Mobile phone charging station state monitoring system for urban rail train

Abstract

The invention provides a mobile phone charging station state monitoring system for an urban rail train, and aims to realize real-time monitoring and management of a power supply state during charging of a mobile phone charging station. The system acquires current, voltage and temperature data of the mobile phone charging station in real time through the sensor module, and transmits the data to the central processing unit through the communication module; the central processing unit integrates a data analysis module, a man-machine interaction module and a database module, and preprocesses, analyzes and stores collected data. The system adopts an auto-encoder to carry out deep learning and identification on data, thereby realizing early fault early warning, and improving the reliability and safety of the mobile phone charging station. According to the system, the intelligent level of urban rail train charging service is improved, decision support is provided for maintenance personnel, and operation management of the mobile phone charging station is optimized.

Description

Technical Field

[0001] The invention relates to the technical field of power monitoring, and in particular to a state monitoring system for a mobile phone charging station for an urban rail train. Background Art

[0002] In the field of urban rail trains, with the popularization of mobile devices, passengers have an increasing demand for charging on trains; however, most of the existing mobile phone charging stations on urban rail trains lack effective status monitoring and fault warning mechanisms; these mobile phone charging stations usually cannot provide real-time status feedback. Once a fault occurs, manual inspection is often required to discover it, which is not only inefficient, but also difficult to ensure timely detection and handling of problems, affecting passengers' user experience and train operation efficiency.

[0003] Due to the lack of a unified supervision platform, the maintenance and management of mobile phone charging stations are often scattered and unsystematic, making it difficult to achieve centralized data analysis and tracking of historical trends; this decentralized management model limits the in-depth understanding and optimization of the performance of mobile phone charging stations, and it is difficult to provide data support for operational decisions; in the field of power supply, although there are some monitoring technologies, they often require complex wiring and high maintenance costs, and have limitations in real-time, accuracy and intelligence; especially in the mobile environment of urban rail trains, how to achieve real-time and accurate monitoring of charging equipment, and how to improve the accuracy and response speed of fault warnings through intelligent means are challenges facing current technologies. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the deficiencies in the prior art, the present invention provides a status monitoring system for a mobile phone charging station for an urban rail train, which solves the problem of how to monitor the power supply status of a mobile phone charging station on an urban rail train in real time.

[0005] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: A mobile phone charging station status monitoring system for urban rail trains, comprising: Mobile phone charging stations are installed in each carriage of the urban rail train. Each mobile phone charging station is equipped with a sensor module to collect the current, voltage and temperature data of the mobile phone charging station in real time; The sensor module is connected to the communication module via the RS485 bus, which is used to receive the data collected by the sensor module and convert it into a digital signal; The communication module communicates with the central processing unit through a network switch and transmits the converted data to the central processing unit to realize the control of the power supply of the mobile phone charging station; The central processing unit integrates the mobile phone charging station power supply monitoring system, including data analysis module, human-computer interaction module and database module, among which: The data analysis module receives and stores the digital signal transmitted from the communication module and pre-processes the received data; the data analysis module identifies the digital signal through the autoencoder to determine the normal working state of the power supply of the mobile phone charging station, thereby realizing early warning of abnormal conditions through the human-computer interaction module; The human-computer interaction module is used to display the power supply information of each mobile phone charging station in real time, and to warn maintenance personnel when the power supply of the mobile phone charging station is abnormal; the human-computer interaction module generates a power supply data chart of the mobile phone charging station based on a visual interface; The database module stores and manages the collected data, analysis results and historical data charts, and is used for query and report generation functions to track and analyze the power supply of mobile phone charging stations over the long term.

[0006] Preferably, the connection between the sensor module and the communication module uses intelligent bridging technology to ensure the efficiency and reliability of data transmission; each sensor module is equipped with a miniaturized RS485 interface chip, which has an embedded edge computing unit to perform real-time analog-to-digital conversion on the analog signal collected by the sensor. The edge computing unit converts the analog signal into a digital signal and optimizes the conversion process through an adaptive quantization algorithm to balance data accuracy and transmission efficiency; the edge computing unit first performs a fast Fourier transform or frequency domain analysis on the analog signal collected by the sensor to determine the characteristics of the signal, such as the amplitude range and the frequency of change; based on these characteristics, the adaptive quantization algorithm dynamically adjusts the quantization level. For signals with faster changes, such as large current fluctuations, a smaller quantization step size is used to improve the accuracy of the signal; and for signals with more stable changes, such as slow changes in voltage and temperature, the quantization step size is increased, the amount of data is reduced, and the transmission efficiency is optimized; in the process of signal quantization, the noise of the signal is suppressed, and by analyzing the statistical characteristics of the digital signal, such as the mean and variance, the adaptive quantization algorithm identifies and filters out the noise components, further improving the quality of the digital signal; the converted digital signal is transmitted to the communication module via the RS485 bus.

[0007] Preferably, the edge computing unit processes analog signals from the sensor module in real time, including current, voltage and temperature; each sensor module converts the analog signal into a digital signal through its built-in digital-to-analog converter, and the edge computing unit pre-processes the data after receiving the digital signal; first, data cleaning is performed to delete redundant data and repeated information to eliminate invalid data caused by sensor fluctuations or noise; the consistency of the data is checked by verification methods, and sensor reading errors are corrected in a timely manner; the data is screened and filtered to remove outliers and low-frequency noise; after completing data cleaning and pre-processing, the edge computing unit will perform heterogeneous computing, that is, different computing methods are used according to different data types and computing requirements; for current and voltage data with high real-time requirements, the edge computing unit uses more efficient and fast algorithms for processing and analysis, such as fast Fourier transform; and for signals with relatively slow changes such as temperature, a lower frequency computing method is used to save processing resources, such as Kalman filtering; by completing these computing tasks locally, the edge computing unit can reduce the network burden of data transmission to the central processing unit, and only upload the processed data and abnormal data, thereby optimizing the use of network bandwidth and improving the overall efficiency and response speed of the system.

[0008] Preferably, the edge computing unit receives analog signals collected by the sensor module, such as current, voltage and temperature signals, and converts the analog signals into digital signals through an adaptive quantization algorithm; the adaptive quantization algorithm ensures that the analog signal maintains its original information during the conversion process by dynamically adjusting the quantization level, while reducing redundancy during data transmission; during the signal conversion process, the edge computing unit performs a fast Fourier transform on the analog signal to determine the frequency distribution of the signal; helping the system to identify the intensity of different frequency components in the signal, especially the frequency components that change dramatically; by analyzing the frequency distribution, the system determines the changing characteristics of the signal. For the signal parts with higher frequency and larger fluctuations, the system will automatically use a smaller quantization step size to ensure that the high-frequency components of the signal are captured accurately, thereby avoiding data loss and reduced accuracy; for the parts with smooth changes and lower frequencies, the system will increase the quantization step size to reduce the amount of data and improve transmission efficiency.

[0009] During the quantization process, the edge computing unit considers the statistical characteristics of the analog signal, including the mean and variance parameters; through real-time calculation of the mean and variance of the analog signal, the noise component is identified and noise suppression is performed. When the fluctuation of the analog signal exceeds a certain variance range, the noise component is identified as abnormal data and filtered to reduce the interference of noise on the quantization result; for example, in the fluctuation of the voltage signal, if the variance of the noise is too large, the system will automatically identify and reduce the quantization of the noise, thereby improving the accuracy of the final result; in order to continuously optimize the quantization, the system dynamically adjusts the strategy in the quantization process based on the error feedback between the quantized digital signal and the original digital signal; for example, if it is found that the quantization error of some digital signals is large, the system will reduce the quantization step to improve the accuracy, otherwise it will increase the quantization step to improve the transmission efficiency.

[0010] Preferably, the data analysis module receives the digital signal from the communication module, and the received digital signal is stored in the buffer of the data analysis module for data preprocessing; the preprocessed data enters the autoencoder for parsing; the autoencoder is composed of an input layer, multiple encoding layers, a bottleneck layer and multiple decoding layers, and the input layer receives data from the sensor, including current, voltage and temperature; the data is sent to the first encoding layer, which uses a piecewise linear activation function , sensitively responding to small changes in current; the output changes linearly when the current value is within the normal range, and increases or decreases rapidly when it approaches the set safety threshold to enhance sensitivity to abnormal current changes; and They are the safety threshold of current, k is a constant greater than 1. If it is not sensitive enough to current, increase the value of k, for example, increase k from 2 to 3 or higher; if it is too sensitive to current, resulting in frequent false alarms, reduce the value of k; the second encoding layer focuses on voltage data, and the adaptive activation function based on voltage history data , where μ and σ are the mean and standard deviation of the voltage data, respectively, and α is an adjustable parameter. In the training phase, the autoencoder is trained using historical data or simulated data, and the parameter α is adjusted and determined by the optimization algorithm in the process; the third encoding layer processes the temperature data and uses the activation function of the temperature gradient change Capture the rapid rise in temperature; the bottleneck layer merges the output of the encoding layer and reduces the data dimension to input the decoding layer; the structure of the decoding layer corresponds to the encoding layer, but the function is opposite, through the reverse activation function, it can accurately reconstruct the input data; when the error when reconstructing the input data exceeds the preset threshold, the system determines that the power supply of the mobile phone charging station is in an abnormal state and issues a warning through the human-computer interaction module.

[0011] Preferably, the human-computer interaction module intuitively presents the results processed by the data analysis module to the maintenance personnel, and provides real-time monitoring and early warning functions; by obtaining real-time data and historical data from the database module, the human-computer interaction module displays the power supply status of each mobile phone charging station, including real-time changes in current, voltage, and temperature; the visualization interface of the human-computer interaction module not only displays the real-time data of the current mobile phone charging station, but also performs trend analysis and prediction; for example, through regression analysis of historical current and voltage data, the system can predict the fluctuation trend of power supply in the future period of time; if the system detects abnormal trends in the power supply of some mobile phone charging stations, such as continuous increase in current, unstable voltage, etc., the human-computer interaction module reminds the maintenance personnel through significant color changes on the chart, such as red marking, and provides fault warnings to facilitate their timely intervention; when the data analysis module detects an abnormality through the autoencoder, the human-computer interaction module will immediately trigger a fault warning; the warning includes audio prompts, visual warnings, such as pop-up warning windows or changes in chart color, and SMS or email notifications.

[0012] Preferably, the database module stores, manages and retrieves all types of data related to the mobile phone charging station, including the current, voltage, temperature collected in real time, and the analysis results after processing by the autoencoder; all raw data from the sensor module, such as current, voltage, and temperature, will be stored in the database in real time; the database not only saves the real-time collected data, but also stores the abnormal data and historical fault records after analysis and processing by the autoencoder; the autoencoder conducts an in-depth analysis of the power supply of each mobile phone charging station and generates corresponding abnormal detection results; the database module uses machine learning to perform pattern recognition and trend analysis on the stored data; machine learning automatically identifies abnormal patterns that may occur in the mobile phone charging station based on the changing trends of historical data; for example, based on the historical current fluctuations, the system detects the trend of abnormal increase in current of a mobile phone charging station, and promptly marks these abnormal data. By deep learning and training the data, the system can continuously optimize the accuracy of pattern recognition and further improve fault prediction capabilities.

[0013] (III) Beneficial effects The present invention provides a mobile phone charging station status monitoring system for urban rail trains, which has the following beneficial effects: 1. The present invention integrates a miniaturized edge computing unit in each sensor module to achieve real-time analog-to-digital conversion and adaptive quantization processing of current, voltage and temperature data; it not only improves data accuracy and transmission efficiency, but also reduces the amount of data transmitted over the network and reduces the system's dependence on central processing capabilities.

[0014] 2. The present invention realizes accurate intelligent recognition and abnormality detection of the power supply status of the mobile phone charging station by combining the autoencoder with the mobile phone charging station; the autoencoder learns the data features under normal conditions, and when the working state of the mobile phone charging station deviates from these features, the system can quickly identify the abnormality, thereby improving the accuracy of fault detection.

[0015] 3. The present invention realizes in-depth analysis and real-time monitoring of data through the data analysis and human-computer interaction modules integrated in the central processing unit, provides decision support for maintenance personnel, and promotes the development of intelligent transportation systems in a more efficient and intelligent direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 This is a data collection flow chart of the mobile phone charging station of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Embodiment 1: See also Figure 1 A mobile phone charging station status monitoring system for urban rail trains realizes the monitoring of the mobile phone charging station through the following steps.

[0019] The connection between the sensor module and the communication module uses intelligent bridging technology to ensure the efficiency and reliability of data transmission. Each sensor module is equipped with a miniaturized RS485 interface chip, which has an embedded edge computing unit to perform real-time analog-to-digital conversion on the analog signals collected by the sensor. The edge computing unit converts these analog signals into digital signals and optimizes the conversion process through an adaptive quantization algorithm to balance data accuracy and transmission efficiency. The edge computing unit first performs fast Fourier transform or frequency domain analysis on the analog signals collected by the sensor to determine the characteristics of the signal, such as amplitude range and change frequency. Based on these characteristics, the adaptive quantization algorithm dynamically adjusts the quantization level. For signals with faster changes, such as large current fluctuations, a smaller quantization step size is used to improve signal accuracy. For signals with more stable changes, such as slow changes in voltage and temperature, the quantization step size is increased to reduce the amount of data and optimize transmission efficiency. In the process of signal quantization, the noise of the signal is suppressed. By analyzing the statistical characteristics of the digital signal, such as the mean and variance, the adaptive quantization algorithm identifies and filters out the noise components to further improve the quality of the digital signal. The converted digital signal is transmitted to the communication module through the RS485 bus.

[0020] The edge computing unit processes analog signals from sensor modules in real time, including current, voltage and temperature. Each sensor module converts analog signals into digital signals through its built-in digital-to-analog converter. After receiving the digital signals, the edge computing unit pre-processes the data. First, data cleaning is performed to delete redundant data and repeated information, and invalid data caused by sensor fluctuations or noise is eliminated. The consistency of the data is checked by verification methods, and sensor reading errors are corrected in a timely manner. The data is screened and filtered to remove outliers and low-frequency noise. After completing data cleaning and pre-processing, the edge computing unit will perform heterogeneous computing, that is, different computing methods are used according to different data types and computing requirements. For current and voltage data with high real-time requirements, the edge computing unit uses more efficient and fast algorithms for processing and analysis, such as fast Fourier transform. For signals that change relatively slowly, such as temperature, a lower frequency computing method is used to save processing resources, such as Kalman filtering. By completing these computing tasks locally, the edge computing unit can reduce the network burden of data transmission to the central processing unit, and only upload processed data and abnormal data, thereby optimizing the use of network bandwidth and improving the overall efficiency and response speed of the system.

[0021] The edge computing unit receives analog signals collected by the sensor module, such as current, voltage and temperature signals, and converts the analog signals into digital signals through an adaptive quantization algorithm; the adaptive quantization algorithm ensures that the analog signal maintains its original information during the conversion process by dynamically adjusting the quantization level, while reducing redundancy during data transmission; during the signal conversion process, the edge computing unit performs a fast Fourier transform on the analog signal to determine the frequency distribution of the signal; helping the system to identify the intensity of different frequency components in the signal, especially the frequency components that change dramatically; by analyzing the frequency distribution, the system determines the changing characteristics of the signal, and then provides guidance for subsequent quantization; for signal parts with higher frequencies and larger fluctuations, the system will automatically use a smaller quantization step size to ensure that the high-frequency components of the signal are captured accurately, thereby avoiding data loss and reduced accuracy; for parts with smooth changes and lower frequencies, the system will increase the quantization step size to reduce the amount of data and improve transmission efficiency.

[0022] During the quantization process, the edge computing unit considers the statistical characteristics of the analog signal, including the mean and variance parameters; through real-time calculation of the mean and variance of the analog signal, the noise component is identified and noise suppression is performed. When the fluctuation of the analog signal exceeds a certain variance range, the noise component is identified as abnormal data and filtered to reduce the interference of noise on the quantization result; for example, in the fluctuation of the voltage signal, if the variance of the noise is too large, the system will automatically identify and reduce the quantization of the noise, thereby improving the accuracy of the final result; in order to continuously optimize the quantization strategy, the system dynamically adjusts the strategy in the quantization process based on the error feedback between the quantized data and the original signal; for example, if it is found that the quantization error of some signals is large, the system will reduce the quantization step to improve the accuracy, otherwise it will increase the quantization step to improve the transmission efficiency.

[0023] The data parsing module receives the digital signal from the communication module. The received digital signal is stored in the buffer of the data parsing module for data preprocessing, including noise removal, missing value filling, and data format standardization. The preprocessed data enters the autoencoder for parsing. The autoencoder consists of an input layer, multiple encoding layers, a bottleneck layer, and multiple decoding layers. The input layer receives data from sensors, including current, voltage, and temperature. The data is sent to the first encoding layer, which uses a piecewise linear activation function. , which responds sensitively to small changes in current; this activation function outputs linear changes when the current value is within the normal range, and increases or decreases rapidly when it approaches the set safety threshold, so as to enhance the sensitivity of the model to abnormal current changes; among them, and are the safety thresholds of current, k is a constant greater than 1; the second encoding layer focuses on voltage data, and the adaptive activation function based on voltage distribution , where μ and σ are the mean and standard deviation of the voltage data, respectively, and α is an adjustable parameter; the third encoding layer processes the temperature data and uses an activation function that changes with the temperature gradient Capture the rapid rise in temperature; the bottleneck layer merges the output of the encoding layer and reduces the data dimension to input the decoding layer; the structure of the decoding layer corresponds to the encoding layer, but the function is opposite, through the reverse activation function, it can accurately reconstruct the input data; when the error in reconstructing the input data exceeds the preset threshold, the system determines that the power supply of the mobile phone charging station is in an abnormal state and triggers the early warning mechanism.

[0024] Preferably, the human-computer interaction module intuitively presents the results processed by the data analysis module to the maintenance personnel, and provides real-time monitoring and early warning functions; by obtaining real-time data and historical data from the database module, the human-computer interaction module displays the power supply status of each mobile phone charging station, including real-time changes in current, voltage, and temperature; the system converts data into an easy-to-understand graphical interface through data visualization, and uses line charts, bar charts, heat maps, etc. to dynamically display the operating status of the mobile phone charging station, helping maintenance personnel to quickly identify potential problems.

[0025] The visualization interface of the human-computer interaction module not only displays the real-time data of the current mobile phone charging station, but also performs trend analysis and prediction. For example, through regression analysis of historical current and voltage data, the system can predict the fluctuation trend of power supply in the future. If the system detects abnormal trends in the power supply of some mobile phone charging stations, such as continuous increase in current and unstable voltage, the human-computer interaction module will remind maintenance personnel through significant color changes on the chart, such as red marking, and provide fault warnings to facilitate timely intervention. When the data analysis module detects an abnormality through the autoencoder, the human-computer interaction module will immediately trigger a fault warning. The warning mechanism includes audio prompts, visual warnings, such as pop-up warning windows or changes in chart color, and SMS or email notifications. These warnings can accurately indicate the mobile phone charging station with problems and the type of abnormality, helping maintenance personnel to locate the problem as soon as possible, reducing the time for manual judgment and improving response efficiency. Maintenance personnel interact directly with the system through the human-computer interaction module to query data, generate reports and view fault records. Maintenance personnel can choose to view the data of mobile phone charging stations in a specific time period and view trend analysis charts generated by the system for later analysis and reporting.

[0026] The database module is responsible for storing, managing and retrieving all types of data related to the mobile phone charging station, including the real-time collected current, voltage, temperature, and the analysis results after processing by the autoencoder; all raw data from the sensor module, such as current, voltage, and temperature, will be stored in the database in real time and updated regularly; the database not only saves these real-time collected data, but also stores abnormal data and historical fault records after analysis and processing by the autoencoder; the autoencoder conducts in-depth analysis of the power supply of each mobile phone charging station and generates corresponding abnormal detection results; the database module uses machine learning to perform pattern recognition and trend analysis on the stored data; machine learning automatically identifies abnormal patterns that may occur in the mobile phone charging station based on the changing trends of historical data; for example, based on historical current fluctuations, the system detects a trend of abnormal increase in current in a mobile phone charging station and promptly marks these abnormal data. By deep learning and training the data, the system can continuously optimize the accuracy of pattern recognition and further improve fault prediction capabilities.

[0027] The database module also provides an efficient query interface to support maintenance personnel to quickly retrieve data according to different needs; these query interfaces can be searched according to multiple conditions such as time period, mobile phone charging station number, abnormality type, etc., to help maintenance personnel quickly obtain relevant information; maintenance personnel can choose to view data of a specific mobile phone charging station or a specific time period, and the system automatically generates trend charts, histograms and other visual charts to facilitate further analysis of the health status of the power supply.

[0028] By storing and managing this data over a long period of time, the database module not only provides maintenance personnel with tools for real-time monitoring and troubleshooting, but also provides data support for long-term tracking and in-depth analysis of the power supply status of the mobile phone charging station. The historical data and trend analysis in the database help maintenance personnel identify potential failure risks, thereby performing preventive maintenance and improving the operating efficiency and safety of the entire system.

[0029] Embodiment 2: Assume that the mobile phone charging station status monitoring system of the present invention is applied to an urban rail train line, which has a total of 30 trains, each train includes 10 carriages, each carriage is equipped with 4 mobile phone charging stations, and a total of 1,200 mobile phone charging stations; in the example application, the system is designed to monitor the current, voltage and temperature of each mobile phone charging station in real time to ensure the safe use of the mobile phone charging station by passengers, and quickly locate and notify maintenance personnel when an abnormality occurs in the equipment.

[0030] See also Figure 2, each mobile phone charging station is equipped with a sensor module for monitoring current, voltage and temperature; the sensor module is connected to the communication module, and the communication module aggregates the data of each sensor and sends it to the central processing unit through the RS485 bus; in the system deployment stage, reasonable current, voltage and temperature safety thresholds are set for the status of each mobile phone charging station to ensure that the equipment operates within a safe range; it is assumed that the current range of each mobile phone charging station under normal circumstances is set to 1.5A to 3A, the voltage range is set to 5V to 5.5V, and the temperature range is set to 20°C to 45°C.

[0031] During operation, the system will monitor the current, voltage and temperature of each mobile phone charging station in real time through the sensor module; for example, when a passenger plugs a mobile phone into the mobile phone charging station, the sensor immediately senses the current rise and starts collecting data; the collected analog signal is converted to digital through the edge computing unit built into the RS485 interface chip, and the edge computing unit uses an adaptive quantization algorithm to dynamically adjust the quantization level to optimize the accuracy and bandwidth occupancy of data transmission; assuming that the current signal of the mobile phone charging station changes dramatically at the beginning of mobile phone charging, the adaptive quantization algorithm will reduce the quantization step size to accurately capture data fluctuations; when the current stabilizes, the quantization step size is increased to reduce the amount of data; for example, when it is detected that the current of a mobile phone charging station rises to 3.2A, which is beyond the normal range, the system amplifies the signal through an adaptive quantization algorithm to ensure that the abnormal signal is accurately transmitted to the central processing unit.

[0032] After receiving the digitized current, voltage and temperature data, the central processing unit passes the data to the data analysis module; the data analysis module first pre-processes the data, removes noise and fills missing values, and standardizes the data into a format that is convenient for subsequent processing; then, the data enters the autoencoder for analysis.

[0033] The autoencoder detects data anomalies through multiple layers of encoding layers, bottleneck layers, and decoding layers. Among them, if the current value is close to 3A or exceeds 3A, the current layer of the autoencoder amplifies the current data through a piecewise linear activation function to identify the current change trend. For example, when the current value reaches 3.2A, the current layer activation function of the autoencoder will amplify it and generate an abnormal signal. For voltage data, the system performs activation processing according to its historical distribution. When the voltage exceeds 5.5V, the system will increase the sensitivity of the voltage activation layer to detect the risk of excessive voltage. For temperature changes, the autoencoder uses a secondary activation function to identify situations where the temperature rises rapidly. For example, when the temperature of the mobile phone charging station exceeds 45°C, the activation function will rapidly increase the output value and generate an abnormal signal.

[0034] When the decoding layer reconstructs the input data, if the reconstruction error is greater than the preset threshold, an abnormal warning is triggered; assuming that the reconstruction error threshold set by the system is 0.1, when the current of a mobile phone charging station soars to 3.5A due to a fault and exceeds the normal range, the reconstruction error of the decoding layer will far exceed the threshold, and the system will immediately judge it as an abnormality and trigger an alarm.

[0035] Once the system detects an abnormal state, the human-computer interaction module will show the maintenance personnel the location of the abnormal mobile phone charging station, the type of abnormality and the specific data detected, such as current 3.5A, voltage 5.6V, and temperature 47°C through a visual interface; the interface will present charts of real-time and historical data to help maintenance personnel analyze the failure trend of the equipment; when an abnormality occurs, the system will issue a warning notification and store the information in the database module for subsequent analysis.

[0036] The database module records the data of all mobile phone charging stations, including real-time data of current, voltage, temperature and abnormal events detected by the system. These data are used to generate reports to help the maintenance team track the health status of the equipment and identify possible failure modes. For example, by performing pattern recognition on the data through machine learning algorithms, the database module analyzes the reasons why the current of some mobile phone charging stations increases abnormally during peak hours, such as the evening commuting peak, helping to prevent possible failures in advance.

[0037] Through the present invention, the system can efficiently and real-time monitor 1,200 mobile phone charging stations in urban rail trains, ensuring the safety of each mobile phone charging station; when an abnormal increase in current occurs at a mobile phone charging station, the system can detect and generate an alarm message within 1 second, allowing maintenance personnel to respond quickly and ensure the stable operation of the mobile phone charging station.

[0038] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A mobile phone charging station status monitoring system for urban rail trains, characterized in that: include: Mobile phone charging stations are installed in each carriage of the urban rail train. Each mobile phone charging station is equipped with a sensor module to collect the current, voltage and temperature data of the mobile phone charging station in real time; The sensor module is connected to the communication module via the RS485 bus, which is used to receive the data collected by the sensor module and convert it into a digital signal; The communication module communicates with the central processing unit through a network switch and transmits the converted data to the central processing unit to realize the control of the power supply of the mobile phone charging station; The central processing unit integrates the mobile phone charging station power supply monitoring system, including data analysis module, human-computer interaction module and database module, among which: The data analysis module receives and stores the digital signal transmitted from the communication module and pre-processes the received data; the data analysis module identifies the digital signal through the autoencoder to determine the normal working state of the power supply of the mobile phone charging station, thereby realizing early warning of abnormal conditions through the human-computer interaction module; The human-computer interaction module is used to display the power supply information of each mobile phone charging station in real time, and to warn maintenance personnel when the power supply of the mobile phone charging station is abnormal; the human-computer interaction module generates a power supply data chart of the mobile phone charging station based on a visual interface; The database module stores and manages the collected data, analysis results and historical data charts, and is used for query and report generation functions to track and analyze the power supply of mobile phone charging stations over the long term.

2. A mobile phone charging station status monitoring system for urban rail trains according to claim 1, characterized in that: The connection between the sensor module and the communication module uses intelligent bridging technology. Each sensor module is equipped with a miniaturized RS485 interface chip, which has a built-in miniature edge computing unit. The edge computing unit performs real-time analog-to-digital conversion on the data collected by the sensor and converts the analog signal into a digital signal. During the conversion process, the edge computing unit uses an adaptive quantization algorithm to adjust the quantization level according to the characteristics of the analog signal. The converted digital signal is transmitted to the communication module via the RS485 bus.

3. A mobile phone charging station status monitoring system for urban rail trains according to claim 2, characterized in that: The microcontroller unit in the edge computing unit preprocesses the digital signal, and the preprocessing includes data cleaning, deleting duplicate information and correcting errors; it also includes data screening and filtering. After data cleaning and preprocessing, the edge computing unit performs heterogeneous computing to adapt to different computing needs; Edge computing units receive data and process it, reducing the network requirements for large amounts of data transmission.

4. A mobile phone charging station status monitoring system for urban rail trains according to claim 2, characterized in that: The adaptive quantization algorithm dynamically adjusts the quantization level to optimize the accuracy and transmission efficiency of the data, and determines the degree of quantization refinement by analyzing the characteristics of the analog signal in real time, including the amplitude range and the frequency of change; The adaptive quantization algorithm first performs a fast Fourier transform on the analog signal to determine its frequency distribution, and then adjusts the quantization step size according to the changing frequency and amplitude of the analog signal.

5. The urban rail train mobile phone charging station status monitoring system according to claim 1 is characterized by: The data analysis module receives the digital signal from the communication module, and the received digital signal is stored in the buffer of the data analysis module, and performs data preprocessing, including removing noise, filling missing values, and standardizing data format; The preprocessed data enters the autoencoder for analysis.

6. A mobile phone charging station status monitoring system for urban rail trains according to claim 5, characterized in that: The autoencoder consists of an input layer, multiple encoding layers, a bottleneck layer, and multiple decoding layers. The input layer receives data from sensors, including current, voltage, and temperature. The data is fed into the first encoding layer, which uses a piecewise linear activation function. , responds sensitively to small changes in current, outputs linear changes when the current value is within the normal range, and rapidly increases or decreases when it approaches the set safety threshold, so as to enhance the model's sensitivity to abnormal current changes; and They are the safety thresholds of current, k is a constant greater than 1; the second encoding layer focuses on voltage data, based on the voltage-based adaptive activation function Identify abnormal changes in voltage, where μ and σ are the mean and standard deviation of the voltage data, respectively, and α is an adjustable parameter; the third encoding layer processes temperature data and uses an activation function that changes in temperature gradient Capture the rapid rise in temperature; the bottleneck layer merges the output of the encoding layer and reduces the data dimension to input the decoding layer; the structure of the decoding layer corresponds to the encoding layer, but the function is opposite, through the reverse activation function, it can accurately reconstruct the input data; when the error in reconstructing the input data exceeds the preset threshold, the system determines that the power supply of the mobile phone charging station is in an abnormal state, and triggers an early warning through the human-computer interaction module.

7. A mobile phone charging station status monitoring system for urban rail trains according to claim 6, characterized in that: The human-computer interaction module intuitively presents the results processed by the data analysis module to the maintenance personnel and provides real-time monitoring and early warning; the historical and real-time data of the mobile phone charging station are obtained from the database module, and then the data is converted into a graphical interface through data visualization to display the current power supply status of the mobile phone charging station, including trend analysis and prediction, to help maintenance personnel quickly identify potential problems; When the data analysis module detects an abnormality, the human-computer interaction module immediately triggers a fault warning to alert maintenance personnel.

8. The urban rail train mobile phone charging station status monitoring system according to claim 1 is characterized by: The database module is responsible for storing, managing and retrieving all data related to the mobile phone charging station; Stores real-time collected data and the analysis results of the autoencoder. The database uses machine learning to perform pattern recognition and trend analysis on the stored data and automatically marks abnormal data analyzed by the autoencoder. The database module provides a query interface to support maintenance personnel to quickly retrieve data and generate customized reports and analysis charts, thereby achieving long-term tracking and in-depth analysis of the power supply status of the mobile phone charging station.

Images