Intelligent maintenance system for rail transit storage battery

Abstract

The invention provides an intelligent maintenance system for a rail transit storage battery, and the system comprises a plurality of sensors which are used for collecting the operation data of the rail transit storage battery, and are used for monitoring the environment temperature, the environment humidity, the material stress, the gas pressure and the gas temperature; the battery failure evaluation module is used for receiving the parameter data acquired by the sensor and calculating a first failure probability of the storage battery through a vector machine model; calculating a second failure probability of the storage battery according to the Bayesian model probability distribution of each feature under the failure condition of the storage battery; the error evaluation module is used for calculating the difference degree of the first failure probability P1 and the second failure probability P2 based on the first failure probability P1 and the second failure probability P2 of the storage battery; and the data analysis module is used for comparing the difference degree with a set threshold value, and triggering an early warning signal when the difference degree exceeds the threshold value. By means of the technical scheme, real-time intelligent monitoring and failure early warning of the rail transit storage battery can be achieved, the timeliness and accuracy of maintenance are improved, and the operation risk caused by storage battery faults is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent maintenance of batteries in rail transit systems, and in particular to an intelligent maintenance system for rail transit batteries, so as to improve the reliability and maintenance efficiency of rail transit batteries. Background Art

[0002] Batteries are widely used as emergency backup power sources in rail transit systems to ensure stable operation of the system when the main power fails. Once a battery fails, it will directly affect the safety of rail transit, especially in emergency situations, which may cause serious operational safety hazards. Therefore, health monitoring and failure warning of rail transit batteries are key links to ensure system safety and continuity.

[0003] Traditional battery maintenance mostly uses regular manual inspections and simple parameter monitoring methods, mainly by monitoring the battery's voltage, current, temperature and other basic physical parameters to determine its status. However, this maintenance method has the following problems:

[0004] Unable to monitor in real time: The manual inspection cycle is long, and it is difficult to capture the rapid aging or failure signs of the battery in real time, resulting in a large risk blind spot.

[0005] Difficult to accurately assess aging status: The health status of the battery is affected by many factors, such as ambient temperature, humidity, pressure, and material aging. It is difficult to accurately predict the risk of failure based solely on parameters such as voltage.

[0006] High maintenance cost: Manual inspections require frequent arrangements for maintenance, which is costly. In addition, due to reliance on subjective human judgment, accuracy and consistency are poor.

[0007] In recent years, with the rapid development of the Internet of Things and intelligent algorithms, battery health management systems based on multi-parameter sensor monitoring and intelligent prediction algorithms have gradually become an effective solution. This method installs multiple sensors on the battery to obtain its voltage, current, temperature and other data in real time, and combines the failure prediction model to analyze and warn the battery status. However, most systems currently rely mainly on a single model (such as a support vector machine model or a logistic regression model) to calculate the failure probability of the battery, and fail to effectively integrate multiple models to comprehensively evaluate the battery health status. Therefore, the prediction accuracy in complex environments is low, and it is difficult to adapt to the stringent safety requirements in rail transit. Summary of the invention

[0008] In view of this, the present invention provides a rail transit battery intelligent maintenance system, the system comprising:

[0009] Multiple sensors that collect rail transit battery operation data to monitor the ambient temperature T env、Ambient humidity H env , material stress S material , Gas pressure P gas and gas temperature T gas ;

[0010] The battery failure assessment module is used to receive parameter data collected by the sensor, calculate the first failure probability P1 of the battery through the vector machine model; and calculate the second failure probability P2 of the battery according to the Bayesian model probability distribution of each feature under the failure condition of the battery;

[0011] An error evaluation module calculates the difference between the first failure probability P1 and the second failure probability P2 of the battery;

[0012] The data analysis module compares the difference with a set threshold and triggers an early warning signal when the difference exceeds the threshold.

[0013] In particular, the first failure probability P1 of the battery is calculated by the vector machine model in the following manner:

[0014]

[0015] In particular, the second failure probability P2 of the battery is calculated according to the Bayesian model probability distribution of each feature under the failure condition of the battery by:

[0016]

[0017] Where P2 represents the second failure probability, P(failure) represents the prior probability of failure, which means the estimated probability of battery failure in the absence of observation data; this value is usually derived based on historical data and prior knowledge; P(x i |failure) means that under the failure condition, the i-th parameter x i The conditional probability of occurrence indicates the possibility of observing this characteristic value when the battery is known to fail; x i Parameters measured by multiple sensors that indicate the battery status, including the ambient temperature T env 、Ambient humidity H env , material stress S material , Gas pressure P gas and gas temperature T gas ; n: the total number of characteristic parameters, that is, the number of sensor parameters of the battery status; ∑ c∈{failure，no failure} In P(c), P (C) represents the prior probability of category c, where c∈{failure,no failure} represents the prior probability of failure or no failure; Represents all i parameters x under a given category ci The joint conditional probability of .

[0018] In particular, in the error evaluation module, the difference between the two is calculated as follows:

[0019]

[0020] Wherein, αα represents the difference control parameter, and β represents the sinusoidal control parameter.

[0021] In particular, the prior probability of failure in the Bayesian model probability distribution can be periodically updated according to historical data to improve the calculation accuracy of the failure probability.

[0022] In particular, the system is bus-powered and distributed in the machine rooms of various stations along the rail transit line. The data of each station is transmitted to the central processing unit through a wireless communication network, thereby realizing unified real-time monitoring and intelligent maintenance decision-making for all batteries.

[0023] In particular, the system further comprises a fault analysis module for analyzing the causes of the early warning, and optimizing the parameters of the Bayesian model and the vector machine model according to the analysis results, so as to improve the prediction accuracy of the two models.

[0024] In particular, the fault analysis module is also used to collect historical warning data of each site, and input the data into the deep learning model for training, so as to automatically learn the fault mode and provide a basis for parameter optimization.

[0025] In particular, an online monitoring terminal is installed in the machine room of each site to display the operating status and health of the batteries at the site in real time and replace faulty batteries according to maintenance instructions.

[0026] In particular, the online monitoring terminal can allow on-site maintenance personnel to manually input historical maintenance records through a human-computer interaction interface so as to update model parameters and improve the accuracy of subsequent predictions.

[0027] Beneficial effects:

[0028] 1. Through the technical solution of the present invention, real-time intelligent monitoring and failure warning of rail transit batteries can be achieved, which improves the timeliness and accuracy of maintenance and reduces the operational risks caused by battery failures.

[0029] 2. Through the technical solution of the present invention, the Bayesian model and the support vector machine model are combined to fully utilize the advantages of the two models. The Bayesian model can update the prior probability to capture long-term trends, and the support vector machine model can predict and capture short-term changes in real time, so it can more accurately evaluate the health status of the battery.

[0030] 3. Through the technical solution of the present invention, an error evaluation module is introduced to perform difference analysis on the prediction results of the two models. When the difference exceeds the threshold, an early warning is triggered, which can effectively detect model anomalies, prompting the system to require manual intervention and optimize model parameters.

[0031] 4. Through the technical solution of the present invention, a fault analysis module is set in the central processing unit, which can automatically analyze historical warning data, learn fault modes, provide a basis for model parameter optimization, and enable the system to have the ability of self-learning and adaptation.

[0032] 5. Through the technical solution of the present invention, the system adopts a distributed architecture, and an online monitoring terminal is set up in the on-site computer room to visualize the battery status of this site. At the same time, it supports manual input of maintenance records, realizes human-computer interaction, forms a closed-loop feedback optimization, and comprehensively improves the intelligence level of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is an intelligent maintenance system for rail transit batteries proposed in the present invention. DETAILED DESCRIPTION

[0034] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0035] To solve the above problems, the present invention proposes an intelligent maintenance system for rail transit batteries based on a combination of a Bayesian model and a support vector machine (SVM) model. The system collects multiple operating parameters of the battery and implements a comprehensive assessment of the probability of battery failure based on the prior probability distribution update mechanism of the Bayesian model and the real-time prediction capability of the support vector machine (SVM) model. It can dynamically adapt to the aging process of the battery and changes in environmental conditions, thereby improving the accuracy of failure prediction and the timeliness of early warning. This innovative solution significantly improves the automation level and reliability of rail transit battery maintenance, and provides a more efficient maintenance and management method for the rail transit system.

[0036] The present invention discloses an intelligent maintenance system for rail transit batteries. Figure 1 As shown, the system comprises:

[0037] Multiple sensors that collect rail transit battery operation data to monitor the ambient temperature T env 、Ambient humidity H env , material stress S material , Gas pressure P gas and gas temperature T gas ;

[0038] The battery failure assessment module is used to receive parameter data collected by the sensor, calculate the first failure probability P1 of the battery through the vector machine model; calculate the second failure probability P2 of the battery according to the Bayesian model probability distribution of each feature under the failure condition of the battery; the calculation of the first failure probability P1 of the battery through the vector machine model is calculated in the following way:

[0039]

[0040] The second failure probability P2 of the battery is calculated according to the Bayesian model probability distribution of each feature under the failure condition of the battery by the following method:

[0041]

[0042] Where P2 represents the second failure probability, P(failure) represents the prior probability of failure, which means the estimated probability of battery failure in the absence of observation data; this value is usually derived based on historical data and prior knowledge; P(x i |failure) means that under the failure condition, the i-th parameter x i The conditional probability of occurrence indicates the possibility of observing this characteristic value when the battery is known to fail; x i Parameters measured by multiple sensors that indicate the battery status, including the ambient temperature T env 、Ambient humidity H env , material stress S material , Gas pressure P gas and gas temperature T gas ; n: the total number of characteristic parameters, that is, the number of sensor parameters of the battery status; ∑ c∈{failure，no failure} In P(c), P (C) represents the prior probability of category c, where c∈{failure,no failure} represents the prior probability of failure or no failure; Represents all i parameters x under a given category c i The joint conditional probability of .

[0043] The prior probability of failure in the Bayesian model probability distribution can be periodically updated based on historical data to improve the calculation accuracy of the failure probability.

[0044] The error evaluation module calculates the difference between the first failure probability P1 and the second failure probability P2 of the battery. In the error evaluation module, the difference between the two is calculated by the following method:

[0045]

[0046] Wherein, α represents the difference control parameter, and β represents the sinusoidal control parameter.

[0047] The data analysis module compares the difference with a set threshold and triggers an early warning signal when the difference exceeds the threshold.

[0048] The system is bus-powered and distributed in the machine rooms of each station along the rail transit line. The data of each station is transmitted to the central processing unit through the wireless communication network, realizing unified real-time monitoring and intelligent maintenance decision-making for all batteries.

[0049] The system also includes a fault analysis module for analyzing the cause of the early warning and optimizing the parameters of the Bayesian model and the SVM model according to the analysis results to improve the prediction accuracy of the two models.

[0050] The fault analysis module is also used to collect historical warning data from each site and input the data into a deep learning model for training, thereby automatically learning fault modes and providing a basis for parameter optimization.

[0051] The computer room of each site is also equipped with an online monitoring terminal, which is used to display the operating status and health of the batteries at the site in real time and replace faulty batteries according to maintenance instructions. The online monitoring terminal allows on-site maintenance personnel to manually input historical maintenance records through a human-computer interaction interface to update model parameters and improve the accuracy of subsequent predictions.

[0052] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0053] It is obvious to those skilled in the art that the embodiments of the present invention are not limited to the details of the above exemplary embodiments, and that the embodiments of the present invention can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present invention is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the embodiments of the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units, modules or devices stated in the system, device or terminal claims can also be implemented by the same unit, module or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0054] Finally, it should be noted that the above implementation modes are only used to illustrate the technical solutions of the embodiments of the present invention and are not intended to limit them. Although the embodiments of the present invention have been described in detail with reference to the above preferred implementation modes, those skilled in the art should understand that the technical solutions of the embodiments of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rail transit battery intelligent maintenance system, characterized in that: The system comprises: Multiple sensors that collect rail transit battery operation data to monitor the ambient temperature T env 、Ambient humidity H env , material stress S material , Gas pressure P gas and gas temperature T gas ; The battery failure assessment module is used to receive parameter data collected by the sensor, calculate the first failure probability P1 of the battery through the vector machine model; and calculate the second failure probability P2 of the battery according to the Bayesian model probability distribution of each feature under the failure condition of the battery; An error evaluation module calculates the difference between the first failure probability P1 and the second failure probability P2 of the battery; The data analysis module compares the difference with a set threshold and triggers an early warning signal when the difference exceeds the threshold.

2. The rail transit battery intelligent maintenance system according to claim 1, characterized in that: The first failure probability P1 of the battery is calculated by the vector machine model in the following way:

3. The rail transit battery intelligent maintenance system according to claim 2, characterized in that: The second failure probability P2 of the battery is calculated according to the Bayesian model probability distribution of each feature under the failure condition of the battery by the following method: Where P2 represents the second failure probability, P(failure) represents the prior probability of failure, which means the estimated probability of battery failure in the absence of observation data; this value is usually derived based on historical data and prior knowledge; P(x i |failure) means that under the failure condition, the i-th parameter x i The conditional probability of occurrence indicates the possibility of observing this characteristic value when the battery is known to fail; x i Parameters measured by multiple sensors that indicate the battery status, including the ambient temperature T env 、Ambient humidity H env , material stress S material , Gas pressure P gas and gas temperature T gas ; n: the total number of characteristic parameters, that is, the number of sensor parameters of the battery status; ∑ c∈{failure，nofailure} In P(c), P (C) represents the prior probability of category c, where c∈{failure,no failure} represents the prior probability of failure or no failure; Represents all i parameters x under a given category c i The joint conditional probability of .

4. The rail transit battery intelligent maintenance system according to claim 3, characterized in that: In the error evaluation module, the difference between the two is calculated in the following way: Wherein, α represents the difference control parameter, and β represents the sinusoidal control parameter.

5. The rail transit battery intelligent maintenance system according to claim 4, characterized in that: The prior probability of failure in the Bayesian model probability distribution can be periodically updated according to historical data to improve the calculation accuracy of the failure probability.

6. The rail transit battery intelligent maintenance system according to claim 5, characterized in that: The system is bus-powered and distributed in the machine rooms of each station along the rail transit line. The data of each station is transmitted to the central processing unit through the wireless communication network, realizing unified real-time monitoring and intelligent maintenance decision-making for all batteries.

7. The intelligent maintenance system for rail transit batteries according to claim 6, characterized in that: The system also includes a fault analysis module for analyzing the cause of the early warning and optimizing the parameters of the Bayesian model and the vector machine model according to the analysis results to improve the prediction accuracy of the two models.

8. The rail transit battery intelligent maintenance system according to claim 7, characterized in that: The fault analysis module is also used to collect historical warning data from each site and input the data into a deep learning model for training, thereby automatically learning fault modes and providing a basis for parameter optimization.

9. The rail transit battery intelligent maintenance system according to any one of claims 6 to 8, characterized in that: There is also an online monitoring terminal in the computer room of each site, which is used to display the operating status and health of the battery at the site in real time and replace faulty batteries according to maintenance instructions.

10. The intelligent maintenance system for rail transit batteries according to claim 9, characterized in that: The online monitoring terminal can allow on-site maintenance personnel to manually input historical maintenance records through a human-computer interaction interface so as to update model parameters and improve the accuracy of subsequent predictions.

Images