A battery management system for fire trucks
By monitoring and predicting battery status in the fire truck battery management system and dynamically adjusting charging parameters, the problem that existing systems cannot fully monitor and optimize charging strategies is solved, extending the battery life and improving charging efficiency.
Patent Information
- Application Number
- CN202410668653.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-28
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-05-28
AI Technical Summary
The existing fire truck battery management system has not yet achieved comprehensive monitoring and in-depth analysis of battery status, and the charging strategy is relatively fixed. It is impossible to optimize charging parameters based on the instant status and historical performance of the battery, resulting in a shortening of the battery life.
The acquisition and processing module, status monitoring module, fault prediction module and charging strategy adjustment module are adopted to monitor the battery status in real time, fault prediction is performed using logistic regression model, and charging parameters are dynamically adjusted according to the battery status, and preprocessing and storage are combined with sensor data. Use the visual module to display battery status and early warning information.
It improves the ability to monitor battery status and fault prediction, dynamically adjusts charging parameters, extends the battery life and optimizes charging efficiency.
Smart Images

Figure CN118405025B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of battery management, in particular to a battery management system for fire trucks. Background Art
[0002] Battery management systems play a vital role in the key operating equipment of fire trucks, especially with the popularity of electric and hybrid fire trucks. Traditional battery management systems mainly focus on basic battery charge and discharge control and simple status monitoring. Modern battery management systems have integrated advanced data acquisition, real-time monitoring, fault diagnosis and intelligent charging management. The current systems on the market can effectively predict battery performance degradation and potential failures through real-time data analysis, so as to take preventive measures in advance to ensure the continuous operation and safety of fire trucks. These systems usually rely on complex algorithms and sensor networks, as well as cloud storage technology to ensure real-time update and analysis of data.
[0003] Although the existing fire truck battery management system has made significant progress, there are still some obvious shortcomings. Most existing systems have not yet achieved comprehensive monitoring and in-depth analysis of battery status. They are usually limited to past data records and basic fault indications and lack the ability to predict future status. The system's charging strategy is often relatively fixed and not flexible enough. It cannot optimize charging parameters based on the battery's immediate status and historical performance, which may cause the battery to be overcharged or undercharged, thereby shortening the battery's service life. Summary of the invention
[0004] In view of the above problems existing in the existing fire truck battery management system, the present invention is proposed.
[0005] Therefore, the problem to be solved by the present invention is that the existing system has not yet achieved comprehensive monitoring and in-depth analysis of the battery status, and the system's charging strategy is often relatively fixed and not flexible enough, and cannot optimize the charging parameters according to the battery's immediate status and historical performance.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: A fire truck battery management system, comprising: a collection and processing module for collecting and preprocessing data of battery cells;
[0007] The status monitoring module is used to monitor the battery operation in real time and judge the battery status according to the safety threshold in real time;
[0008] A fault prediction module, used to predict faults using a logistic regression model;
[0009] A charging strategy adjustment module is used to dynamically adjust charging parameters according to the battery status;
[0010] A storage module, used to store the used battery data;
[0011] Visualization module, used to visualize battery status and warning information.
[0012] As a preferred solution of the fire truck battery management system described in the present invention, the data collection of the battery cell refers to installing a piezoelectric voltage sensor, a thermistor temperature sensor, a Hall effect current sensor and an electrochemical impedance spectrum internal resistance sensor on the battery cell, and using a GPS clock synchronization sensor to collect the voltage, temperature, charge and discharge current and internal resistance data of the battery cell in real time.
[0013] As a preferred solution of the fire truck battery management system of the present invention, wherein: preprocessing the data collected from the battery unit refers to using a Kalman filter to perform denoising on the data collected by the sensor;
[0014] The data collected by the sensor is processed by minimum standardization, and the formula is:
[0015]
[0016] where z j is the data after minimum standardization of the jth sensor, x j (t) is the data collected by the jth sensor at time t, λ is the attenuation factor, μ dj is the mean value of the data collected by the jth sensor, N is the number of sensor data, T is the time window, x ji is the value of the i-th data of the j-th sensor, and t is the time variable;
[0017] The IQR method was used to identify outliers and the detected outlier data were deleted.
[0018] As a preferred solution of the fire truck battery management system of the present invention, the real-time monitoring of battery operation and judging the battery status according to the real-time comparison of the safety threshold refers to performing arithmetic averaging fusion on the pre-processed sensor data to obtain a comprehensive battery status index, and the formula is:
[0019]
[0020] Where s is the state indicator after fusion, n is the number of sensors, β is the coefficient, max(z j ) and min(z j ) is the maximum and minimum value of the jth data in the sensor;
[0021] Using statistical methods, combining manufacturer data and historical performance data to set the safety threshold of the parameter, the formula is:
[0022] θ=μ p +k·σ p
[0023] Where θ is the set safety threshold, μ p is the state mean based on historical data, k is the safety factor, σ p is the state standard deviation based on historical data;
[0024] The fused status indicator s is compared with the established threshold in real time. If s>θ, the battery status is judged to be abnormal, an alarm is automatically triggered, and the maintenance personnel are notified. If s≤θ, the battery status is judged to be normal and real-time monitoring continues.
[0025] As a preferred solution of the fire truck battery management system of the present invention, the fault prediction using the logistic regression model refers to extracting features from the fused state indicators using the time window dynamic feature extraction technology, and the formula is:
[0026]
[0027] Where F(t) is the extracted feature at time t, W is the size of the time window, t0 is the current time, and s(t) is the indicator of the fusion state at time t;
[0028] The fault prediction is performed using a logistic regression model, expressed as:
[0029]
[0030] Where R(y=1|X) is the probability of failure under the feature vector X, ω T is the transpose of the vector ω, and b is the bias term;
[0031] The gradient descent method is used to optimize the parameters of the logistic regression model. The formula is:
[0032]
[0033] Where α is the learning rate, is the gradient of the loss function with respect to ω;
[0034]
[0035] where y i is the true class label of the i-th sample, X i is the feature vector of the i-th sample, m is the total number of training samples, is the partial derivative of the loss function J with respect to the bias b;
[0036] when When , stop the iteration;
[0037] Use the average probability method of historical data to set the fault prediction threshold. The formula is:
[0038]
[0039] in is the fault prediction threshold, g is the number of records in the historical fault data, r i is the fault occurrence in the i-th historical record, 0 means normal and 1 means fault;
[0040] The real-time fused status indicators are used with the trained logistic regression model to predict the fault status of the current status data of the battery. If a fault occurs, an alarm is triggered immediately. Continue real-time monitoring.
[0041] As a preferred solution of the fire truck battery management system of the present invention, the dynamic adjustment of charging parameters according to the state of the battery refers to using an adaptive control strategy to dynamically adjust the charging power according to the real-time state of the battery. The formula is:
[0042] p c (t) = p b ·(1+δ·tanh(γ·(s(t)-s1(t))))
[0043] where p c (t) is the charging power at time t, p b is the basic charging power, δ is the adjustment coefficient, γ is the sensitivity factor, and s1(t) is the target state of the battery;
[0044] s1(t)=μ s +λ·σ s
[0045] where μ s is the average value of the battery status indicator based on historical data, λ is the adjustment factor, σ s is the standard deviation of the battery status indicator.
[0046] As a preferred solution of the fire truck battery management system described in the present invention, the storage of the used battery data refers to storing the data collected by the sensor, the preprocessed data, the prediction results of the battery status, the results of the real-time monitoring of the battery status and the results of the safety threshold comparison in a time series database.
[0047] As a preferred solution of the fire truck battery management system of the present invention, wherein: the visual display of battery status and warning information refers to using the modern UI design tool Sketch to create an interface to display battery status and warning information;
[0048] The real-time monitoring results of the battery status in the stored data and the predicted results of the battery status are displayed in real time using the JavaScript chart library Chart.js, and the JavaScript library SweetAlert2 is used to pop up an alarm window and notify the maintenance personnel via email.
[0049] The beneficial effects of the present invention are as follows: the present invention collects data from battery cells, monitors battery operation in real time, judges battery status based on real-time comparison of safety thresholds, uses a logistic regression model to predict faults, and dynamically adjusts charging parameters according to the battery status, thereby improving the ability of preventive maintenance, optimizing battery charging efficiency, and extending battery service life. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0051] Figure 1 This is a flow chart of the battery management system for fire trucks.
[0052] Figure 2 This is a schematic diagram of the structure of the fire truck battery management system. DETAILED DESCRIPTION
[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.
[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0055] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.
[0056] Example 1
[0057] Reference Figure 1 and Figure 2, which is the first embodiment of the present invention, and provides a fire truck battery management system, the fire truck battery management system comprising:
[0058] S1, acquisition and processing module, used to collect data of battery cells and perform preprocessing;
[0059] It should be noted that collecting battery cell data refers to installing piezoelectric voltage sensors, thermistor temperature sensors, Hall effect current sensors and electrochemical impedance spectroscopy internal resistance sensors on the battery cells, and using GPS clock synchronization sensors to collect the battery cell voltage, temperature, charge and discharge current and internal resistance data in real time.
[0060] Through GPS clock synchronization, the time stamps of data collection are accurate and consistent, avoiding analysis errors caused by data asynchrony, improving data reliability and accuracy of subsequent processing. By real-time monitoring of multiple key parameters of the battery, the system can quickly detect signs of battery performance degradation, such as abnormal voltage, increased internal resistance, etc., so as to perform fault diagnosis and preventive maintenance in a timely manner. Accurate real-time data enables the battery management system to effectively adjust the charging and discharging strategy, extend the battery life, and improve the energy efficiency of the entire system. Integrated sensor technology provides a large amount of real-time data for battery management, supporting data-based decision making, such as optimization of battery charging status, load adjustment, and formulation of maintenance plans.
[0061] Furthermore, preprocessing the data collected from the battery unit refers to using a Kalman filter to perform denoising on the data collected by the sensor;
[0062] The data collected by the sensor is processed by minimum standardization, and the formula is:
[0063]
[0064] where z j is the data after minimum standardization of the jth sensor, x j (t) is the data collected by the jth sensor at time t, λ is the attenuation factor, μ dj is the mean value of the data collected by the jth sensor, N is the number of sensor data, T is the time window, x ji is the value of the i-th data of the j-th sensor, and t is the time variable;
[0065] The IQR method was used to identify outliers and the detected outlier data were deleted.
[0066] By using Kalman filters and minimum standardization processing, the quality of data can be significantly improved, errors and deviations can be reduced, and the battery management system can more accurately monitor and evaluate the performance and status of the battery. Through precise data preprocessing and effective identification and processing of outliers, the system can more sensitively and accurately identify potential faults and problems, thereby intervening and maintaining them in a timely manner. The QR method can effectively identify statistical outliers in real data sets, which may be caused by incorrect measurements, data entry errors, or abnormal events. In quality control, financial analysis, and any field that requires precise data input, accurately identifying and processing outliers is the key to ensuring the reliability of data analysis results.
[0067] S2, status monitoring module, used to monitor battery operation in real time and judge the battery status according to the safety threshold in real time;
[0068] It should be noted that real-time monitoring of battery operation and judging the battery status based on real-time comparison of safety thresholds refers to performing arithmetic averaging fusion on the pre-processed sensor data to obtain a comprehensive battery status indicator, and the formula is:
[0069]
[0070] Where s is the state indicator after fusion, n is the number of sensors, β is the coefficient, max(z j ) and min(z j ) is the maximum and minimum value of the jth data in the sensor;
[0071] Using statistical methods, combining manufacturer data and historical performance data to set the safety threshold of the parameter, the formula is:
[0072] θ=μ p +k·σ p
[0073] Where θ is the set safety threshold, μ p is the state mean based on historical data, k is the safety factor, σ p is the state standard deviation based on historical data;
[0074] The fused status indicator s is compared with the established threshold in real time. If s>θ, the battery status is judged to be abnormal, an alarm is automatically triggered, and the maintenance personnel are notified. If s≤θ, the battery status is judged to be normal and real-time monitoring continues.
[0075] Real-time monitoring allows the system to instantly capture changes in battery status and promptly identify potential performance degradation or failures, thereby reducing system downtime and maintenance costs. By comparing with the preset safety threshold, it can quickly determine whether the battery is within the normal operating range, effectively preventing accidents caused by battery problems. The safety threshold set in combination with historical data is more scientific and real-time, and can adjust the threshold according to the actual use of the battery, thereby improving the adaptability and accuracy of the entire monitoring system. Timely anomaly detection and alarm triggering can prevent further deterioration of battery damage, ensure user safety and maintain normal operation of the system.
[0076] S3, a fault prediction module, used for fault prediction using a logistic regression model;
[0077] Furthermore, using the logistic regression model for fault prediction refers to extracting features from the fused state indicators using the time window dynamic feature extraction technology. The formula is:
[0078]
[0079] Where F(t) is the extracted feature at time t, W is the size of the time window, t0 is the current time, and s(t) is the indicator of the fusion state at time t;
[0080] The fault prediction is performed using a logistic regression model, expressed as:
[0081]
[0082] Where R(y=1|X) is the probability of failure under the feature vector X, ω T is the transpose of the vector ω, and b is the bias term;
[0083] The gradient descent method is used to optimize the parameters of the logistic regression model. The formula is:
[0084]
[0085] Where α is the learning rate, is the gradient of the loss function with respect to ω;
[0086]
[0087] where y i is the true class label of the i-th sample, X i is the feature vector of the i-th sample, m is the total number of training samples, is the partial derivative of the loss function J with respect to the bias b;
[0088] when When , stop the iteration;
[0089] Use the average probability method of historical data to set the fault prediction threshold. The formula is:
[0090]
[0091] in is the fault prediction threshold, g is the number of records in the historical fault data, r i is the fault occurrence in the i-th historical record, 0 means normal and 1 means fault;
[0092] The real-time fused status indicators are used with the trained logistic regression model to predict the fault status of the current status data of the battery. If a fault occurs, an alarm is triggered immediately. Continue real-time monitoring.
[0093] Through the time window dynamic feature extraction technology, key features that are helpful for prediction can be effectively extracted from the fused state indicators. These features help capture the time series changes of the state and improve the accuracy of the prediction. The logistic regression model can predict the probability of fault occurrence based on the extracted features, providing a scientific basis for maintenance decisions. Using the gradient descent method to optimize parameters can ensure that the logistic regression model achieves the best performance on a given data set and reduce prediction errors. The fault prediction threshold calculated from historical data can be used to customize the early warning system based on actual operating experience, improving the adaptability and accuracy of the system response. The real-time prediction and alarm mechanism can respond to potential faults immediately and minimize damage.
[0094] S4, a charging strategy adjustment module, used to dynamically adjust charging parameters according to the battery status;
[0095] It should be noted that dynamically adjusting the charging parameters according to the battery status refers to using an adaptive control strategy to dynamically adjust the charging power according to the real-time status of the battery. The formula is:
[0096] p c (t) = p b ·(1+δ·tanh(γ·(s(t)-s1(t))))
[0097] where p c (t) is the charging power at time t, p b is the basic charging power, δ is the adjustment coefficient, γ is the sensitivity factor, and s1(t) is the target state of the battery;
[0098] s1(t)=μ s +λ·σ s
[0099] where μ sis the average value of the battery status indicator based on historical data, λ is the adjustment factor, σ s is the standard deviation of the battery status indicator.
[0100] By monitoring the battery status in real time and adjusting the charging power accordingly, the charging efficiency and life of the battery can be maximized. Dynamic adjustment helps prevent the battery from overcharging or undercharging, thereby protecting the health of the battery and extending its service life. Real-time adjustment of charging parameters makes the charging process more precise. Adjustment based on the current status of the battery, such as voltage, temperature, and capacity, helps to improve the energy utilization of the battery and reduce energy loss. Using historical data to set charging parameters can provide a more stable and scientific charging process. By considering the battery's past performance, its future behavior can be predicted, making the charging process safer and more efficient. Effective battery management can reduce energy waste and maintenance costs, and improve overall economic benefits.
[0101] S5, a storage module, used to store the used battery data;
[0102] Furthermore, storing the used battery data means storing the data collected by the sensor, the preprocessed data, the prediction results of the battery status, the results of the real-time monitoring of the battery status and the results of the safety threshold comparison in a time series database.
[0103] Storing battery-related data such as sensor data and preprocessed data in a time series database can efficiently manage a large amount of time-related data and optimize data query and analysis speed. The time series database supports the writing of high-frequency data and fast aggregate query, making data processing more efficient. By centrally storing battery data at different stages, including the original sensor data, preprocessed data, and predicted results of battery status, the entire life cycle of the battery can be comprehensively managed and analyzed. This method helps to identify trends and potential problems in battery performance and improve battery efficiency and safety. Storing the results of safety threshold comparison in a time series database can track the comparison of battery status with preset safety standards in real time, and promptly discover and respond to potential safety risks. This real-time monitoring and data recording strategy improves the ability to prevent and respond to emergencies.
[0104] S6, visualization module, used for visually displaying battery status and warning information;
[0105] It should be noted that visual display of battery status and warning information refers to using the modern UI design tool Sketch to create an interface to display battery status and warning information;
[0106] The real-time monitoring results of the battery status in the stored data and the predicted results of the battery status are displayed in real time using the JavaScript chart library Chart.js, and the JavaScript library SweetAlert2 is used to pop up an alarm window and notify the maintenance personnel via email.
[0107] By using the Sketch design tool, you can create an intuitive, user-friendly interface to display battery status and warning information, which helps improve the user experience and enables users to quickly understand the current condition of the battery. Using Chart.js to visualize the real-time monitoring results and prediction results of the battery status can provide a dynamic, easy-to-understand data display and enhance the functionality of the decision support system. The alert pop-up window provided by SweetAlert2 not only has a modern appearance, but also contains rich interactive functions, which can attract users' attention at critical moments and ensure that safety warnings are noticed and handled in time. When the system detects a critical battery problem, it can automatically notify maintenance personnel via email, which not only improves the response speed, but also enhances the ability of remote monitoring and management.
[0108] Example 2
[0109] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.
[0110] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.
[0111] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0112] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
Claims
1. A fire truck battery management system, characterized in that: include, An acquisition and processing module is used to acquire data of battery cells and perform pre-processing; The status monitoring module is used to monitor the battery operation in real time and judge the battery status according to the safety threshold in real time; A fault prediction module, used to predict faults using a logistic regression model; A charging strategy adjustment module is used to dynamically adjust charging parameters according to the battery status; A storage module, used to store the used battery data; Visualization module, used to visualize battery status and warning information; The data collection of the battery cell refers to installing a piezoelectric voltage sensor, a thermistor temperature sensor, a Hall effect current sensor and an electrochemical impedance spectrum internal resistance sensor on the battery cell, and using a GPS clock synchronization sensor to collect the voltage, temperature, charge and discharge current and internal resistance data of the battery cell in real time; Preprocessing the data collected from the battery unit refers to using a Kalman filter to perform denoising on the data collected by the sensor; The data collected by the sensor is processed by minimum standardization, and the formula is: where z j is the data after minimum standardization of the jth sensor, x j (t) is the data collected by the jth sensor at time t, λ is the attenuation factor, μ dj is the mean value of the data collected by the jth sensor, N is the number of sensor data, T is the time window, x ji is the value of the i-th data of the j-th sensor, and t is the time variable; The IQR method was used to identify outliers and the detected outlier data were deleted.
2. The fire truck battery management system according to claim 1, characterized in that: The real-time monitoring of battery operation and judging the battery status according to the safety threshold in real time refers to performing arithmetic averaging fusion on the pre-processed sensor data to obtain a comprehensive battery status indicator, and the formula is: Where s is the state indicator after fusion, n is the number of sensors, β is the coefficient, max(z j ) and min(z j ) is the maximum and minimum value of the jth data in the sensor; Using statistical methods, combining manufacturer data and historical performance data to set the safety threshold of the parameter, the formula is: θ=μ p +k·s p Where θ is the set safety threshold, μ p is the state mean based on historical data, k is the safety factor, σ p is the state standard deviation based on historical data; The fused status indicator s is compared with the established threshold in real time. If s>θ, the battery status is judged to be abnormal, an alarm is automatically triggered, and the maintenance personnel are notified. If s≤θ, the battery status is judged to be normal and real-time monitoring continues.
3. The fire truck battery management system according to claim 2, characterized in that: The use of a logistic regression model for fault prediction refers to using a time window dynamic feature extraction technique to extract features from the fused state indicators, and the formula is: Where F(t) is the extracted feature at time t, W is the size of the time window, t0 is the current time, and s(t) is the indicator of the fusion state at time t; The fault prediction is performed using a logistic regression model, expressed as: Where R(y=1|X) is the probability of failure under the feature vector X, ω T is the transpose of the vector ω, and b is the bias term; The gradient descent method is used to optimize the parameters of the logistic regression model. The formula is: Where α is the learning rate, is the gradient of the loss function with respect to ω; where y i is the true class label of the i-th sample, X i is the feature vector of the i-th sample, m is the total number of training samples, is the partial derivative of the loss function J with respect to the bias b; when When , stop the iteration; Use the average probability method of historical data to set the fault prediction threshold. The formula is: Where θ is the fault prediction threshold, g is the number of records in the historical fault data, and r i is the fault occurrence in the i-th historical record, 0 means normal and 1 means fault; The real-time fused status indicators are used with the trained logistic regression model to predict the fault status of the current status data of the battery. If R(y=1|X)>θ, a fault is predicted and an alarm is triggered immediately; if R(y=1|X)≤θ, real-time monitoring continues.
4. The fire truck battery management system according to claim 3, characterized in that: The dynamically adjusting charging parameters according to the battery status refers to using an adaptive control strategy to dynamically adjust the charging power according to the real-time status of the battery. The formula is: p c (t)=p b ·(1+δ·tanh(γ·(s(t)-s1(t)))) where p c (t) is the charging power at time t, p b is the basic charging power, δ is the adjustment coefficient, γ is the sensitivity factor, and s1(t) is the target state of the battery; s1(t)=μ s +l·s s where μ s is the average value of the battery status indicator based on historical data, λ is the adjustment factor, σ s is the standard deviation of the battery status indicator.
5. The fire truck battery management system according to claim 4, characterized in that: The storing of the used battery data refers to storing the data collected by the sensor, the preprocessed data, the prediction result of the battery status, the result of the real-time monitoring of the battery status and the result of the safety threshold comparison in a time series database.
6. The fire truck battery management system according to claim 5, characterized in that: The visual display of battery status and warning information refers to using a modern UI design tool Sketch to create an interface to display battery status and warning information; The real-time monitoring results of the battery status in the stored data and the predicted results of the battery status are displayed in real time using the JavaScript chart library Chart.js, and the JavaScript library SweetAlert2 is used to pop up an alarm window and notify the maintenance personnel via email.
