BMS battery management system

Abstract

The invention discloses a BMS battery management system, relates to the technical field of battery management systems, and solves the technical problem that the fault type of a zinc-nickel battery cannot be accurately identified and predicted in the prior art. According to the invention, the operation state evaluation coefficient and the working environment evaluation coefficient of the battery are calculated respectively through the operation data and the environment data, and whether the battery operation state early warning is carried out is judged based on the operation state evaluation coefficient and the preset operation state threshold. Based on the working environment evaluation coefficient and a preset environment threshold value, whether environment early warning is carried out or not is judged, an artificial intelligence model is trained through historical battery data, and a battery fault prediction model is obtained; the operation data of the zinc-nickel battery are detected in real time based on the battery fault prediction model, and the technical problem that an existing BMS cannot accurately recognize and predict the fault type of the zinc-nickel battery is solved.

Description

Technical Field

[0001] This invention belongs to the field of battery management systems, specifically a BMS battery management system. Background Technology

[0002] A Battery Management System (BMS) is an electronic system used to monitor and manage the health of battery packs, and is particularly crucial in electric vehicles, energy storage systems, and other applications using rechargeable batteries. However, most current BMS systems are designed for lithium-ion batteries and lack effective management systems for nickel-zinc batteries. Existing BMS systems can only improve the utilization rate of nickel-zinc batteries to a limited extent, and cannot accurately monitor overcharging and over-discharging phenomena, thus failing to guarantee the health and lifespan of the batteries.

[0003] The prior art (invention application with publication number CN116387645A) discloses a zinc-nickel battery BMS power management system, including: a battery protection module for controlling the battery's voltage, current, power, and temperature; a parameter detection module for detecting battery parameters and obtaining first detection data; an equalization charging module for equalizing the charging of the battery to ensure that each individual cell in the battery is in a synchronous working state; a power management module for measuring the charging and discharging power of the battery and obtaining measurement data; a SOC module for monitoring the remaining power of the battery to ensure that the battery operates in a safe state; a temperature management module for detecting the real-time temperature of the battery and obtaining second detection data; and a SOH module for determining the battery's health status based on the first detection data, measurement data, and second detection data. This invention application can achieve power equalization of the battery pack at any time during charging and discharging, ensuring the health status of the battery pack and greatly extending the battery pack's service life. However, the prior art does not consider that the fault diagnosis algorithms and protection strategies preset in existing BMS systems are mainly based on common fault modes of lithium-ion batteries, such as overcharging, over-discharging, and short circuits. Zinc-nickel batteries have unique failure modes, such as metal deposition and electrolyte decomposition, which makes it impossible for existing BMS to accurately identify and predict these specific failures.

[0004] Therefore, this invention proposes a BMS battery management system to solve the above-mentioned technical problems. Summary of the Invention

[0005] This invention aims to address at least one of the technical problems existing in the prior art. To this end, this invention proposes a Battery Management System (BMS) to solve the problem that the pre-set fault diagnosis algorithms and protection strategies in the prior art are mainly based on common fault modes of lithium-ion batteries, such as overcharge, over-discharge, and short circuit. However, zinc-nickel batteries have their own unique fault modes, such as metal deposition and electrolyte decomposition, which makes it difficult for existing BMS to accurately identify and predict these specific faults.

[0006] To achieve the above objectives, a first aspect of the present invention provides a BMS battery management system, comprising: a data acquisition module, a data analysis module, a battery early warning module, and a fault prediction module; Data acquisition module: used to acquire operational data of zinc-nickel batteries and environmental data of the zinc-nickel battery's working environment; Data analysis module: Calculates the battery's operating status evaluation coefficient based on operational data; calculates the battery's working environment evaluation coefficient based on environmental data; Battery warning module: Based on the operating status evaluation coefficient and preset operating status threshold, it determines whether to issue a battery operating status warning; if yes, it sends a warning message to the user terminal; otherwise, it continues to monitor and judge; wherein, the user terminal includes: mobile phone or computer; and, The system determines whether to issue an environmental warning based on the work environment assessment coefficient and preset environmental thresholds; if yes, it sends a warning message to the user; otherwise, it continues to monitor and determine the warning. Fault prediction module: An artificial intelligence model is trained based on historical battery data to obtain a battery fault prediction model; the operating data of the zinc-nickel battery is monitored in real time based on the battery fault prediction model to obtain prediction results; the prediction results include: battery health status, battery state of charge, and predicted fault type; the artificial intelligence model includes: convolutional neural network or deep belief network; and, The battery is processed based on early warning information and prediction results.

[0007] Preferably, the data acquisition module is communicatively and / or electrically connected to the data analysis module and the fault prediction module, respectively; the data analysis module is communicatively and / or electrically connected to the battery warning module and the fault prediction module, respectively.

[0008] Preferably, the acquisition of the operating data of the zinc-nickel battery and the environmental data of the zinc-nickel battery's operating environment includes: The system collects real-time operational data and environmental data of the zinc-nickel battery's operating environment using data acquisition equipment. This equipment includes temperature sensors, current sensors, voltage sensors, internal resistance detectors, vibration sensors, and humidity sensors. The operational data includes battery temperature, battery current, battery voltage, battery internal resistance, and battery vibration frequency. The environmental data includes ambient temperature and ambient humidity.

[0009] It should be noted that the vibration frequency of the battery refers to the vibration of the zinc-nickel battery caused by the vibration of the equipment during operation. For example, during the flight of a drone, especially when performing complex maneuvers, encountering unstable airflow, or during take-off and landing, it will experience significant mechanical vibration. Continuous or high-intensity vibration may cause the internal structure of the battery to loosen, the separator to be damaged, the electrode material to fall off or break, thereby causing internal short circuits, leakage or battery failure.

[0010] Preferably, the battery operating status evaluation coefficient calculated based on operating data includes: The battery's internal resistance is labeled R, and its vibration frequency is labeled Z. The battery's operating state evaluation coefficient is calculated using the formula: PY = A × ln(R+1) + B × e^Z / (e^Z+1); where PY is the battery's operating state evaluation coefficient, A and B are proportional coefficients, and ln( ) is a logarithmic function with the natural number e as its base.

[0011] Preferably, the calculation of the battery's operating environment evaluation coefficient based on environmental data includes: The ambient temperature is labeled W, and the ambient humidity is labeled H; The battery's operating state evaluation coefficient is calculated using the formula: PH = C × ln[(W - ZW)^2 + 1] + D × tanhH × lnH; where PH is the battery's operating state evaluation coefficient, C and D are proportional coefficients, and tanh( ) is the hyperbolic tangent function, and ZW is the optimal ambient temperature for the zinc-nickel battery under operating conditions.

[0012] Preferably, the step of determining whether to issue a battery operating status warning based on the operating status evaluation coefficient and a preset operating status threshold includes: Determine whether the operating status evaluation coefficient is greater than the preset operating status threshold; if yes, generate an operating status warning message; otherwise, continue to monitor and judge; the preset operating status threshold is set based on the actual operating data of the battery during operation.

[0013] This invention collects real-time operating data of zinc-nickel batteries, calculates the battery's operating status evaluation coefficient based on the operating data, and determines whether the operating status evaluation coefficient is greater than a preset operating status threshold; if yes, an operating status warning message is generated; otherwise, continuous monitoring and judgment are performed; this solves the problem that existing technologies do not provide warnings for the operating status of zinc-nickel batteries.

[0014] Preferably, the step of determining whether to issue an environmental warning based on a working environment assessment coefficient and a preset environmental threshold includes: Determine whether the working environment assessment coefficient is greater than the preset environmental threshold; if yes, generate an environmental warning message; otherwise, continue monitoring and judgment; the preset environmental threshold is set according to the actual degree of environmental influence on the battery.

[0015] High temperatures accelerate the aging of zinc-nickel batteries and disrupt their internal chemical balance, triggering side reactions and shortening battery life. Regarding environmental humidity, high humidity leads to electrode corrosion, passivation, and changes in electrolyte composition, affecting battery stability and lifespan. This invention collects environmental data on the zinc-nickel battery's operating environment in real time, calculates an environmental evaluation coefficient based on this data, and determines whether the evaluation coefficient exceeds a preset environmental threshold. If yes, an environmental warning is generated; otherwise, continuous monitoring and judgment are performed. This solves the problem of existing technologies lacking early warning mechanisms for the operating environment of zinc-nickel batteries.

[0016] Preferably, the training of the artificial intelligence model based on historical battery data includes: S1: Mark the initial time as t0 and the predicted time as t, and obtain the prediction time interval. t = t - t0; S2: Combine historical battery data at time t0 with the predicted time interval. The data at time t is integrated into standard input data; historical battery data at time t is integrated into standard output data; historical battery data includes: operating data, operating status evaluation coefficient, operating environment evaluation coefficient, and environmental data; the initial time is the time before prediction, and the prediction time is the time to be predicted; S3: Train an artificial intelligence model based on standard input data and standard output data to obtain a data prediction model; S4: Based on the data prediction model, predict the battery's operating data to obtain the data prediction results; among which, the data prediction results include: operating data at time t, operating status evaluation coefficient, working environment evaluation coefficient, and environmental data; S5: Train an artificial intelligence model based on the data prediction results to obtain a battery failure prediction model.

[0017] It should be noted that the historical battery data refers to historical data collected on the operation of zinc-nickel batteries and the environmental data of the zinc-nickel battery's working environment, as well as historical data on the calculation of operating status evaluation coefficients and working environment evaluation coefficients based on the operation data and environmental data.

[0018] Preferably, the step of training an artificial intelligence model based on data prediction results to obtain a battery failure prediction model includes: S1: The data prediction results are integrated into standard input data, and the battery health status, battery state of charge, and predicted fault type at the corresponding time are integrated into standard output data. Among them, the predicted fault types include: overcharge, over-discharge, short circuit, metal deposition, electrolyte decomposition, electrode active material shedding, and electrode passivation. S2: Train an artificial intelligence model based on standard input data and standard output data to obtain a battery fault prediction model.

[0019] It should be noted that the battery health status is estimated by using an existing BMS system to estimate the percentage of remaining charge relative to the total battery capacity by accumulating the total amount of charge flowing into or out of the battery; the battery state of charge is estimated by using an existing BMS system to compare the current performance of the battery with its best performance at the time of manufacture or throughout its entire life cycle; and the predicted fault type is manually identified based on historical data.

[0020] This invention integrates historical operating data of zinc-nickel batteries and their environmental factors, and utilizes artificial intelligence technology to achieve accurate prediction of battery performance from a specific initial moment to the predicted moment. It not only predicts the battery's operating state and environmental influence conditions, but also establishes a fault prediction model, effectively improving the predictability and maintenance efficiency of battery management, reducing operational risks caused by unexpected failures, and significantly extending the lifespan of zinc-nickel batteries and ensuring stable system operation.

[0021] Preferably, the battery processing based on early warning information and prediction results includes: Warning information and prediction results are sent to the user terminal via wireless transmission technology and displayed on the user terminal; the battery is processed based on the operation status warning information, environmental warning or prediction results; the wireless transmission technology includes: 4G / 5G or WIFI.

[0022] Compared with the prior art, the beneficial effects of the present invention are: 1. Existing technologies do not consider that the fault diagnosis algorithms and protection strategies preset in existing BMS systems are mainly based on common fault modes of lithium-ion batteries, such as overcharge, over-discharge, and short circuit. However, zinc-nickel batteries have their own unique fault modes, such as metal deposition and electrolyte decomposition, which makes it difficult for existing BMS systems to accurately identify and predict these specific faults. This invention trains an artificial intelligence model using historical battery data to obtain a battery fault prediction model. Based on the battery fault prediction model, the operating data of zinc-nickel batteries is monitored in real time to obtain prediction results, thus solving the above problems.

[0023] 2. This invention uses historical battery data at time t0 and the predicted time interval. The system integrates the data into standard input data (t); integrates the historical battery data at time t into standard output data; trains an artificial intelligence model based on the standard input and standard output data to obtain a data prediction model; predicts the battery's operating data based on the data prediction model to obtain the data prediction result; and solves the problem that existing technologies do not consider predicting the operating status and environmental data of zinc-nickel batteries, which makes it impossible to intervene before the battery malfunctions or fails.

[0024] 3. This invention calculates the battery's operating status evaluation coefficient and working environment evaluation coefficient based on operational data and environmental data, respectively. It then determines whether to issue a battery operating status warning based on the operating status evaluation coefficient and a preset operating status threshold, and determines whether to issue an environmental warning based on the working environment evaluation coefficient and a preset environmental threshold. This helps prevent potential battery failures and performance degradation, improving safety and lifespan. It also solves the problem that existing technologies do not consider monitoring and warning of the operating status of zinc-nickel batteries and the environmental conditions in which they are located, leading to undetected abnormal operating states (such as internal short circuits, metal dendrite growth) and adverse environmental conditions (such as excessively high temperature or humidity) that may cause battery expansion and leakage, or even explosion, posing serious safety hazards. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the system modules according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the specific process of an embodiment of the present invention; Figure 3 This is a flowchart illustrating the construction process of the battery fault prediction model in an embodiment of the present invention. Detailed Implementation

[0027] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Please see Figures 1-2The first aspect of the present invention provides a BMS battery management system, including: a data acquisition module, a data analysis module, a battery early warning module and a fault prediction module; Data acquisition module: used to acquire operational data of zinc-nickel batteries and environmental data of the zinc-nickel battery's working environment; Data analysis module: Calculates the battery's operating status evaluation coefficient based on operational data; calculates the battery's working environment evaluation coefficient based on environmental data; Battery warning module: Determines whether to issue a battery operation status warning based on the operation status evaluation coefficient and preset operation status threshold; if yes, a warning message is sent to the user terminal; if no, it continues to monitor and judge; wherein, the user terminal includes: mobile phone or computer; and determines whether to issue an environmental warning based on the working environment evaluation coefficient and preset environmental threshold; if yes, a warning message is sent to the user terminal; if no, it continues to monitor and judge. The system collects real-time operational data and environmental data of the zinc-nickel battery's operating environment using data acquisition equipment. This equipment includes temperature sensors, current sensors, voltage sensors, internal resistance detectors, vibration sensors, and humidity sensors. The operational data includes battery temperature, battery current, battery voltage, battery internal resistance, and battery vibration frequency. The environmental data includes ambient temperature and ambient humidity.

[0029] The battery's operational status evaluation coefficient is calculated based on operational data, including: The battery's internal resistance is labeled R, and its vibration frequency is labeled Z. The battery's operating state evaluation coefficient is calculated using the formula: PY = A × ln(R+1) + B × e^Z / (e^Z+1); where PY is the battery's operating state evaluation coefficient, A and B are proportional coefficients, and ln( ) is a logarithmic function with the natural number e as its base.

[0030] The battery's operating environment evaluation coefficient is calculated based on environmental data, including: The ambient temperature is labeled W, and the ambient humidity is labeled H; The battery's operating state evaluation coefficient is calculated using the formula: PH = C × ln[(W - ZW)^2 + 1] + D × tanhH × lnH; where PH is the battery's operating state evaluation coefficient, C and D are proportional coefficients, and tanh( ) is the hyperbolic tangent function, and ZW is the optimal ambient temperature for the zinc-nickel battery under operating conditions.

[0031] Whether to issue a battery operation status warning is determined based on the operation status evaluation coefficient and preset operation status threshold, including: Determine whether the operating status evaluation coefficient is greater than the preset operating status threshold; if yes, generate an operating status warning message; otherwise, continue to monitor and judge; the preset operating status threshold is set based on the actual operating data of the battery during operation.

[0032] Whether to issue an environmental warning is determined based on the work environment assessment coefficient and preset environmental thresholds, including: Determine whether the working environment assessment coefficient is greater than the preset environmental threshold; if yes, generate an environmental warning message; otherwise, continue monitoring and judgment; the preset environmental threshold is set according to the actual degree of environmental influence on the battery.

[0033] For example: Suppose a drone uses a zinc-nickel battery as its power source, and its data acquisition module is equipped with a temperature sensor, current sensor, voltage sensor, internal resistance detector, vibration sensor, and humidity sensor. During a flight mission, the following data is collected at a certain point in time: Running data: Temperature (T): 45℃; Current (I): 10A; Voltage (V): 12V; Internal resistance (R): 0.1Ω; Vibration frequency (Z): 20Hz; Environmental data: Ambient temperature (W): 40℃; Ambient humidity (H): 60%; The battery's operating state evaluation coefficient is calculated using the formula: PY = A × ln(R+1) + B × e^Z / (e^Z+1). Assuming proportional coefficients A = 0.5 and B = 0.9, substituting them into the formula... PY=0.5×ln(0.1+1)+0.9×e^20 / (e^20+1)≈1.3; The battery's operating state evaluation coefficient is calculated using the formula: PH = C × ln[(W - ZW)^2 + 1] + D × tanhH × lnH. Assuming a proportional gain of C = 0.2, D = 0.7, and the optimal operating temperature (ZW) for the zinc-nickel battery is 30℃, the formula is substituted as follows: PH=0.2×ln[(40-30)^2+1]+0.7×tanh60×ln60≈3.79; Assume the preset operating state threshold is 1 and the preset environment threshold is 3.

[0034] Compare the calculation results: The operating status evaluation coefficient (PY≈1.3) is greater than the preset threshold (1), so the system will generate an operating status warning message to remind the user that the battery operating status may be abnormal and needs to be paid attention to or measures should be taken.

[0035] If the working environment assessment coefficient (PH≈3.79) is greater than the preset environmental threshold (3), the system will also generate environmental warning information to remind the user that the current environment may be detrimental to battery performance and lifespan, and to suggest adjusting the working environment or taking protective measures.

[0036] This invention monitors the operating status and working environment of zinc-nickel batteries in real time, promptly identifies potential problems and issues early warnings, effectively preventing battery failures and performance degradation, demonstrating the practical application value of this invention in improving battery management and maintenance efficiency.

[0037] Fault prediction module: Trains an artificial intelligence model based on historical battery data to obtain a battery fault prediction model; performs real-time monitoring of the operating data of zinc-nickel batteries based on the battery fault prediction model to obtain prediction results; the prediction results include: battery health status, battery state of charge, and predicted fault type; the artificial intelligence model includes: convolutional neural network or deep belief network; and processes the battery based on early warning information and prediction results.

[0038] Please see Figure 3 The artificial intelligence model is trained based on historical battery data, including: S1: Mark the initial time as t0 and the predicted time as t, and obtain the prediction time interval. t = t - t0; S2: Combine historical battery data at time t0 with the predicted time interval. The data at time t is integrated into standard input data; historical battery data at time t is integrated into standard output data; historical battery data includes: operating data, operating status evaluation coefficient, operating environment evaluation coefficient, and environmental data; the initial time is the time before prediction, and the prediction time is the time to be predicted; S3: Train an artificial intelligence model based on standard input data and standard output data to obtain a data prediction model; S4: Based on the data prediction model, predict the battery's operating data to obtain the data prediction results; among which, the data prediction results include: operating data at time t, operating status evaluation coefficient, working environment evaluation coefficient, and environmental data; S5: Train an artificial intelligence model based on the data prediction results to obtain a battery failure prediction model.

[0039] An artificial intelligence model is trained based on the data prediction results to obtain a battery failure prediction model, including: S1: The data prediction results are integrated into standard input data, and the battery health status, battery state of charge, and predicted fault type at the corresponding time are integrated into standard output data. Among them, the predicted fault types include: overcharge, over-discharge, short circuit, metal deposition, electrolyte decomposition, electrode active material shedding, and electrode passivation. S2: Train an artificial intelligence model based on standard input data and standard output data to obtain a battery fault prediction model.

[0040] Battery processing is based on early warning information and prediction results, including: Warning information and prediction results are sent to the user terminal via wireless transmission technology and displayed on the user terminal; the battery is processed based on the operation status warning information, environmental warning or prediction results; the wireless transmission technology includes: 4G / 5G or WIFI.

[0041] For example, suppose a company focused on operating electric buses faces service disruptions and increased maintenance costs due to battery failures. To address these issues, the company decides to deploy an AI-based battery failure prediction system to improve battery reliability and extend its lifespan. The specific steps are as follows: 1. Historical data collection and integration; Initial time (t0): 08:00 AM every day.

[0042] Predicted time (t): 09:00 AM every day.

[0043] Prediction time interval ( t): 1 hour.

[0044] Historical battery data includes: temperature, current, voltage, internal resistance, vibration frequency, ambient temperature, and ambient humidity for each vehicle's battery. In addition, it includes operating condition evaluation coefficients and working environment evaluation coefficients calculated from historical operating data and historical environmental data.

[0045] 2. Data preprocessing; Historical battery data is preprocessed; this preprocessing includes normalization and feature extraction. Historical battery data at time t0 and t is integrated into the standard input data format.

[0046] The historical battery data at time t is integrated into a standard output data format.

[0047] 3. Model training; Deep Belief Networks (DBNs) were chosen as the artificial intelligence model because they perform well in processing time series data and prediction tasks.

[0048] The DBN model is trained using standard input and output data. The model parameters are optimized through multiple iterations to ensure that the model can accurately predict the future state of the battery, thus obtaining a data prediction model.

[0049] 4. Prediction; The trained data prediction model is used to predict the battery data at time t0, thus obtaining the battery data at time t.

[0050] 5. Fault prediction model training; The data prediction results (battery data at time t) are integrated into the new standard input data.

[0051] The battery health status, state of charge, and predicted fault types (such as overheating, overcharging, short circuit, etc.) are integrated into standard output data.

[0052] Then, a battery failure prediction model is obtained by training with DBN.

[0053] 6. Real-time monitoring and early warning; Real-time data acquisition equipment is installed on buses to continuously collect vehicle battery data and send the data to a central processing system.

[0054] The battery data collected from the vehicle is predicted using a data prediction model to obtain the battery data one hour later. The battery fault prediction model is used to predict battery faults one hour later, and the battery health status, state of charge, and predicted fault type are obtained one hour later.

[0055] 7. Processing and Response; Based on the alerts received by the company's maintenance team, team members can remotely access the system to view detailed battery status and forecast results.

[0056] Based on the early warning information, the maintenance team can arrange necessary inspections and maintenance work in advance, such as adjusting charging strategies, replacing parts, or performing preventative maintenance.

[0057] At the same time, the system sends notifications to fleet managers via a mobile application to ensure they are aware of the real-time status and maintenance needs of the vehicles.

[0058] This invention utilizes artificial intelligence technology to effectively predict and manage faults in zinc-nickel batteries, thereby improving battery performance and safety while reducing operating costs.

[0059] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. 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.

[0060] Working principle of the invention: This invention addresses the technical problem of existing BMS systems failing to accurately identify and predict zinc-nickel battery fault types. It involves acquiring real-time operational data and environmental data of the zinc-nickel battery's working environment. The system calculates an operational status evaluation coefficient based on the operational data and an environmental environment evaluation coefficient based on the environmental data. It then determines whether to issue a battery operational status warning based on the operational status evaluation coefficient and a preset operational status threshold. If yes, a warning message is sent to the user terminal; otherwise, continuous monitoring and judgment are performed. Additionally, it determines whether to issue an environmental warning based on the working environment evaluation coefficient and a preset environmental threshold. If yes, a warning message is sent to the user terminal; otherwise, continuous monitoring and judgment are performed. An artificial intelligence model is trained based on historical battery data to obtain a battery fault prediction model. The system then performs real-time monitoring of the zinc-nickel battery's operational data based on this model to obtain prediction results. Finally, the system processes the battery based on the warning information and prediction results. This solution solves the problem that existing technologies do not consider the fault diagnosis algorithms and protection strategies in existing BMS systems, which are mainly based on common fault modes of lithium-ion batteries, such as overcharge, over-discharge, and short circuits. Zinc-nickel batteries have unique fault modes, such as metal deposition and electrolyte decomposition, leading to the inability of existing BMS systems to accurately identify and predict zinc-nickel battery fault types.

[0061] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A battery management system (BMS), characterized in that, include: Data acquisition module, data analysis module, battery warning module, and fault prediction module; Data acquisition module: used to acquire operational data of zinc-nickel batteries and environmental data of the zinc-nickel battery's working environment; Data analysis module: Calculates the battery's operating status evaluation coefficient based on the operating data; The battery's operating environment evaluation coefficient is calculated based on environmental data; Battery warning module: Determines whether to issue a battery operating status warning based on the operating status evaluation coefficient and preset operating status threshold; If yes, then a warning message will be sent to the user's device; No, then continue monitoring and judgment; the user end includes: mobile phone or computer; and, The system determines whether to issue an environmental warning based on the work environment assessment coefficient and preset environmental thresholds; if yes, it sends a warning message to the user; otherwise, it continues to monitor and determine the warning. Fault prediction module: An artificial intelligence model is trained based on historical battery data to obtain a battery fault prediction model; the operating data of the zinc-nickel battery is monitored in real time based on the battery fault prediction model to obtain prediction results; the prediction results include: battery health status, battery state of charge, and predicted fault type; the artificial intelligence model includes: convolutional neural network or deep belief network; and, The battery is processed based on early warning information and prediction results.

2. The BMS battery management system according to claim 1, characterized in that, The data acquisition module is communicatively and / or electrically connected to the data analysis module and the fault prediction module, respectively; the data analysis module is communicatively and / or electrically connected to the battery warning module and the fault prediction module, respectively.

3. The BMS battery management system according to claim 1, characterized in that, The acquisition of operational data and environmental data of the zinc-nickel battery includes: The data acquisition equipment collects real-time operating data of the zinc-nickel battery and environmental data of the battery's working environment. The operating data includes the battery's temperature, current, voltage, internal resistance, and vibration frequency. The environmental data includes ambient temperature and humidity.

4. A BMS battery management system according to claim 1, characterized in that, The battery operating status evaluation coefficient calculated based on operating data includes: The battery's internal resistance is labeled R, and its vibration frequency is labeled Z. The battery's operating state evaluation coefficient is calculated using the formula: PY = A × ln(R+1) + B × e^Z / (e^Z+1); where PY is the battery's operating state evaluation coefficient, A and B are proportional coefficients, and ln( ) is a logarithmic function with the natural number e as its base.

5. A BMS battery management system according to claim 1, characterized in that, The battery's operating environment evaluation coefficient, calculated based on environmental data, includes: The ambient temperature is labeled W, and the ambient humidity is labeled H; The battery's operating state evaluation coefficient is calculated using the formula: PH = C × ln[(W - ZW)^2 + 1] + D × tanhH × lnH; where PH is the battery's operating state evaluation coefficient, C and D are proportional coefficients, and tanh( ) is the hyperbolic tangent function, and ZW is the optimal ambient temperature for the zinc-nickel battery under operating conditions.

6. A BMS battery management system according to claim 1, characterized in that, The step of determining whether to issue a battery operating status warning based on the operating status evaluation coefficient and a preset operating status threshold includes: Determine whether the operating status evaluation coefficient is greater than the preset operating status threshold; if yes, generate an operating status warning message; otherwise, continue to monitor and judge; the preset operating status threshold is set based on the actual operating data of the battery during operation.

7. A BMS battery management system according to claim 1, characterized in that, The method of determining whether to issue an environmental warning based on a working environment assessment coefficient and a preset environmental threshold includes: Determine whether the working environment assessment coefficient is greater than the preset environmental threshold; if yes, generate an environmental warning message; otherwise, continue monitoring and judgment; the preset environmental threshold is set according to the actual degree of environmental influence on the battery.

8. A BMS battery management system according to claim 1, characterized in that, The training of the artificial intelligence model based on historical battery data includes: S1: Mark the initial time as t0 and the predicted time as t, and obtain the prediction time interval. t = t - t0; S2: Combine historical battery data at time t0 with the predicted time interval. The data at time t is integrated into standard input data; historical battery data at time t is integrated into standard output data; historical battery data includes: operating data, operating status evaluation coefficient, operating environment evaluation coefficient, and environmental data; the initial time is the time before prediction, and the prediction time is the time to be predicted; S3: Train an artificial intelligence model based on standard input data and standard output data to obtain a data prediction model; S4: Based on the data prediction model, predict the battery's operating data to obtain the data prediction results; among which, the data prediction results include: operating data at time t, operating status evaluation coefficient, working environment evaluation coefficient, and environmental data; S5: Train an artificial intelligence model based on the data prediction results to obtain a battery failure prediction model.

9. A BMS battery management system according to claim 8, characterized in that, The process of training an artificial intelligence model based on data prediction results to obtain a battery failure prediction model includes: S1: The data prediction results are integrated into standard input data, and the battery health status, battery state of charge, and predicted fault type at the corresponding time are integrated into standard output data. Among them, the predicted fault types include: overcharge, over-discharge, short circuit, metal deposition, electrolyte decomposition, electrode active material shedding, and electrode passivation. S2: Train an artificial intelligence model based on standard input data and standard output data to obtain a battery fault prediction model.

10. A BMS battery management system according to claim 1, characterized in that, The battery processing based on early warning information and prediction results includes: Warning information and prediction results are sent to the user terminal via wireless transmission technology and displayed on the user terminal; the battery is processed based on the operation status warning information, environmental warning or prediction results; the wireless transmission technology includes: 4G / 5G or WIFI.

Images