Real-time monitoring system and method for traction force and conversion resistance of point switch, electronic equipment and storage medium

Abstract

The invention discloses a real-time monitoring system and method for traction force and conversion resistance of a point switch, electronic equipment and a storage medium. The monitoring system comprises at least one force transducer, at least one acquisition extension set and at least one monitoring host, the force measuring sensors are used for sensing traction force data and conversion resistance data of the corresponding point switches; the acquisition extension set is used for acquiring traction data and conversion resistance data of the point switches sensed by the force sensors, filtering and analyzing the acquired data, and judging whether the point switches are abnormal or not; the monitoring host is used for monitoring the traction data and the conversion resistance data of the point switches acquired by the acquisition extensions, and risk prediction is carried out on the point switches according to changes of the traction data and the conversion resistance data of the point switches. The state of traction force and conversion resistance of the switch machine can be obtained in real time, the debugging work difficulty and time are reduced, and the work efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of electronic information technology, and relates to a monitoring system, and in particular to a real-time monitoring system, method, electronic equipment and storage medium for traction and conversion resistance of a switch machine. Background Art

[0002] The operating status of railway turnout switching equipment is one of the main issues affecting railway transportation safety and efficiency. The switch machine is the executive equipment of the turnout control system. Therefore, whether the switch machine can work normally directly affects the switch conversion and affects the safety and efficiency of train operation. With the overall speed increase of the main railway lines and the construction and development of high-speed railways, a lot of manpower and material resources have been invested in the maintenance time and workload of the switch machine on site. In addition, both high-speed and conventional railway lines require skylights to work on the track. In particular, the maintenance skylights of the equipment on high-speed lines are all operated at night, which brings certain difficulties to the on-site maintenance personnel.

[0003] The traction and conversion resistance of the switch machine will be affected by improper adjustment, climate influence, and line changes. The current situation is that the traction and conversion resistance of the switch machine cannot be mastered on a daily basis, and it is necessary to use a portable switch machine turnout parameter tensile tester to test it on the track in the skylight point to understand it. In order to solve the problem that during the maintenance of the switch machine on site, the bolts of the action rod and the turnout action connecting rod (close adjustment rod) are disassembled on the track, and a pin-type sensor is installed to test the traction of the switch machine, which not only affects the normal use of the equipment, but also requires maintenance of the skylight and daily mastery of the traction and conversion resistance of the switch machine. We chose to study the automatic real-time monitoring of the traction and conversion resistance of the switch machine, and use scientific and technological means to ensure that the conversion of the switch machine is in a reliably controlled state.

[0004] In view of this, there is an urgent need to design a new switch machine monitoring method to overcome at least some of the above-mentioned defects of the existing switch machine monitoring method. Summary of the invention

[0005] The present invention provides a real-time monitoring system, method, electronic equipment and storage medium for the traction force and conversion resistance of a switch machine, which can obtain the status of the traction force and conversion resistance of the switch machine in real time, reduce the difficulty and time of debugging work, and improve work efficiency.

[0006] To solve the above technical problem, according to one aspect of the present invention, the following technical solution is adopted:

[0007] A real-time monitoring system for traction and conversion resistance of a switch machine, the monitoring system comprising:

[0008] At least one force sensor is arranged at the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine;

[0009] At least one data collection extension connected to at least one force sensor for collecting the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and filtering and analyzing the collected data to determine whether each switch machine has an abnormality;

[0010] At least one monitoring host is connected to at least one collecting extension to monitor the traction force data and conversion resistance data of the switch machines collected by each collecting extension, and to perform risk prediction on each switch machine based on the changes in the traction force data and conversion resistance data of each switch machine.

[0011] As an embodiment of the present invention, the monitoring system further comprises: a switch machine fault analysis model construction module and an analysis model training module, wherein the switch machine fault analysis model construction module is used to construct a switch machine fault analysis model, and the analysis model training module is used to train the switch machine fault analysis model;

[0012] The switch machine fault analysis model building module is used to extract features from historical data, extract the change process data of the switch machine's starting resistance, intermediate process resistance and locking resistance within a set time before the switch machine fails, and the switch machine fault characteristic data, and establish a mathematical model of the association between the change process data of the switch machine's starting resistance, intermediate process resistance and locking resistance and the switch machine fault characteristic data; wherein the switch machine fault characteristic data includes the time from the switch machine failure to the change of the switch machine's starting resistance or / and intermediate process resistance or / and locking resistance within the set time reaching a set threshold;

[0013] The monitoring host performs risk prediction on each switch machine through the switch machine fault analysis model constructed by the switch machine fault analysis model construction module, and determines whether the switch machine is in danger of failure based on the change process data of the starting resistance, intermediate process resistance and locking resistance of the switch machine sensed by each force sensor.

[0014] As an embodiment of the present invention, the monitoring system further comprises:

[0015] The server is connected to at least one monitoring host, and is used to receive the traction force data and conversion resistance data of the switch machine sent by each monitoring host, and receive the risk prediction data of the switch machine sent by the monitoring host;

[0016] At least one terminal, connected to the server, for receiving the fault information and risk prediction information of the switch machine sent by the server;

[0017] The monitoring system further includes at least one communication transmission module, each communication transmission module collects data processed by at least one acquisition extension and uploads the data to the monitoring host; the force sensor is arranged at the action connecting rod corresponding to the switch machine.

[0018] As an embodiment of the present invention, the acquisition sub-machine includes a data processing module for processing the traction force data and conversion resistance data sensed by the force sensor;

[0019] The data processing module comprises:

[0020] A data acquisition unit is used to acquire the traction force data and conversion resistance data of the switch machine in real time from the force measuring sensor; the traction force data and conversion resistance data include force value, state, data type and possible switching data;

[0021] The data processing unit is used to clean, convert and analyze the collected raw data; the mathematical formula for the traction force and conversion resistance value of the switch machine is expressed as: Where N is the number of observations, CJZ i is the true value of the i-th observation data, CJZ ^ i is the corresponding estimated value; QTZ refers to a specific parameter or variable, i.e., a value after the traction force of the switch machine is converted into resistance;

[0022] The data encoding unit is used to convert the processed data into a format suitable for transmission; the corresponding encoding method is determined according to the data type, state and force value;

[0023] The data transmission unit is used to send the encoded data to the host computer or server for further analysis and storage through the communication protocol. During the transmission process, the integrity and accuracy of the data must be ensured. After receiving the data, the monitoring host or server will decode and analyze it. According to the type and status of the data, corresponding processing will be performed. At the same time, the monitoring system also needs to have error handling and early warning mechanisms. When erroneous data is received, the system should be able to identify and take corresponding processing measures, such as re-requesting data or issuing an alarm. The early warning mechanism can promptly alarm abnormal data according to the preset threshold to ensure the safety and stability of railway operation.

[0024] As an embodiment of the present invention, the monitoring host includes a risk prediction module, and the risk prediction module is used to predict the risk of the switch machine;

[0025] The risk prediction module includes:

[0026] A data collection unit for collecting traction data and conversion resistance data within a set time, which are daily, hourly or minute data points;

[0027] The data preprocessing unit is used to preprocess the data after it is collected, including removing outliers and filling missing values ​​to ensure the accuracy and consistency of the data;

[0028] Trend analysis unit, which uses statistical methods or machine learning algorithms to analyze historical data and identify trends in force values; including seasonal trends, cyclical trends, or long-term growth / decrease trends;

[0029] Trend prediction unit, used to predict future force trend based on the analysis results of historical data using appropriate prediction algorithms;

[0030] The monitoring host includes a data monitoring module, and the data monitoring module is used to monitor whether there is a risk in the traction data and conversion resistance data of the switch machine;

[0031] The alarm determination formula of the data monitoring module is expressed as: Among them, Sjz is the actual switch machine traction force and conversion resistance data, which is the size of the switch machine traction force and conversion resistance measured in real time by the force sensor; bhl is the rate of change of the switch machine traction force and conversion resistance, which indicates how fast the switch machine traction force and conversion resistance values ​​change over a period of time, which helps to predict whether the switch machine traction force and conversion resistance may deteriorate rapidly; hjcs is the environmental parameters including temperature and humidity, which may affect the stability of the switch machine traction force and conversion resistance.

[0032] As an embodiment of the present invention, the monitoring host includes an offset range prediction module for predicting possible offsets of the traction force and conversion resistance of the turnout machine;

[0033] In the railway turnout switch traction and conversion resistance monitoring system, the offset range prediction algorithm is used to improve the accuracy of the switch traction and conversion resistance position detection; the possible offsets of the turnout switch traction and conversion resistance are divided into multiple intervals, and the intervals to which the switch traction and conversion resistance offsets belong are predicted, and finally the specific switch traction and conversion resistance positions are determined by calculating the expected offsets within the intervals;

[0034] The prediction process of the offset range prediction module includes:

[0035] Offset interval division: Based on historical data and system requirements, the possible offset range of turnout traction and conversion resistance is divided into several intervals;

[0036] Interval classification prediction: Assume that the probability that the classifier predicts that the target offset belongs to the i-th interval is p_i; train a classifier, use the extracted features as input, and predict the interval to which the traction force and conversion resistance offset of the switch machine belong;

[0037] Calculation of expected offset: For each interval, the expected value or average value of all possible offsets in the interval is calculated based on historical data or system requirements; when the classifier predicts the interval to which the traction force and conversion resistance offset of the switch machine belongs, the corresponding interval is selected and the expected offset of the interval is used as the final prediction value;

[0038] Interval classification prediction: Assume that the classifier predicts that the probability that the traction force and conversion resistance offset of the switch machine belongs to the i-th interval is p_i; where i = 1, 2, ..., n, and n is the number of intervals;

[0039] Calculation of expected offsets: For each interval i, calculate the expected E_i of all offsets within it.

[0040] As an embodiment of the present invention, the monitoring host includes a curve generating module, and the curve generating module is used to generate a curve of the traction force data and the conversion resistance data of the turnout machine;

[0041] The generation process of the curve generation module includes:

[0042] Data collection and preprocessing steps: Collect the traction and conversion resistance displacement data of the switch machine from the monitoring equipment, usually including the timestamp and the corresponding displacement value; remove outliers in the data and fill in missing values ​​to ensure the accuracy and consistency of the data;

[0043] Data analysis and modeling steps: The traction force and conversion resistance displacement data of the switch machine are regarded as time series data, and statistical methods are used to analyze their trends and periodicity; let D_t be the traction force and conversion resistance displacement value of the switch machine at time t, and N is the window size of the moving average; then the moving average displacement value OL at time t is calculated by the following formula:

[0044] Trend prediction step: Based on historical data, use linear regression, machine learning or deep learning models to predict future switch machine traction and conversion resistance displacement trends.

[0045] According to another aspect of the present invention, the following technical solution is adopted: a real-time monitoring method for traction force and conversion resistance of a switch machine, the monitoring method comprising:

[0046] The force sensor is arranged on the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine;

[0047] The collection extension is connected to at least one force sensor, collects the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and filters and analyzes the collected data to determine whether each switch machine has an abnormality;

[0048] The monitoring host is connected to at least one collection extension, monitors the traction data and conversion resistance data of the switch machines collected by each collection extension, and performs risk prediction on each switch machine based on the changes in the traction data and conversion resistance data of each switch machine.

[0049] According to another aspect of the present invention, the following technical solution is adopted: an electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0050] According to another aspect of the present invention, the following technical solution is adopted: a storage medium stores computer program instructions, and the computer program instructions implement the steps of the above method when executed by a processor.

[0051] The beneficial effects of the present invention are as follows: the system, method, electronic device and storage medium for real-time monitoring of the traction force and conversion resistance of a switch machine proposed in the present invention can obtain the status of the traction force and conversion resistance of the switch machine in real time, reduce the difficulty and time of debugging work, and improve work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The figure is a schematic diagram of the composition of a real-time monitoring system for traction force and conversion resistance of a switch machine in one embodiment of the present invention.

[0053] Figure 2 Flow chart of a real-time monitoring method in one embodiment of the present invention.

[0054] Figure 3 FIG. 4 is a schematic diagram of the composition of an electronic device in one embodiment of the present invention. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0056] In order to further understand the present invention, preferred embodiments of the present invention are described below in conjunction with examples. However, it should be understood that these descriptions are only for further illustrating the features and advantages of the present invention, rather than limiting the claims of the present invention.

[0057] The description in this section is only for several typical embodiments, and the present invention is not limited to the scope of the embodiments. The same or similar prior art means and some technical features in the embodiments are mutually replaced within the scope of the present invention.

[0058] The description of the steps in each embodiment in the specification is only for the convenience of explanation, and the implementation method of the present application is not limited by the order of implementation of the steps.

[0059] The term “connection” in the specification includes both direct connection and indirect connection.

[0060] The present invention discloses a real-time monitoring system for traction force and conversion resistance of a switch machine. Figure 1 This is a schematic diagram of the composition of a real-time monitoring system for traction and conversion resistance of a switch machine in one embodiment of the present invention; please refer to Figure 1 The monitoring system includes: at least one force sensor 3, at least one acquisition extension 2 and at least one monitoring host 1.

[0061] Each force sensor 3 is arranged at a corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine;

[0062] Each collection sub-machine 2 is connected to at least one force sensor to collect the traction force data and conversion resistance data of the switch machine sensed by each force sensor 3, and filter and analyze the collected data to determine whether each switch machine has an abnormality;

[0063] Each monitoring host 1 is connected to at least one collection extension 2 to monitor the traction force data and conversion resistance data of the switch machines collected by each collection extension, and to perform risk prediction on each switch machine based on the changes in the traction force data and conversion resistance data of each switch machine.

[0064] In one embodiment of the present invention, the monitoring system further comprises: a switch machine fault analysis model construction module and an analysis model training module, wherein the switch machine fault analysis model construction module is used to construct a switch machine fault analysis model, and the analysis model training module is used to train the switch machine fault analysis model;

[0065] The switch machine fault analysis model building module is used to extract features from historical data, extract the change process data of the switch machine's starting resistance, intermediate process resistance and locking resistance within a set time before the switch machine fails, and the switch machine fault characteristic data, and establish a mathematical model of the association between the change process data of the switch machine's starting resistance, intermediate process resistance and locking resistance and the switch machine fault characteristic data; wherein the switch machine fault characteristic data includes the time from the switch machine failure to the change of the switch machine's starting resistance or / and intermediate process resistance or / and locking resistance within the set time reaching a set threshold;

[0066] The monitoring host performs risk prediction on each switch machine through the switch machine fault analysis model constructed by the switch machine fault analysis model construction module, and determines whether the switch machine is in danger of failure based on the change process data of the starting resistance, intermediate process resistance and locking resistance of the switch machine sensed by each force sensor.

[0067] In one embodiment, the monitoring system further includes: a server and at least one terminal; the server is connected to at least one monitoring host to receive the switch machine traction data and conversion resistance data sent by each monitoring host, and receive the switch machine risk prediction data sent by the monitoring host; the terminal is connected to the server to receive the switch machine fault information and risk prediction information sent by the server.

[0068] The monitoring system further includes at least one communication transmission module, each communication transmission module collects data processed by at least one acquisition extension and uploads the data to the monitoring host; the force sensor is arranged at the action connecting rod corresponding to the switch machine.

[0069] The acquisition sub-machine may include a data processing module for processing the traction data and conversion resistance data sensed by the force sensor.

[0070] The data processing module includes: a data acquisition unit, a data processing unit, a data encoding unit and a data transmission unit.

[0071] The data acquisition unit is used to collect the traction data and conversion resistance data of the switch machine from the force measuring sensor in real time; the traction data and conversion resistance data include force value, state, data type and possible pulling data;

[0072] The data processing unit is used to clean, convert and analyze the collected raw data; the mathematical formula for the traction force and conversion resistance value of the switch machine is expressed as: Where N is the number of observations, CJZ i is the true value of the i-th observation data, CJZ ^ i is the corresponding estimated value; QTZ refers to a specific parameter or variable, i.e., a value after the traction force of the switch machine is converted into resistance; this usually involves optimizing the parameters in f(x);

[0073] The data encoding unit is used to convert the processed data into a format suitable for transmission; the corresponding encoding method is determined according to the data type, state and force value;

[0074] The data transmission unit is used to send the encoded data to the host computer or server through the communication protocol for further analysis and storage; during the transmission process, the integrity and accuracy of the data must be ensured; after receiving the data, the monitoring host or server will decode and analyze it; according to the type and status of the data, corresponding processing will be performed; for example, for the switching data, the number of switching times and the changes in the traction and conversion resistance of the switch machine after each switching will be analyzed; at the same time, the monitoring system also needs to have an error handling and early warning mechanism; when receiving erroneous data, the system should be able to identify and take corresponding processing measures, such as re-requesting data or issuing an alarm; the early warning mechanism can promptly alarm abnormal data according to the preset threshold to ensure the safety and stability of railway operation.

[0075] The monitoring host may include a risk prediction module, which is used to predict the risk of the switch machine. The risk prediction module includes: a data collection unit, a data preprocessing unit, a trend analysis unit and a trend prediction unit.

[0076] The data collection unit is used to collect traction data and conversion resistance data within a set time, which are daily, hourly or minute data points;

[0077] The data preprocessing unit is used to preprocess the data after it is collected, including removing outliers and filling missing values ​​to ensure the accuracy and consistency of the data;

[0078] The trend analysis unit uses statistical methods or machine learning algorithms to analyze historical data and identify trends in force values; including seasonal trends, cyclical trends, or long-term growth / decrease trends;

[0079] The trend prediction unit is used to predict future force value trends based on the analysis results of historical data using appropriate prediction algorithms.

[0080] The monitoring host may include a data monitoring module, which is used to monitor whether there is a risk in the traction data and conversion resistance data of the switch machine. The alarm determination formula of the data monitoring module is expressed as: Among them, Sjz is the actual switch machine traction force and conversion resistance data, which is the size of the switch machine traction force and conversion resistance measured in real time by the force sensor; bhl is the rate of change of the switch machine traction force and conversion resistance, which indicates how fast the switch machine traction force and conversion resistance values ​​change over a period of time, which helps to predict whether the switch machine traction force and conversion resistance may deteriorate rapidly; hjcs is the environmental parameters including temperature and humidity, which may affect the stability of the switch machine traction force and conversion resistance.

[0081] The monitoring host may also include an offset range prediction module to predict the possible offset of the traction force and conversion resistance of the turnout switch. In the railway turnout switch traction force and conversion resistance monitoring system, the offset range prediction algorithm is used to improve the accuracy of the detection of the traction force and conversion resistance position of the turnout switch; the possible offset of the traction force and conversion resistance of the turnout switch is divided into multiple intervals, and the interval to which the traction force and conversion resistance offset of the turnout switch belongs is predicted, and finally the specific traction force and conversion resistance position of the turnout switch is determined by calculating the expected offset within the interval.

[0082] The prediction process of the offset range prediction module includes:

[0083] Offset interval division: Based on historical data and system requirements, the possible offset range of turnout traction and conversion resistance is divided into several intervals;

[0084] Interval classification prediction: Assume that the probability that the classifier predicts that the target offset belongs to the i-th interval is p_i; train a classifier, use the extracted features as input, and predict the interval to which the traction force and conversion resistance offset of the switch machine belong;

[0085] Calculation of expected offset: For each interval, the expected value or average value of all possible offsets in the interval is calculated based on historical data or system requirements; when the classifier predicts the interval to which the traction force and conversion resistance offset of the switch machine belongs, the corresponding interval is selected and the expected offset of the interval is used as the final prediction value;

[0086] Interval classification prediction: Assume that the classifier predicts that the probability that the traction force and conversion resistance offset of the switch machine belongs to the i-th interval is p_i; where i = 1, 2, ..., n, and n is the number of intervals;

[0087] Calculation of expected offsets: For each interval i, calculate the expected E_i of all offsets within it.

[0088] In one embodiment of the present invention, the monitoring host further comprises a curve generating module, and the curve generating module is used to generate a curve of the traction force data and the conversion resistance data of the turnout machine.

[0089] The generation process of the curve generation module includes:

[0090] Data collection and preprocessing steps: Collect the traction and conversion resistance displacement data of the switch machine from the monitoring equipment, usually including the timestamp and the corresponding displacement value; remove outliers in the data and fill in missing values ​​to ensure the accuracy and consistency of the data;

[0091] Data analysis and modeling steps: The traction force and conversion resistance displacement data of the switch machine are regarded as time series data, and statistical methods are used to analyze their trends and periodicity; let D_t be the traction force and conversion resistance displacement value of the switch machine at time t, and N is the window size of the moving average; then the moving average displacement value OL at time t is calculated by the following formula:

[0092] Trend prediction step: Based on historical data, use linear regression, machine learning or deep learning models to predict future switch machine traction and conversion resistance displacement trends.

[0093] The present invention further discloses a real-time monitoring method for traction force and conversion resistance of a switch machine, the monitoring method comprising:

[0094] [Step S1] A force sensor is disposed on the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine;

[0095] [Step S2] The collection extension is connected to at least one force sensor to collect the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and the collected data is filtered and analyzed to determine whether each switch machine has an abnormality;

[0096] [Step S3] The monitoring host is connected to at least one collection extension, monitors the traction data and conversion resistance data of the switch machines collected by each collection extension, and performs risk prediction for each switch machine based on changes in the traction data and conversion resistance data of each switch machine.

[0097] The force sensor is a resistance strain sensor, which is a functional element that uses a metal elastic body as a force conversion into strain. It is installed on the external locking action connecting rod of the ZD(J)9, S700K, and ZYJ7 switch machine installation device; it is installed on the internal locking close-fitting adjustment rod (close to the switch machine end) of the ZD(J)9, ZD6, and ZYJ switch machine installation device. One force sensor is installed on the action connecting rod (close-fitting rod) of each switch machine.

[0098] The data collection extension uses RS485 communication to connect to the resistance strain gauge force sensor and is responsible for data collection and analysis. It is installed in the HZ24 junction box next to the switch machine. One data collection extension is installed on each switch machine.

[0099] The wireless handheld device communicates wirelessly with the data collection extension and is used for on-site maintenance to monitor and display the data and graphics of the traction and conversion resistance of the switch machine in real time. Each station is equipped with a handheld device as needed.

[0100] Communication transmission equipment, collects data from multiple acquisition extensions and uploads the data to the monitoring host. One per channel, each channel can connect 30 acquisition extensions, and is placed in the monitoring host cabinet in the station signal machinery room.

[0101] The monitoring host has the functions of background processing and data analysis, displaying various parameter data of the switch machine, collecting extension maintenance and user maintenance, issuing instructions, data processing, trend analysis, station management, displaying the traction force and conversion resistance curve of the switch machine, providing alarm function and trend analysis data, and can be networked with the centralized signal monitoring. One unit is installed at each station, which can connect to 4 communication transmission equipment and is placed in the monitoring host cabinet of the station signal machinery room.

[0102] For the power supply, a 50Hz online UPS power supply is installed, which has the functions of power safety isolation, voltage stabilization and power failure data storage. One unit is installed at each station, and is placed in the monitoring host cabinet of the station signal machinery room.

[0103] 1. Data processing and communication of switch machine traction and conversion resistance monitoring

[0104] The processing and communication of switch machine traction and conversion resistance monitoring data involves multiple links, including data collection, processing, encoding and transmission; these links use a series of algorithms and formulas to ensure data accuracy and transmission efficiency; the following is a detailed introduction to the algorithms and mathematical formulas related to the processing and communication of switch machine traction and conversion resistance monitoring data:

[0105] Data collection is the first step of the monitoring system. Sensors are used to obtain various data on the traction and switching resistance of the switch machine in real time. These data include force value, state, data type, and possible switching data.

[0106] Data processing is the process of cleaning, converting and analyzing the collected raw data; the mathematical formula for the traction force and conversion resistance value of the switch machine is expressed as:

[0107]

[0108] Where N is the number of observations, CJZ i is the true value of the i-th observation data, CJZ ^ i is the corresponding estimated value; QTZ refers to a specific parameter or variable, i.e., a value after the traction force of the switch machine is converted into resistance. This usually involves optimizing the parameters in f(x).

[0109] Data encoding is the process of converting processed data into a format suitable for transmission. Depending on the data type, state, and force value, the encoding method will also be different; for example, for the pulling data, it is necessary to record the number of pulls and the 10-second data after each pull;

[0110] Data transmission is to send the encoded data to the host computer or server through the communication protocol for further analysis and storage; during the transmission process, the integrity and accuracy of the data need to be ensured;

[0111] After receiving the data, the host computer or server will decode and analyze it; according to the type and status of the data, it will perform corresponding processing; for example, for the switching data, the number of switching times and the changes in the traction force and conversion resistance of the switch machine after each switching will be analyzed;

[0112] At the same time, the monitoring system also needs to have error handling and early warning mechanisms; when receiving erroneous data, the system should be able to identify and take appropriate measures, such as re-requesting data or issuing an alarm; the early warning mechanism can promptly alarm abnormal data according to preset thresholds to ensure the safety and stability of railway operations;

[0113] 2. The trend prediction formula of the traction force and conversion resistance value of the switch machine:

[0114] In the switch machine traction and conversion resistance monitoring system, the prediction of force trend is the key to ensure that the system can detect potential problems in advance and take measures. The following is a description of a force trend prediction algorithm based on historical data, with the corresponding formula attached:

[0115] 1. Algorithm Description

[0116] a. Data Collection

[0117] First, the system needs to collect force data over a period of time, which can be daily, hourly, or minute-by-minute data points, depending on the accuracy and needs of the monitoring.

[0118] b. Data preprocessing

[0119] After the data is collected, it needs to be preprocessed, including removing outliers, filling missing values, etc., to ensure the accuracy and consistency of the data.

[0120] c. Trend analysis

[0121] Use statistical methods or machine learning algorithms to analyze historical data and identify trends in force values. This can include seasonal trends, cyclical trends, or long-term growth / decrease trends.

[0122] d. Trend prediction

[0123] Based on the analysis results of historical data, an appropriate forecasting algorithm is used to predict future force value trends. This can be a simple linear forecasting model or a more complex machine learning model.

[0124] 2. Formula description

[0125] A prediction formula model based on linear regression.

[0126] Assume that we have n historical data points, each of which contains the time tr and the corresponding switch machine traction and conversion resistance displacement dz. The linear regression model attempts to find a straight line that minimizes the sum of the distances from all data points to this straight line.

[0127] The formula of the linear regression model can be expressed as:

[0128] d_t=α+β*tr

[0129] in:

[0130] d_t is the predicted value of force at time tr;

[0131] α is the intercept term, which represents the traction force and conversion resistance displacement value of the switch machine when tr = 0 (it may not make sense in actual situations, but it is part of the model);

[0132] β is the slope term, which indicates the change in the traction force and the conversion resistance displacement dz of the switch machine when the time tr increases by 1 unit;

[0133] To find the best values ​​of α and β, we usually fit the model using the method of least squares, which involves solving the following system of equations:

[0134] Σ(d_t-α-β*tr)^2=minimum value

[0135] Here, Σ represents the sum of all n data points.

[0136] By solving this set of equations, the optimal α and β values ​​can be found, and then these values ​​can be used to predict future switch machine traction and conversion resistance displacement trends; the quality and quantity of data have a great impact on the accuracy of the prediction, so it is very important to ensure the accuracy and completeness of the data; the prediction is only an estimate of the future, and the actual results may be different, so it needs to be adjusted and corrected according to the actual situation.

[0137] 3. The formula and mechanism of traction and conversion resistance of switch machine

[0138] Switch machine traction and switching resistance monitoring

[0139] First, the traction force and conversion resistance of the switch machine are monitored in real time through sensors installed on the switch machine; these sensors can measure the actual size of the traction force and conversion resistance of the switch machine with high accuracy and transmit the data to the central processing unit;

[0140] Alarm threshold setting

[0141] The alarm threshold is set based on railway safety standards and historical data analysis. Usually, this threshold takes into account the following factors:

[0142] Safety margin: ensuring that there is enough time to intervene before the traction and switching resistance values ​​of the switch machine reach dangerous levels.

[0143] Historical data: Analyze the changing trends of the traction force and conversion resistance of the switch machine over a period of time to determine the reasonable alarm threshold;

[0144] Environmental factors: Consider the impact of environmental factors such as temperature and humidity on the traction and conversion resistance of the switch machine;

[0145] Alarm judgment formula

[0146] The alarm determination is not only based on the single traction force and conversion resistance value of the switch machine, but also combines the change rate of the traction force and conversion resistance value of the switch machine and other related parameters; the specific alarm determination formula can be expressed as:

[0147]

[0148] in:

[0149] Sjz is the actual traction force and conversion resistance value of the switch machine: the traction force and conversion resistance of the switch machine measured in real time by the sensor;

[0150] BHL switch traction and conversion resistance change rate: indicates how fast the switch traction and conversion resistance values ​​change over a period of time, which helps predict whether the switch traction and conversion resistance may deteriorate rapidly;

[0151] hjcs environmental parameters: including temperature, humidity, etc., which may affect the stability of the switch machine's traction and conversion resistance.

[0152] Multi-level alarm system

[0153] In addition to the basic alarm thresholds, you can also set up a multi-level alarm system so that different responses can be taken in situations of different severity. For example:

[0154] Level 1 alarm: It is triggered when the actual switch machine traction and conversion resistance values ​​are close to but have not exceeded the alarm threshold, alerting the staff to pay attention and possibly perform preventive maintenance.

[0155] Second level alarm: triggered when the actual switch machine traction and conversion resistance values ​​exceed the alarm threshold, immediate measures need to be taken to prevent accidents.

[0156] Level 3 alarm: Based on the level 2 alarm, if the traction and conversion resistance values ​​of the switch machine continue to deteriorate rapidly, triggering a higher level alarm, emergency measures such as emergency stopping may be required.

[0157] Alarm response mechanism

[0158] Once an alarm is triggered, the system will automatically execute a series of response measures, including but not limited to:

[0159] Send alarm information to relevant personnel, including location, switch machine traction and conversion resistance values ​​and recommended countermeasures.

[0160] Display alarm information in the control center so that managers can make decisions quickly.

[0161] Automatically record alarm events for subsequent analysis and improvement.

[0162] Through this complex alarm formula and mechanism, the traction and conversion resistance values ​​of railway switches can be monitored more effectively, and potential safety hazards can be discovered and dealt with in a timely manner, thereby ensuring the safety and reliability of railway operations.

[0163] 4. Offset range optimization detection accuracy

[0164] In the railway turnout traction and transfer resistance monitoring system, the offset range prediction algorithm is used to improve the accuracy of the turnout traction and transfer resistance position detection. The core idea of ​​the algorithm is to divide the possible offsets of the turnout traction and transfer resistance into multiple intervals, predict the intervals to which the offsets of the turnout traction and transfer resistance belong, and finally determine the specific positions of the turnout traction and transfer resistance by calculating the expected offsets within the intervals.

[0165] Offset interval division:

[0166] According to historical data and system requirements, the possible offset ranges of the turnout traction and conversion resistance are divided into several intervals; for example, suppose we divide the offset range of [-5,5] into 7 intervals, namely [-∞,-5), [-5,-4), [-4,-2), [-2,0), [0,2), [2,4), [4,+∞), and represent them with labels 0 to 6 respectively.

[0167] Interval classification prediction:

[0168] Assume that the classifier predicts that the probability that the target offset belongs to the i-th interval is p_i (i=0,1,...,6);

[0169] Train a classifier, taking the extracted features as input, to predict the interval to which the traction force and the switching resistance offset of the switch machine belong;

[0170] Calculation of expected offset:

[0171] For each interval, the expectation (or average) of all possible offsets within the interval is calculated based on historical data or system requirements.

[0172] After the classifier predicts the interval to which the traction force and conversion resistance offset of the switch machine belong, the corresponding interval is selected, and the offset expectation of the interval is used as the final prediction value.

[0173] formula

[0174] Interval classification prediction:

[0175] Assume that the probability that the classifier predicts that the traction force and conversion resistance offset of the switch machine belongs to the i-th interval is p_i (i=1,2,...,n, n is the number of intervals);

[0176] Calculation of expected offset:

[0177] For each interval i, calculate the expected E_i of all offsets within it;

[0178] For example, for the interval [-2,0), the expected E_i may be -1 (assuming that the offsets are uniformly distributed in the interval).

[0179] The final prediction offset is: OP = Σ(p_i*E_i)(i = 0, 1, ..., 6).

[0180] Key Points

[0181] Interval division: Reasonable interval division can balance the accuracy of classification and the granularity of regression, making the algorithm more effective;

[0182] Classifier training: Using high-quality historical data and advanced machine learning algorithms to train classifiers can improve the accuracy of predictions;

[0183] Calculation of expected offset: Calculating the expected offset within a range based on historical data or system requirements can provide more accurate prediction results;

[0184] In the railway turnout machine traction and conversion resistance monitoring system, the application of the deviation range prediction algorithm can monitor the deviation of the turnout machine traction and conversion resistance in real time, detect and handle abnormal situations in time, and ensure the safety and smoothness of railway traffic;

[0185] 5. Turnout machine traction and conversion resistance curve

[0186] The switch machine traction and transfer resistance displacement monitoring system usually involves real-time monitoring and data analysis of the switch machine traction and transfer resistance displacement of the switch machine or other similar equipment. Different data processing and visualization techniques may be used when constructing a curve algorithm to show the trend of the switch machine traction and transfer resistance displacement. The following is a description of the switch machine traction and transfer resistance displacement monitoring system curve algorithm based on statistical analysis and data visualization, and attempts to list the relevant formulas:

[0187] 1. Data collection and preprocessing

[0188] Data collection: Collect switch machine traction and conversion resistance displacement data from monitoring equipment, usually including timestamps and corresponding displacement values;

[0189] Data preprocessing: remove outliers, fill in missing values, and ensure data accuracy and consistency;

[0190] 2. Data Analysis and Modeling

[0191] Time series analysis: The traction force and conversion resistance displacement data of the switch machine are regarded as time series data, and statistical methods (such as moving average, exponential smoothing, etc.) can be used to analyze their trends and periodicity;

[0192] Formula Description

[0193] Let (D_t) be the displacement value of the switch machine traction and conversion resistance at time (t), and (N) be the window size of the moving average;

[0194] Then the moving average displacement value (OL) of time (t) can be calculated by the following formula:

[0195]

[0196] Trend prediction: Based on historical data, linear regression, machine learning, or deep learning models can be used to predict future trends in switch machine traction and switching resistance displacement.

[0197] 3. Data Visualization

[0198] Curve drawing: Use the chart library to draw a curve graph of the switch machine traction and conversion resistance displacement changing with time.

[0199] Curve type: may include line graphs, smooth curve graphs (such as using spline interpolation), etc.

[0200] Example (taking spline interpolation as an example, although the mathematical formula is not given directly here, the concept of interpolation is described):

[0201] Spline interpolation is a method of constructing a smooth curve through a set of known data points. It uses piecewise polynomial functions to approximate the original data and provide smooth transitions between data points.

[0202] 4. Algorithm Summary

[0203] Data preprocessing: ensure data accuracy and consistency.

[0204] Data Analysis: Use time series analysis and statistical models to identify trends and cyclicalities.

[0205] Data visualization: Draw a graph to visually display the changes of the switch machine traction and conversion resistance displacement over time.

[0206] 5. Notes

[0207] Data quality: The accuracy and completeness of data are critical to the analysis results.

[0208] Model selection: Select appropriate analysis models and visualization methods based on the characteristics and requirements of the data.

[0209] Real-time: The monitoring system needs to be able to collect, analyze and display data in real time.

[0210] Scalability: As the system scales up and demands increase, the algorithm needs to have good scalability.

[0211] The present invention also discloses an electronic device, Figure 3 FIG. 1 is a schematic diagram of the composition of an electronic device in an embodiment of the present invention; please refer to FIG. Figure 3 At the hardware level, the electronic device includes a memory, a processor and at least one network interface; the processor may be a microprocessor, and the memory may include a memory, such as a random access memory (RAM), or a non-volatile memory (non-volatile memory), etc. Of course, the electronic device may also be provided with other hardware as required.

[0212] The processor, network interface and memory may be interconnected via an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus may include an address bus, a data bus, a control bus, etc. The memory is used to store programs (which may include operating system programs and application programs); the program may include program code, which may include computer operation instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.

[0213] In one embodiment, the processor can read the corresponding program from the non-volatile memory into the memory and then run it; the processor can execute the program stored in the memory and is specifically used to perform the following operations (such as Figure 2 shown):

[0214] [Step S1] A force sensor is disposed on the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine;

[0215] [Step S2] The collection extension is connected to at least one force sensor to collect the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and the collected data is filtered and analyzed to determine whether each switch machine has an abnormality;

[0216] [Step S3] The monitoring host is connected to at least one collection extension, monitors the traction data and conversion resistance data of the switch machines collected by each collection extension, and performs risk prediction for each switch machine based on changes in the traction data and conversion resistance data of each switch machine.

[0217] The present invention further discloses a storage medium having computer program instructions stored thereon, which, when executed by a processor, implements the following steps of the method of the present invention (eg Figure 1 shown):

[0218] [Step S1] A force sensor is disposed on the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine;

[0219] [Step S2] The collection extension is connected to at least one force sensor to collect the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and the collected data is filtered and analyzed to determine whether each switch machine has an abnormality;

[0220] [Step S3] The monitoring host is connected to at least one collection extension, monitors the traction data and conversion resistance data of the switch machines collected by each collection extension, and performs risk prediction for each switch machine based on changes in the traction data and conversion resistance data of each switch machine.

[0221] In summary, the real-time monitoring system, method, electronic device and storage medium for the traction force and conversion resistance of the switch machine proposed in the present invention can obtain the status of the traction force and conversion resistance of the switch machine in real time, reduce the difficulty and time of debugging work, and improve work efficiency.

[0222] It should be noted that the present application can be implemented in software and / or a combination of software and hardware; for example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device. In some embodiments, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) can be stored in a computer-readable recording medium; for example, a RAM memory, a magnetic or optical drive or a floppy disk and the like. In addition, some steps or functions of the present application can be implemented in hardware; for example, as a circuit that cooperates with a processor to perform various steps or functions.

[0223] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0224] The description and application of the present invention here are illustrative, and it is not intended to limit the scope of the present invention to the above-mentioned embodiments. The effects or advantages involved in the embodiments may not be embodied in the embodiments due to interference from various factors, and the description of the effects or advantages is not used to limit the embodiments. The deformation and change of the embodiments disclosed here are possible, and the replacement of the embodiments and the various equivalent parts are well known to those of ordinary skill in the art. It should be clear to those skilled in the art that the present invention can be implemented in other forms, structures, arrangements, proportions, and with other components, materials and parts without departing from the spirit or essential features of the present invention. Other deformations and changes can be made to the embodiments disclosed here without departing from the scope and spirit of the present invention.

Claims

1. A real-time monitoring system for traction and conversion resistance of a switch machine, characterized in that: The monitoring system comprises: At least one force sensor is arranged on the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine; At least one data collection extension connected to at least one force sensor for collecting the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and filtering and analyzing the collected data to determine whether each switch machine has an abnormality; At least one monitoring host is connected to at least one collecting extension to monitor the traction force data and conversion resistance data of the switch machines collected by each collecting extension, and to perform risk prediction on each switch machine based on the changes in the traction force data and conversion resistance data of each switch machine.

2. The real-time monitoring system for traction force and conversion resistance of a switch machine according to claim 1 is characterized in that: The monitoring system further comprises: a switch machine fault analysis model construction module and an analysis model training module, wherein the switch machine fault analysis model construction module is used to construct a switch machine fault analysis model, and the analysis model training module is used to train the switch machine fault analysis model; The switch machine fault analysis model building module is used to extract features from historical data, extract the change process data of the switch machine's starting resistance, intermediate process resistance and locking resistance within a set time before the switch machine fails, and the switch machine fault characteristic data, and establish a mathematical model of the association between the change process data of the switch machine's starting resistance, intermediate process resistance and locking resistance and the switch machine fault characteristic data; wherein the switch machine fault characteristic data includes the time from the switch machine failure to the change of the switch machine's starting resistance or / and intermediate process resistance or / and locking resistance within the set time reaching a set threshold; The monitoring host performs risk prediction on each switch machine through the switch machine fault analysis model constructed by the switch machine fault analysis model construction module, and determines whether the switch machine is in danger of failure based on the change process data of the starting resistance, intermediate process resistance and locking resistance of the switch machine sensed by each force sensor.

3. The real-time monitoring system for traction force and conversion resistance of a switch machine according to claim 1 is characterized in that: The monitoring system further comprises: The server is connected to at least one monitoring host, and is used to receive the traction force data and conversion resistance data of the switch machine sent by each monitoring host, and receive the risk prediction data of the switch machine sent by the monitoring host; At least one terminal, connected to the server, for receiving the fault information and risk prediction information of the switch machine sent by the server; The monitoring system further includes at least one communication transmission module, each communication transmission module collects data processed by at least one acquisition extension and uploads the data to the monitoring host; the force sensor is arranged at the action connecting rod corresponding to the switch machine.

4. The real-time monitoring system for traction force and conversion resistance of a switch machine according to claim 1 is characterized in that: The acquisition sub-machine includes a data processing module for processing the traction force data and conversion resistance data sensed by the force sensor; The data processing module comprises: A data acquisition unit is used to acquire the traction force data and conversion resistance data of the switch machine in real time from the force measuring sensor; the traction force data and conversion resistance data include force value, state, data type and possible switching data; The data processing unit is used to clean, convert and analyze the collected raw data; the mathematical formula for the traction force and conversion resistance value of the switch machine is expressed as: Where N is the number of observations, CJZ i is the true value of the i-th observation data, CJZ ^ i is the corresponding estimated value; QTZ refers to a specific parameter or variable, i.e., a value after the traction force of the switch machine is converted into resistance; The data encoding unit is used to convert the processed data into a format suitable for transmission; the corresponding encoding method is determined according to the data type, state and force value; The data transmission unit is used to send the encoded data to the host computer or server for further analysis and storage through the communication protocol. During the transmission process, the integrity and accuracy of the data must be ensured. After receiving the data, the monitoring host or server will decode and analyze it. According to the type and status of the data, corresponding processing will be performed. At the same time, the monitoring system also needs to have error handling and early warning mechanisms. When erroneous data is received, the system should be able to identify and take corresponding processing measures, such as re-requesting data or issuing an alarm. The early warning mechanism can promptly alarm abnormal data according to the preset threshold to ensure the safety and stability of railway operation.

5. The real-time monitoring system for traction force and conversion resistance of a switch machine according to claim 1 is characterized in that: The monitoring host includes a risk prediction module, and the risk prediction module is used to predict the risk of the switch machine; The risk prediction module includes: A data collection unit for collecting traction data and conversion resistance data within a set time, which are daily, hourly or minute data points; The data preprocessing unit is used to preprocess the data after it is collected, including removing outliers and filling missing values ​​to ensure the accuracy and consistency of the data; Trend analysis unit, which uses statistical methods or machine learning algorithms to analyze historical data and identify trends in force values; including seasonal trends, cyclical trends, or long-term growth / decrease trends; Trend prediction unit, used to predict future force trend based on the analysis results of historical data using appropriate prediction algorithms; The monitoring host includes a data monitoring module, and the data monitoring module is used to monitor whether there is a risk in the traction data and conversion resistance data of the switch machine; The alarm determination formula of the data monitoring module is expressed as: Among them, Sjz is the actual traction force and conversion resistance data of the switch machine, which is the traction force and conversion resistance of the switch machine measured in real time by the force sensor; bhl is the change rate of the traction force and conversion resistance of the switch machine, which indicates how fast the traction force and conversion resistance of the switch machine change over a period of time, which helps to predict whether the traction force and conversion resistance of the switch machine may deteriorate rapidly; hjcs is the environmental parameter including temperature and humidity, which may affect the stability of the switch machine's traction and conversion resistance.

6. The real-time monitoring system for traction force and conversion resistance of a switch machine according to claim 1 is characterized in that: The monitoring host includes an offset range prediction module for predicting the possible offset of the traction force and conversion resistance of the turnout machine; In the railway turnout switch traction and conversion resistance monitoring system, the offset range prediction algorithm is used to improve the accuracy of the switch traction and conversion resistance position detection; the possible offsets of the turnout switch traction and conversion resistance are divided into multiple intervals, and the intervals to which the switch traction and conversion resistance offsets belong are predicted, and finally the specific switch traction and conversion resistance positions are determined by calculating the expected offsets within the intervals; The prediction process of the offset range prediction module includes: Offset interval division: Based on historical data and system requirements, the possible offset range of turnout traction and conversion resistance is divided into several intervals; Interval classification prediction: Assume that the probability that the classifier predicts that the target offset belongs to the i-th interval is p_i; train a classifier, use the extracted features as input, and predict the interval to which the traction force and conversion resistance offset of the switch machine belong; Calculation of expected offset: For each interval, the expected value or average value of all possible offsets in the interval is calculated based on historical data or system requirements; when the classifier predicts the interval to which the traction force and conversion resistance offset of the switch machine belongs, the corresponding interval is selected and the expected offset of the interval is used as the final prediction value; Interval classification prediction: Assume that the classifier predicts that the probability that the traction force and conversion resistance offset of the switch machine belongs to the i-th interval is p_i; where i = 1, 2, ..., n, and n is the number of intervals; Calculation of expected offsets: For each interval i, calculate the expected E_i of all offsets within it.

7. The real-time monitoring system for traction force and conversion resistance of a switch machine according to claim 1 is characterized in that: The monitoring host comprises a curve generating module, and the curve generating module is used to generate a curve of the traction force data and the conversion resistance data of the turnout machine; The generation process of the curve generation module includes: Data collection and preprocessing steps: Collect the traction and conversion resistance displacement data of the switch machine from the monitoring equipment, usually including the timestamp and the corresponding displacement value; remove outliers in the data and fill in missing values ​​to ensure the accuracy and consistency of the data; Data analysis and modeling steps: The traction force and conversion resistance displacement data of the switch machine are regarded as time series data, and statistical methods are used to analyze their trends and periodicity; let D_t be the traction force and conversion resistance displacement value of the switch machine at time t, and N is the window size of the moving average; then the moving average displacement value OL at time t is calculated by the following formula: Trend prediction step: Based on historical data, use linear regression, machine learning or deep learning models to predict future switch machine traction and conversion resistance displacement trends.

8. A real-time monitoring method for traction and conversion resistance of a switch machine, characterized in that: The monitoring method comprises: The force sensor is arranged on the corresponding switch machine to sense the traction force data and conversion resistance data of the corresponding switch machine; The collection extension is connected to at least one force sensor, collects the traction force data and conversion resistance data of the switch machine sensed by each force sensor, and filters and analyzes the collected data to determine whether each switch machine has an abnormality; The monitoring host is connected to at least one collection extension, monitors the traction data and conversion resistance data of the switch machines collected by each collection extension, and performs risk prediction on each switch machine based on the changes in the traction data and conversion resistance data of each switch machine.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to claim 8 are implemented.

10. A storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the steps of the method according to claim 8 are implemented.

Images