthermore, calculating the monitored aging increment of the current time cluster according to the monitoring time, the most recent charging time point, and several historical remaining power durations in the current time cluster includes:

[0026] Obtain the monitoring time, the most recent charging time point, and several historical remaining power durations in the current time cluster in real time;

[0027] Obtain the charge and discharge duration CFSC corresponding to the monitoring time; divide the charge and discharge duration into P ΔSCs, that is, P × ΔSC = CFSC; P is an integer, P > 0;

[0028] When the number of historical remaining power durations is 0, calculate the monitored aging increment ΔJLX through the formula Calculate the monitored aging increment ΔJLX jWhen the number corresponding to the historical remaining power duration is not 0, obtain the number G corresponding to the historical remaining power duration, its corresponding charge-discharge data and environmental data; the charge-discharge data includes the charge-discharge rate C and the depth of discharge DoD.

[0029] Calculate the monitored aging increment ΔJLX through the formula j ;

[0030]

[0031] where f j represents the correction function for the j-th battery: |I p | represents the current in the p-th time period, DSC represents the unit charge-discharge duration, Q j,max represents the nominal capacity of the j-th battery; k represents the proportionality coefficient, k ∈ (0, 1), a and b represent the exponential coefficients, a and b ∈ (0, 1); E represents the activation energy; R represents the gas constant; α is the proportionality coefficient, α ∈ (0, 1), β is the exponential coefficient, β ∈ (0, 1); DS represents the remaining power duration; DT represents the unit duration.

[0032] When evaluating the battery aging status in this application, taking the status of the battery at a specific monitoring time point as a consideration factor, it not only carefully analyzes the tiny aging effects on the battery during the charging process, but also equally attaches importance to the subtle aging effects accumulated during the discharging process, making the evaluation of the battery aging status at the monitoring time point more accurate, providing a solid and reliable data basis and support for subsequent operations and maintenance.

[0033] Further, after calculating the monitored battery pack aging coefficient based on several monitored battery aging coefficients and generating the estimated service life of the battery pack, it includes:

[0034] Obtain several monitored battery aging coefficients JLX j ;

[0035] Calculate the monitored battery pack aging coefficient JBLX through the formula ; where ω represents the weakest monomer weight coefficient, ω ∈ (0, 1), ω j represents the monomer weight coefficient, and the monomer weight coefficient is calculated through the internal resistance; γ represents the consistency penalty coefficient, γ ∈ (0, 1); ΔJLX = max(JLX j ) - min(JLX j ); max() and min() respectively represent the operations of taking the maximum value and the minimum value;

[0036] Obtain the historical monitored battery pack aging coefficients, historical environmental data, and historical battery pack data corresponding to several historical monitoring times;

[0037] Integrate the historical monitoring battery pack aging coefficients, historical environmental data, and historical battery pack data corresponding to several historical monitoring time points into a prediction sequence;

[0038] Input the prediction sequence into a life prediction model to obtain the predicted service life of the battery pack; the life prediction model is constructed through a machine learning model.

[0039] Furthermore, the monomer weight coefficient is calculated through internal resistance, including:

[0040] Obtain the internal resistance and environmental data corresponding to several batteries;

[0041] Through the formula Calculate the monomer weight coefficient ω j ; where NZ j represents the internal resistance of the jth battery, T j represents the temperature of the jth battery, YT represents the temperature threshold, E represents the activation energy; R represents the gas constant.

[0042] In this application, by adaptively assigning values to the monomer weight coefficient and taking temperature into account, the calculated monomer weight coefficient is more accurate, providing accurate data support for subsequent battery life prediction.

[0043] Furthermore, the life prediction model is constructed through a machine learning model, including:

[0044] Obtain several historical prediction sequences and their corresponding historical predicted service lives of the battery pack;

[0045] Divide several historical prediction sequences and their corresponding historical predicted service lives of the battery pack into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set;

[0046] Select a machine learning model as the basic model;

[0047] Train the basic model through the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0048] Verify the pre-trained model on the test set, and finally obtain a life prediction model that inputs several prediction sequences and outputs the predicted service life of the battery pack.

[0049] Furthermore, generating an alarm signal according to the estimated service life of the battery pack and several monitoring battery aging coefficients includes:

[0050] Obtain the estimated service life of the battery pack, several monitored battery aging coefficients, battery parameters, and environmental data; the battery parameters include voltage; the environmental data includes temperature;

[0051] Judge whether several voltages are within the voltage range. If so, do nothing; if not, generate a battery voltage anomaly alarm signal;

[0052] Judge whether several temperatures are within the temperature range. If so, do nothing; if not, generate a battery temperature anomaly alarm signal;

[0053] Judge whether several monitored battery aging coefficients are within the normal coefficient range. If so, do nothing; if not, generate a battery aging alarm signal;

[0054] Judge whether the estimated service life of the battery pack is greater than the life threshold. If so, do nothing; if not, judge whether the estimated service life of the battery pack is greater than H times the life threshold. If so, generate a low battery pack life warning signal; if not, generate a battery pack about to be scrapped alarm signal; where H is a proportionality coefficient, H ∈ (0, 1).

[0055] Another aspect of the present invention provides an automatic monitoring method for a lead-acid battery, including:

[0056] S0: Obtain battery pack data and environmental data; the battery pack data includes several battery IDs and battery parameters; the battery parameters include current, voltage, internal resistance, and charge and discharge data; the charge and discharge data includes charge time points;

[0057] S1: Generate several time clusters according to the charge and discharge data; calculate several battery aging coefficients according to the several time clusters;

[0058] S2: Obtain the monitoring time in real time, and calculate the monitored battery aging coefficient according to the monitoring time, battery aging coefficient, and the most recent charge time point;

[0059] S3: Calculate the monitored battery pack aging coefficient according to several monitored battery aging coefficients, and then generate the estimated service life of the battery pack;

[0060] S4: Generate an alarm signal according to the estimated service life of the battery pack and several monitored battery aging coefficients; give a prompt according to the alarm signal and contact the management personnel.

[0061] Compared with the prior art, the beneficial effects of this application are:

[0062] 1. This application generates several time clusters based on charge and discharge data; calculates several battery aging coefficients according to the several time clusters; obtains the monitoring time in real time, calculates the monitored battery aging coefficient based on the monitoring time and the most recent charging time point; generates the estimated service life of the battery pack after calculating the monitored battery pack aging coefficient based on several monitored battery aging coefficients; generates an alarm signal based on the estimated service life of the battery pack and several monitored battery aging coefficients, converges the charge and discharge behaviors of the battery, considers the impacts of multiple factors such as frequent charging and the charging and discharging processes on battery aging, calculates the states of each stage of the battery life cycle, effectively grasps the situation of the battery life cycle, and improves the accuracy and efficiency of the battery automatic monitoring system.

[0063] 2. When this application calculates the aging coefficient of the battery within different time clusters, it considers the aging increment within the current time cluster and calculates it by combining the multi-source parameters within the current time cluster with the number of frequent charging times, enabling the battery aging states within different time clusters to be more accurately reflected and improving the accuracy of the monitoring system.

[0064] 3. By also considering the state of the battery at the monitoring time point as a factor in the overall battery aging, this application not only considers the subtle aging caused during the charging process of the battery but also the subtle aging caused during the discharging process, making the battery aging state at the monitoring time point more accurate and providing good data support for subsequent operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0066] Figure 1 It is a schematic diagram of the principle of an automatic monitoring system for a lead-acid battery of the present application;

[0067] Figure 2 It is a flowchart of an automatic monitoring method for a lead-acid battery of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] The following will clearly and completely describe the technical solutions of the present application in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0069] Please refer to Figure 1, an embodiment of the first aspect of the present application provides an automatic monitoring system for a lead-acid battery, including: a data acquisition module, a data analysis module, an early warning module, and a database; the data acquisition module is electrically and / or communicatively connected to the data analysis module; the data analysis module is electrically and / or communicatively connected to the early warning module; the database is electrically and / or communicatively connected to the data acquisition module, the data analysis module, and the early warning module respectively;

[0070] Data acquisition module: Obtain battery pack data and environmental data through data acquisition devices; the battery pack data includes a number of battery IDs and battery parameters; the battery parameters include current, voltage, internal resistance, and charge and discharge data; the data acquisition devices include a number of sensors, etc.; the charge and discharge data includes charging time points;

[0071] Data analysis module: Generate a number of time clusters according to the charge and discharge data, where a time cluster refers to a time cluster obtained by dividing according to the time distance between the two most recent charging time points; calculate a number of battery aging coefficients according to the number of time clusters, where the battery aging coefficient refers to the degree of battery aging, and the larger the battery aging coefficient, the higher the degree of battery aging, which is used to calculate the battery aging coefficient corresponding to the historical time cluster; obtain the monitoring time in real time, and calculate the monitored battery aging coefficient according to the monitoring time, the battery aging coefficient, and the most recent charging time point, where the monitored battery aging coefficient refers to the aging coefficient of the battery at the monitoring time point; generate the estimated service life of the battery pack after calculating the monitored battery pack aging coefficient according to a number of monitored battery aging coefficients, where the monitored battery pack aging coefficient refers to the aging coefficient of the battery pack at the monitoring time point, and the estimated service life of the battery pack refers to the remaining normal service life of the battery pack at the monitoring time point; generate an alarm signal according to the estimated service life of the battery pack and a number of monitored battery aging coefficients;

[0072] Early warning module: Make a prompt according to the alarm signal and contact the management personnel; the alarm signals include battery voltage abnormal alarm signal, battery temperature abnormal alarm signal, battery aging alarm signal, battery pack about-to-be-scrapped alarm signal, etc.;

[0073] The database is used to store the data of each module and the historical data required for training the model.

[0074] Generating a number of time clusters according to the charge and discharge data in this embodiment includes:

[0075] Obtain the charge and discharge data; the charge and discharge data includes the number of charge and discharge times, charging time points, and charging durations;

[0076] Obtain the standby duration DS of a number of charging time points by performing a difference operation on adjacent two charging time points i ;

[0077] Create a list SCL of time clusters; the data stored in the time cluster list may be the data corresponding to a single power-off duration, or may be the data corresponding to multiple power-off durations.

[0078] Select DS in sequence i ;

[0079] When SCL is an empty list, put DS i into SCL;

[0080] Otherwise, judge whether DS i is less than the duration threshold, and the duration threshold is set according to experience; if so, put DS i into SCL; if not, create a new SCL.

[0081] In this embodiment, by dividing the historical charge and discharge behaviors of the battery and converging the frequent charging times according to the power-off durations between adjacent charging time points, a good time range can be provided when calculating the battery aging situation subsequently, improving the evaluation accuracy of the battery aging state.

[0082] Calculating several battery aging coefficients according to several time clusters in this embodiment includes:

[0083] Obtain several time clusters and their corresponding battery IDs, battery parameters, power-off durations, and environmental data; the battery parameters include charge and discharge data; the charge and discharge data include the charge and discharge rate C and the depth of discharge DoD; the environmental data includes the temperature T; the temperature T is expressed as the absolute temperature;

[0084] Through the formula calculate the aging coefficient LX of several batteries j,n ; where j represents the number corresponding to the battery ID, n represents the number of the time cluster, m represents the number of the power-off duration within the time cluster; k represents the proportionality coefficient, k ∈ (0, 1), a and b represent the exponential coefficients, a and b ∈ (0, 1), and the specific values are set according to experience. The setting of k is to reflect the overall sensitivity of the battery chemical system to the aging rate. The larger the k value, the faster the battery aging speed under the same working conditions. In this embodiment, k is set to 1.2×10 -5 , a is set to 0.8, and b is set to 0.5; E represents the activation energy, reflecting the sensitivity of temperature to the aging reaction. In this embodiment, E is set to 3×10 4 J / mol; R represents the gas constant, and the specific value is 8.314 J / (mol·K); f j,n represents the time cluster correction function, and the specific manifestation is: Among them, α is a proportionality coefficient, α ∈ (0, 1), β is an exponential coefficient, β ∈ (0, 1), and the specific values are set according to experience and are used to strengthen the battery aging caused by frequent charging. In this embodiment, α is set to 0.1 and β is set to 0.5; ST j,n represents the total duration within the nth time cluster of the jth battery ID; DT represents the unit duration, and the specific value is set according to experience; given the aging coefficient of the (n - 1)th time cluster, as the number of charging times within the nth time cluster increases, along with the increase in the charge-discharge rate and discharge depth, the corresponding aging coefficient will increase accordingly.

[0085] In this embodiment, in order to calculate the aging coefficient of the battery in different time clusters more precisely, the index of the aging increment within the current time cluster is adopted. This not only comprehensively considers various parameter sources within the current time cluster but also focuses on combining the number of frequent charging times to ensure that the assessment of the battery aging state can more accurately reflect the actual situation in each time interval, thereby improving the accuracy of the monitoring system in capturing and reflecting the battery aging state.

[0086] Calculating the monitored battery aging coefficient according to the monitoring time, battery aging coefficient, and the most recent charging time point in this embodiment includes:

[0087] Obtaining in real time the monitoring time, the most recent charging time point, several historical standby durations within the current time cluster, and the aging coefficient LX corresponding to the most recent historical time cluster j ;

[0088] Calculating the monitored aging increment of the current time cluster according to the monitoring time, the most recent charging time point, and several historical standby durations within the current time cluster; the monitored aging increment refers to the aging increment from the most recent time cluster to the monitoring time point;

[0089] Through the formula JLX j = LX j + ΔJLX j Calculate the monitored battery aging coefficient JLX j ; where, ΔJLX j represents the monitored aging increment of the jth battery ID corresponding to the time cluster of the monitoring time; at the monitoring time point, if there is a charge-discharge process between the monitoring time point and the most recent time cluster, a corresponding aging increment will be generated, causing the monitored battery aging coefficient to increase accordingly.

[0090] Calculating the monitored aging increment of the current time cluster according to the monitoring time, the most recent charging time point, and several historical standby durations within the current time cluster in this embodiment includes:

[0091] Obtaining in real time the monitoring time, the most recent charging time point, and several historical standby durations within the current time cluster;

[0092] Obtain the charge and discharge duration CFSC corresponding to the monitoring time; divide the charge and discharge duration into P ΔSCs, i.e., P×ΔSC = CFSC; P is an integer, P > 0, and the specific value is set according to experience;

[0093] When the number corresponding to the historical remaining power duration is 0, calculate the monitoring aging increment ΔJLX through the formula ; when the number corresponding to the historical remaining power duration is not 0, obtain the number G corresponding to the historical remaining power duration, its corresponding charge and discharge data, and environmental data; the charge and discharge data include the charge and discharge rate C and the depth of discharge DoD; the monitoring aging increment increases with the increase of parameters such as current and temperature, that is, the larger the charge and discharge current, the higher the degree of battery aging; j ;

[0094] Calculate the monitoring aging increment ΔJLX through the formula j ;

[0095]

[0096] Among them, f j represents the correction function of the j-th battery: |I p | represents the current in the p-th time period, DSC represents the unit charge and discharge duration, Q j,max represents the nominal capacity of the j-th battery; k represents the proportionality coefficient, k ∈ (0, 1), a and b represent the exponential coefficients, a and b ∈ (0, 1), and the specific values are set according to experience. The setting of k is to reflect the overall sensitivity of the battery chemical system to the aging rate. The larger the k value, the faster the battery aging speed under the same working conditions. In this embodiment, k is set to 1.2×10 -5 , a is set to 0.8, and b is set to 0.5; E represents the activation energy, which reflects the sensitivity of temperature to the aging reaction. In this embodiment, E is set to 3×10 4 J / mol; R represents the gas constant, and the specific value is 8.314 J / (mol·K); α is the proportionality coefficient, α ∈ (0, 1), β is the exponential coefficient, β ∈ (0, 1), and the specific values are set according to experience, which is used to strengthen the battery aging caused by frequent charging. In this embodiment, α is set to 0.1 and β is set to 0.5; DS represents the remaining power duration; DT represents the unit duration, and the specific value is set according to experience; when the number of historical remaining power durations is not 0, it means that there are several frequent charge and discharge behaviors before the monitoring time point, and it is necessary to consider the battery aging impact brought by the frequent charge and discharge behaviors, which is more equivalent to the combined aging impact of some historical remaining power durations in a time cluster and a current charge and discharge behavior.

[0097] After calculating the aging coefficients of several monitoring batteries in this embodiment to generate the predicted service life of the battery pack, it includes:

[0098] Obtain the aging coefficients JLX of several monitoring batteries j ;

[0099] Through the formula Calculate the aging coefficient JBLX of the monitoring battery pack; where, ω represents the weakest monomer weight coefficient, ω ∈ (0, 1), and the specific value is set according to the battery pack structure. If it is used for power batteries and is dominated by series connection, in this embodiment, ω is set to 0.7 to emphasize the short-board effect; if it is used for energy storage systems and is dominated by parallel connection, in this embodiment, ω is set to 0.3 to weaken the short-board effect; ω j Represents the monomer weight coefficient, and the monomer weight coefficient is calculated through the internal resistance; γ represents the consistency penalty coefficient, γ ∈ (0, 1), and the specific value is set according to experience. In this embodiment, γ is set to 0.2; ΔJLX = max(JLX j ) - min(JLX j ); max() and min() respectively represent the operations of taking the maximum value and the minimum value; when the weakest monomer weight coefficient is determined, as the aging degree corresponding to the highest value of the battery aging coefficient increases, and the difference between the maximum value and the minimum value of several battery aging coefficients is larger, then the aging coefficient of the battery pack will also increase accordingly;

[0100] Obtain the historical aging coefficients of the monitoring battery pack, historical environmental data, and historical battery pack data corresponding to several historical monitoring times;

[0101] Integrate the historical aging coefficients of the monitoring battery pack, historical environmental data, and historical battery pack data corresponding to several historical monitoring times into a prediction sequence;

[0102] Input the prediction sequence into the life prediction model to obtain the predicted service life of the battery pack; the life prediction model is constructed through a machine learning model.

[0103] In this embodiment, the aging coefficient of the entire battery pack is obtained through the aging coefficients of several batteries. When calculating the aging coefficient of the battery pack, it is considered that the performance of the battery pack is greatly affected by the worst performance among several batteries. Therefore, the weakest monomer constraint term is added, making the aging state of the battery pack more in line with the actual situation and improving the accuracy of the battery pack aging coefficient.

[0104] The monomer weight coefficient in this embodiment is calculated through the internal resistance, including:

[0105] Obtain the internal resistances and environmental data corresponding to several batteries;

[0106] Through the formula Calculate the monomer weight coefficient ω j ; where NZ j represents the internal resistance of the j-th battery, and T j represents the temperature of the j-th battery, YT represents the temperature threshold, and the specific value is set according to experience. In this embodiment, the temperature threshold is set to 298.15K. In this embodiment, the measured temperature and the temperature threshold are both expressed in absolute temperature. E represents the activation energy, which reflects the sensitivity of the aging reaction to temperature. In this embodiment, E is set to 3×10 4 J / mol; R represents the gas constant, and the specific value is 8.314 J / (mol·K); the denominator term in the above formula represents the actual battery internal resistance corrected by temperature using the Arrhenius model; as the actual internal resistance increases, the corresponding monomer weight coefficient will decrease accordingly.

[0107] The life prediction model in this embodiment is constructed by a machine learning model, including:

[0108] Obtain a number of historical prediction sequences and their corresponding predicted service lives of historical battery packs;

[0109] Divide a number of historical prediction sequences and their corresponding predicted service lives of historical battery packs into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; the ratio among the training set, the test set, and the validation set is 7:2:1;

[0110] Select a machine learning model as the basic model; the machine learning model includes an LSTM model, etc.;

[0111] Train the basic model through the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0112] Verify the pre-trained model on the test set, and finally obtain a life prediction model that inputs a number of prediction sequences and outputs the predicted service life of the battery pack.

[0113] Generating an alarm signal according to the predicted service life of the battery pack and several monitored battery aging coefficients in this embodiment includes:

[0114] Obtain the predicted service life of the battery pack, several monitored battery aging coefficients, battery parameters, and environmental data; the battery parameters include voltage; the environmental data includes temperature;

[0115] Judge whether several voltages are within the voltage range. The voltage range refers to the voltage range during normal operation, and the specific range is set according to experience; if yes, do nothing; if not, generate a battery voltage abnormal alarm signal;

[0116] Determine whether several temperatures are within the temperature range, where the temperature range refers to the temperature range during normal operation, and the specific temperature range is set according to experience; if yes, do nothing; if no, generate a battery temperature anomaly alarm signal;

[0117] Determine whether several monitored battery aging coefficients are within the normal coefficient range; the normal coefficient range means that the aging of the battery belongs to the normal coefficient range, and the specific values are set according to experience; if yes, do nothing; if no, generate a battery aging alarm signal;

[0118] Determine whether the estimated service life of the battery pack is greater than the life threshold, where the life threshold is set according to the battery material; if yes, do nothing; if no, determine whether the estimated service life of the battery pack is greater than H times the life threshold; if yes, generate a low battery pack life warning signal; if no, generate a battery pack about to be scrapped alarm signal; where H is a proportionality coefficient, H ∈ (0, 1), and the specific value is set according to experience. In this embodiment, H is set to 0.6.

[0119] Please refer to Figure 2 , another embodiment of the present application provides an automatic monitoring method for a lead-acid battery, including:

[0120] S0: Obtain battery pack data and environmental data; the battery pack data includes several battery IDs and battery parameters; the battery parameters include current, voltage, internal resistance, and charge and discharge data; the charge and discharge data includes charge time points;

[0121] S1: Generate several time clusters according to the charge and discharge data; calculate several battery aging coefficients according to the several time clusters;

[0122] S2: Obtain the monitoring time in real time, and calculate the monitored battery aging coefficient according to the monitoring time, battery aging coefficient, and the most recent charge time point;

[0123] S3: Calculate the monitored battery pack aging coefficient according to several monitored battery aging coefficients, and then generate the estimated service life of the battery pack;

[0124] S4: Generate an alarm signal according to the estimated service life of the battery pack and several monitored battery aging coefficients; make a prompt according to the alarm signal and contact the management personnel.

[0125] Some of the data in the above formula are calculated by removing the dimension and taking their numerical values. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the real situation; the preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0126] Working principle of this application: By obtaining battery pack data and environmental data; S1: Generate several time clusters according to charge and discharge data; Calculate several battery aging coefficients based on the several time clusters; Obtain the monitoring time in real time, and calculate the monitoring battery aging coefficient according to the monitoring time and the most recent charging time point; Generate the estimated service life of the battery pack after calculating the monitoring battery pack aging coefficient based on the several monitoring battery aging coefficients; Generate an alarm signal based on the estimated service life of the battery pack and the several monitoring battery aging coefficients; Make a prompt according to the alarm signal, contact the management personnel, converge the charge and discharge behaviors of the battery, consider the influence of multiple factors such as frequent charging and the charging and discharging process on battery aging, calculate the state of each stage of the battery life cycle, effectively master the situation of the battery life cycle, improve the accuracy and efficiency of the battery automatic monitoring system, and avoid the problem that the existing technology does not fully consider the influence of frequent charge and discharge and its corresponding interval duration on battery aging, resulting in low accuracy and efficiency of the battery automatic monitoring system.

[0127] The above embodiments are only used to illustrate the technical method of this application and not to limit it. Although this application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of this application can be modified or equivalently replaced without departing from the spirit and scope of the technical method of this application.

Claims

1. An automatic monitoring system for lead-acid batteries, characterized in that: include: Interconnected data acquisition module and data analysis module; The data acquisition module: acquires battery pack data and environmental data through a data acquisition device; the battery pack data includes several battery IDs and battery parameters; the battery parameters include current, voltage and internal resistance as well as charge and discharge data; the charge and discharge data includes charging time points; The data analysis module: generates a number of time clusters according to the charging and discharging data; calculates a number of battery aging coefficients according to the number of time clusters; obtains the monitoring time in real time, and calculates the monitoring battery aging coefficient according to the monitoring time, the battery aging coefficient and the most recent charging time point; calculates the monitoring battery pack aging coefficient according to the number of monitoring battery aging coefficients and then generates an estimated service life of the battery pack; generates an alarm signal according to the estimated service life of the battery pack and the number of monitoring battery aging coefficients.

2. The automatic monitoring system for lead-acid batteries according to claim 1, characterized in that: The generating of a plurality of time clusters according to the charge and discharge data includes: Acquiring charging and discharging data; the charging and discharging data includes the number of charging and discharging times, charging time points and charging duration; By performing a difference calculation on two adjacent charging time points, the standby time DS at several charging time points is obtained. i ; Create a time cluster list SCL; Select DS i ; When SCL is an empty list, DS i Place in SCL; Otherwise, judge DS i Is it less than the duration threshold? If yes, DS i Place in the SCL; if not, create a new SCL.

3. The automatic monitoring system for lead-acid batteries according to claim 1, characterized in that: The calculating of a plurality of battery aging coefficients according to a plurality of time clusters includes: Acquire several time clusters and their corresponding battery IDs, battery parameters, standby time and environmental data; the battery parameters include charge and discharge data; the charge and discharge data include charge and discharge rate C and discharge depth DoD; the environmental data includes temperature T; By formula Calculate the aging factor LX of several batteries j,n ; Wherein, j represents the number corresponding to the battery ID, n represents the number of the time cluster, m represents the number of the waiting time in the time cluster; k represents the proportionality coefficient, k∈(0,1), a and b represent the exponential coefficients, a and b∈(0,1); E represents the activation energy; R represents the gas constant; f j,n It is expressed as a time cluster correction function, which is specifically expressed as: Among them, α is the proportional coefficient, α∈(0,1), β is the exponential coefficient, β∈(0,1); ST j,n It is represented as the total duration of the j-th battery ID in the n-th time cluster; DT is represented as the unit duration.

4. The automatic monitoring system for lead-acid batteries according to claim 1, characterized in that: The method of calculating the monitoring battery aging coefficient according to the monitoring time, the battery aging coefficient and the most recent charging time point includes: Real-time acquisition of monitoring time, the most recent charging time point, several historical standby times in the current time cluster, and the aging coefficient LX corresponding to the most recent historical time cluster j ; Calculate the monitoring aging increment of the current time cluster according to the monitoring time, the most recent charging time point, and several historical standby times in the current time cluster; By Formula JLX j =LX j +ΔJLX j Calculate and monitor battery aging coefficient JLX j ; Among them, ΔJLX j It is represented by the monitoring aging increment of the jth battery ID in the time cluster corresponding to the monitoring time.

5. The automatic monitoring system for lead-acid batteries according to claim 4, characterized in that: The step of calculating the monitoring aging increment of the current time cluster according to the monitoring time, the most recent charging time point, and several historical power waiting times in the current time cluster includes: Real-time acquisition of monitoring time, most recent charging time point, and several historical standby times in the current time cluster; Obtain the charge and discharge duration CFSC corresponding to the monitoring time; divide the charge and discharge duration into P ΔSC, that is, P×ΔSC=CFSC; P is an integer, P>0 When the number of historical standby time is 0, the formula Calculate the monitoring aging increment ΔJLX j ; When the number corresponding to the historical standby time is not 0, obtain the number G corresponding to the historical standby time and its corresponding charging and discharging data and environmental data; the charging and discharging data includes the charging and discharging rate C and the discharge depth DoD; The aging increment ΔJLX is calculated by the formula j ; Among them, f j Expressed as the correction function of the jth battery: |I p | represents the current in the pth time period, DSC represents the unit charge and discharge time, Q j,max It represents the nominal capacity of the j-th battery; k represents the proportionality coefficient, k∈(0,1); a and b represent the exponential coefficients, a and b∈(0,1); E represents the activation energy; R represents the gas constant; α is the proportionality coefficient, α∈(0,1); β is the exponential coefficient, β∈(0,1); DS represents the waiting time; DT represents the unit time.

6. The automatic monitoring system for lead-acid batteries according to claim 1, characterized in that: The step of calculating the monitoring battery pack aging coefficient according to the plurality of monitoring battery aging coefficients and then generating the estimated service life of the battery pack includes: Get some monitoring battery aging coefficients JLX j ; By formula Calculate and monitor the battery pack aging coefficient JBLX; where ω represents the weight coefficient of the weakest cell, ω∈(0,1), ω j It is represented as the monomer weight coefficient, and the monomer weight coefficient is calculated by the internal resistance; γ is represented as the consistency penalty coefficient, γ∈(0,1); ΔJLX=max(JLX j )-min(JLX j ); max() and min() represent the maximum and minimum value operations respectively; Obtain historical monitoring battery pack aging coefficients, historical environmental data, and historical battery pack data corresponding to several historical monitoring times; Integrate the historical monitoring battery pack aging coefficient, historical environmental data and historical battery pack data corresponding to several historical monitoring times into a prediction sequence; The predicted sequence is input into the life prediction model to obtain the predicted service life of the battery pack; the life prediction model is constructed through a machine learning model.

7. The automatic monitoring system for lead-acid batteries according to claim 6, characterized in that: The monomer weight coefficient is calculated by internal resistance, including: Obtain internal resistance and environmental data corresponding to several batteries; By formula Calculate the monomer weight coefficient ω j Among them, NZ j Expressed as the internal resistance of the jth battery, T j It represents the temperature of the j-th battery, YT represents the temperature threshold, E represents the activation energy; R represents the gas constant.

8. The automatic monitoring system for lead-acid batteries according to claim 6, characterized in that: The life prediction model is constructed by a machine learning model, including: Obtain several historical prediction sequences and their corresponding historical battery pack prediction service life; Divide several historical prediction sequences and their corresponding historical battery pack prediction service life into training data, verification data and test data; perform data preprocessing on the training data, verification data and test data to obtain a training set, a verification set and a test set; Select a machine learning model as the base model; Train the basic model with the training set, and adjust the learning rate and hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally get a life prediction model that inputs several prediction sequences and outputs the predicted service life of the battery pack.

9. The automatic monitoring system for lead-acid batteries according to claim 1, characterized in that: The generating of an alarm signal according to the estimated service life of the battery pack and a number of monitored battery aging coefficients includes: Obtaining the estimated service life of the battery pack, several monitored battery aging coefficients and battery parameters, and environmental data; the battery parameters include voltage; the environmental data includes temperature; Determine whether a number of voltages are within a voltage range, if yes, do nothing; if no, generate a battery voltage abnormality alarm signal; Determine whether a certain temperature is within the temperature range, if yes, do not do anything; if no, generate a battery temperature abnormality alarm signal; Determine whether the aging coefficients of several monitored batteries are within the normal coefficient range; if yes, do not perform any operation; if no, generate a battery aging alarm signal; Determine whether the estimated service life of the battery pack is greater than the life threshold. If yes, do not do anything. If no, determine whether the estimated service life of the battery pack is greater than H times the life threshold. If yes, generate a warning signal that the battery pack is low in service life. If no, generate an alarm signal that the battery pack is about to be scrapped. Wherein, H is the proportional coefficient, H∈(0,1).

10. An automatic monitoring method for a lead-acid battery, applied to an automatic monitoring system for a lead-acid battery according to any one of claims 1 to 9, characterized in that: include: S0: Acquire battery pack data and environmental data; the battery pack data includes several battery IDs and battery parameters; the battery parameters include current, voltage and internal resistance as well as charge and discharge data; the charge and discharge data includes charging time point; S1: Generate a number of time clusters according to the charge and discharge data; calculate a number of battery aging coefficients according to the number of time clusters; S2: Acquire the monitoring time in real time, and calculate the monitoring battery aging coefficient according to the monitoring time, the battery aging coefficient and the most recent charging time point; S3: Calculating the monitoring battery pack aging coefficient based on a number of monitoring battery aging coefficients to generate an estimated service life of the battery pack; S4: generating an alarm signal based on the estimated service life of the battery pack and a number of monitored battery aging factors; Respond to alarm signals and contact management.

Images