Intelligent monitoring method for rail deformation of terminal equipment

Abstract

The present invention discloses an intelligent monitoring method for track deformation of terminal equipment, and relates to the technical field of track deformation monitoring. The present invention aims to realize high-precision monitoring of the deformation state of the track of terminal equipment, firstly obtains the predicted value of the track deformation at the current moment, and then obtains a more reliable predicted mean value of the track deformation at the current moment by using a scoring and weighting method, so as to provide a high-precision standard for subsequent calculation and analysis, and then obtains the residual value of the track deformation at the current moment by comparing the difference between the predicted mean value and the actual measured value, and then identifies the residual value of the track deformation at the current moment by using a double threshold form, so as to more effectively and accurately identify the abnormal point of the track deformation at the current moment, thereby providing a good advance condition for the triggering of the subsequent alarm mechanism, ensuring that no misjudgment occurs, and then effectively preventing potential safety hazards and ensuring smooth operation and personnel safety on the premise of effectively improving the scientificity and accuracy of data utilization.

Description

Technical Field

[0001] The present invention relates to the technical field of track deformation monitoring, and in particular to an intelligent monitoring method for track deformation of terminal equipment. Background Art

[0002] As global trade continues to grow, port terminals, as key nodes for cargo transportation, have become particularly important for their equipment safety and reliability. Especially in the application of heavy machinery and equipment such as rail cranes, track deformation not only affects operating efficiency, but may also bring serious safety hazards. Therefore, intelligent monitoring of track deformation of terminal equipment has become an important part of ensuring the safety of port operations.

[0003] Currently, the monitoring of rail deformation of terminal equipment mainly relies on traditional regular manual inspections and sensor-based automatic detection systems. These methods have their own characteristics:

[0004] Regular manual inspection: This method relies on professional technicians to use measuring tools such as levels, total stations, etc. to conduct a comprehensive inspection of the track. Although some obvious deformation problems can be found, early or subtle deformations are often not detected in time due to high labor costs, long time consumption, and difficulty in covering all details.

[0005] Sensor-based automatic detection system: In recent years, with the development of Internet of Things technology, more and more terminals have begun to use various sensors installed on the track, such as displacement sensors, accelerometers, etc., to achieve continuous monitoring of the track status. This type of system can collect data in real time and upload it to the data center through a wireless network for analysis and processing. However, the defects of the existing system are mainly reflected in the following aspects:

[0006] First, the existing technology only relies on the data provided by a certain type of sensor to make decisions, which easily ignores other important reference indicators. For example, factors such as ambient temperature and inclination angle will also affect track deformation. Even if the existing technology uses multiple types of data sources, it will also affect the accuracy of the final identification conclusion if there is no reasonable weight given to each factor.

[0007] 2. In the prior art, most systems have an alarm function based on threshold determination, that is, the final recognition conclusion is determined by using the threshold, and whether to apply the alarm function is selected based on the determination result. Although the current application efficiency is relatively good, it is still often easy to gradually highlight the irrationality of the existing threshold due to the singleness of the existing threshold setting, which often leads to misjudgment and false alarm, thereby reducing the effectiveness of the existing alarm mechanism.

[0008] Therefore, the prior art urgently needs a technical solution for an intelligent monitoring method for track deformation of terminal equipment. Summary of the invention

[0009] In order to solve the above technical problems, the present invention provides an intelligent monitoring method for track deformation of terminal equipment, which is characterized by comprising the following steps:

[0010] Step S1, collecting real-time track data through sensors deployed at key positions of the track, and transmitting the real-time data to a cloud server through wireless communication, and then the cloud server pre-processes the collected real-time data;

[0011] Step S2: using the pre-processed real-time data to obtain the track deformation prediction value at the current moment;

[0012] Step S21, calling the track history data stored in the cloud server, performing preprocessing, and performing feature engineering on the preprocessed track history data to extract key features of track deformation;

[0013] Step S22, the pre-processed track history data and key features are collected into a data set, and the data set is divided according to a preset ratio, and any part of the divided data set is obtained as a sample set;

[0014] Step S23: construct a prediction model, and use the sample set to train the prediction model to obtain a trained prediction model;

[0015] Step S24, input the pre-processed real-time data into the trained prediction model to obtain the track deformation prediction value at the current moment;

[0016] Step S3, obtaining the actual measured value of the track deformation at the current moment, and comparing it with the predicted value of the track deformation at the current moment, to obtain the residual value of the track deformation at the current moment, and using the residual value of the track deformation at the current moment in combination with the threshold to identify whether there is a track deformation abnormal point at the current moment;

[0017] Step S3a1, obtaining the real-time data and performing calculations to obtain the actual measurement value of the track deformation at the current moment;

[0018] Step S3a2, calling the track deformation prediction value at the current moment obtained in step S24, and analyzing the influence of the track deformation prediction value at the current moment on the current track deformation state, assigning a score to the track deformation prediction value at the current moment based on the analysis result, and assigning a weight to the track deformation prediction value at the current moment based on the score;

[0019] Step S3a3, combining the score and weight of the track deformation prediction value at the current moment to obtain the comprehensive weight of the track deformation prediction value at the current moment;

[0020] Step S3a4, obtaining the predicted mean value of the track deformation at the current moment according to the weight of the predicted value of the track deformation at the current moment, the comprehensive weight of the predicted value of the track deformation at the current moment, and the predicted value of the track deformation at the current moment;

[0021] Among them, the calculation formula for obtaining the predicted mean value of track deformation at the current moment is:

[0022] ;

[0023] in, Represents the current The predicted mean value of track deformation at the time; Represents the current The predicted value of track deformation at the time; Represents the current The weight of the track deformation prediction value at the moment; Represents the current The comprehensive weight of the track deformation prediction value at the moment;

[0024] Step S3a5, obtaining the track deformation residual value at the current moment according to the predicted mean value of the track deformation at the current moment and the actual measured value of the track deformation at the current moment;

[0025] Among them, the calculation formula for the track deformation residual value at the current moment is:

[0026] ;

[0027] in, Represents the current Track deformation residual value at time; Represents the current The predicted mean value of track deformation at the time; Represents the current The actual measured value of track deformation at the time;

[0028] Step S3b1, obtaining multiple groups of track deformation residual values, calculating the mean of the multiple groups of track deformation residual values, and then obtaining the standard deviation of the multiple groups of track deformation residual values ​​through the mean value;

[0029] Step S3b2, obtaining the lower threshold limit and the upper threshold limit of the track deformation residual value through the means and standard deviations of multiple groups of track deformation residual values;

[0030] Among them, the formula group for obtaining the lower threshold and upper threshold of the track deformation residual value is:

[0031] ;

[0032] ;

[0033] in, Represents the lower threshold of the track deformation residual value; Represents the upper threshold of the track deformation residual value; Represents the mean of multiple groups of track deformation residual values; represents the standard deviation of multiple sets of track deformation residual values; k represents a constant;

[0034] Step S3b3: if the track deformation residual value at the current moment is less than the lower threshold value of the track deformation residual value or greater than the upper threshold value of the track deformation residual value, it is identified that there is a track deformation abnormal point at the current moment; if the track deformation residual value at the current moment is greater than or equal to the lower threshold value of the track deformation residual value and less than or equal to the upper threshold value of the track deformation residual value, it is identified that there is no track deformation abnormal point at the current moment; the track deformation abnormal point at the current moment is identified by using the track deformation residual value at the current moment in combination with the threshold value;

[0035] Step S4: If it is identified that there is an abnormal point of track deformation at the current moment, the alarm mechanism is triggered and a detailed abnormality report is generated. If it is identified that there is no abnormal point of track deformation at the current moment, there is no need to trigger the alarm mechanism.

[0036] The embodiments of the present invention have the following technical effects:

[0037] The present invention aims to achieve high-precision monitoring of the deformation state of the track of terminal equipment. First, the predicted value of the track deformation at the current moment is obtained through a prediction model. In order to ensure the stability and reliability of the predicted value of the track deformation at the current moment, the present application obtains a more reliable predicted mean value of the track deformation at the current moment by using scoring and weighting methods. This method can make it possible to obtain a relatively stable and reliable predicted mean value even in a complex and changeable environment, providing a high-precision standard for subsequent calculation and analysis. Then, by comparing the difference between the predicted mean value and the actual measured value, the track deformation residual value at the current moment is obtained, and then a reasonable and scientific double threshold value is obtained by calculation. The track deformation residual value at the current moment is identified in the form of a double threshold value, which can more effectively and accurately identify the abnormal point of the track deformation at the current moment, thereby providing good advance conditions for the triggering of the subsequent alarm mechanism, ensuring that there will be no misjudgment, and then effectively preventing potential safety hazards and ensuring smooth operation and personnel safety on the premise of effectively improving the scientificity and accuracy of data utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 It is a flow chart of a method for intelligently monitoring deformation of rails of terminal equipment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. 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 belong to the scope of protection of the present invention.

[0041] Embodiment 1: Figure 1 As shown, the present invention provides a method for intelligently monitoring deformation of a terminal equipment track, comprising the following steps:

[0042] Step S1, collecting real-time track data through sensors deployed at key positions of the track, and transmitting the real-time data to a cloud server through wireless communication, and then the cloud server pre-processes the collected real-time data;

[0043] It is worth noting that sensors are deployed at key positions of the track, including but not limited to track connections, track curves, track fixing points, track joints, and frequently stressed areas; for track connections: these positions are prone to stress concentration and are high-incidence areas of deformation; for track curves: curves are prone to greater deformation due to the effect of steering force; for track fixing points: such as near sleepers or bases, the stability of these places will directly affect the overall stability of the track; for track joints: especially those joints that use different materials or structures; for frequently stressed areas: for example, areas where heavy equipment such as cranes are frequently moved;

[0044] In addition, the sensors deployed and the corresponding real-time data collected include but are not limited to displacement sensors: used to measure the displacement of the track in the horizontal and vertical directions; strain gauges: installed on the track to monitor the slight deformation of the track caused by external forces; temperature sensors: monitor changes in ambient temperature, because temperature has a significant impact on the expansion and contraction of metal materials; vibration sensors: detect the vibration of the track and its supporting structure, which helps to identify abnormal vibration patterns; inclinometers: measure the inclination angle of the track to evaluate whether the track remains level;

[0045] Further, in terms of the selection of displacement sensors, preferably, an NDW type potentiometer displacement meter is installed in the lateral position of the track, usually fixed on the side of the track, to ensure that the displacement of the track in the horizontal direction can be measured, and an RW type capacitive displacement meter is installed in the longitudinal position of the track, usually fixed below the track, to ensure that the displacement of the track in the vertical direction can be measured; in terms of the selection of inclinometers, preferably, an NJX11-B type beam inclinometer is installed in a key position of the track, usually on the center line of the track, to ensure that the inclination angle of the track can be measured;

[0046] The NDW type potentiometer displacement meter measures the displacement by the change of the potentiometer. When the track is horizontally displaced, the resistance value of the potentiometer will change. By measuring the change of the resistance value, the horizontal displacement of the track can be calculated. The NDW type potentiometer displacement meter outputs an analog voltage signal, which is proportional to the displacement; the RW type capacitive displacement meter measures the displacement by the change of the capacitance. When the track is vertically displaced, the capacitance value will change. By measuring the change of the capacitance value, the vertical displacement of the track can be calculated. The RW type capacitive displacement meter outputs an analog voltage signal, which is proportional to the displacement; the NJX11-B type beam inclinometer measures the inclination angle of the track through a built-in inclination sensor such as a MEMS accelerometer. When the track is tilted, the output value of the accelerometer will change. By measuring the output value of the accelerometer, the inclination angle of the track can be calculated. The NJX11-B type beam inclinometer outputs an analog voltage signal, which is proportional to the inclination angle;

[0047] The wireless communication method used is 4G / 5G cellular network, which means that there is good cellular signal coverage on site, and then the existing infrastructure is used for data transmission to ensure high speed and reliability;

[0048] After the collected real-time data is transmitted to the cloud server via wireless communication, a series of preprocessing steps are required to ensure data quality and prepare for subsequent analysis. The preprocessing steps include but are not limited to: data cleaning: remove invalid values ​​such as real-time data that exceeds a reasonable range and fill in missing values; data standardization / normalization: convert real-time data from different sources to the same scale to facilitate subsequent model training and comparison; noise filtering: use filtering algorithms such as Kalman filtering and median filtering to reduce noise interference; time synchronization: ensure that the real-time data collected by all sensors are compared under the same time reference, that is, the timestamp of the real-time data is unified to UTC time; anomaly detection: preliminarily check whether there are obviously abnormal data points and mark them for further investigation.

[0049] Step S2: using the pre-processed real-time data to obtain the track deformation prediction value at the current moment;

[0050] Step S21, calling the track history data stored in the cloud server, performing preprocessing, and performing feature engineering on the preprocessed track history data to extract key features of track deformation;

[0051] All relevant sensor data from the past period of time are extracted from the cloud server database, including displacement, strain value, temperature, vibration intensity and tilt angle, as well as the actual measurement value of track deformation, as historical data; then the historical data is preprocessed, including data cleaning: deleting invalid or erroneous data points, such as readings that are obviously beyond the normal range; missing value processing: using interpolation methods, such as linear interpolation, to fill in missing data; denoising: applying a low-pass filter to remove high-frequency noise; normalization / standardization: converting different types of data to the same scale for subsequent analysis, for example, using minimum-maximum scaling or Z-Score standardization; followed by feature engineering processing, including cross-features: generating interactive features between different sensor data, such as the relationship between displacement and temperature, the correlation between vibration intensity and strain value, etc.; trend features: calculating the change trend over a period of time through a sliding window, such as the speed of rise or fall; periodic features: identifying and extracting periodic patterns in the data, such as daily cycles, weekly cycles, etc.; finally, key feature selection: using feature selection algorithms, such as recursive feature elimination RFE, LASSO regression, etc., to screen out the most important features for track deformation prediction.

[0052] Step S22, the pre-processed track history data and key features are collected into a data set, and the data set is divided according to a preset ratio, and any part of the divided data set is obtained as a sample set;

[0053] The preprocessed historical data and the extracted key features are then merged into a complete data set, ensuring that each record contains input features, such as displacement, strain value, etc., and output targets, i.e., the actual measured values ​​of track deformation. The data set is then randomly divided into 80% training set and 20% test set ratios. It can also be further divided into training set (70%), validation set (10%), and test set (20%) to better evaluate the model performance, maintain the consistency and randomness of data division, and ensure the effectiveness of model training and testing.

[0054] Step S23: construct a prediction model, and use the sample set to train the prediction model to obtain a trained prediction model;

[0055] For the selection of prediction models, it can be linear regression, support vector machine, random forest, gradient boosting tree, long short-term memory network, etc., and then use the above training set data to train the selected prediction model. During the training process, the prediction model parameters are continuously adjusted to minimize the prediction error, and then the prediction model performance is evaluated on the validation set. The prediction model structure or parameters are adjusted according to the evaluation results. Repeat the above process until the prediction model reaches satisfactory performance indicators, such as mean square error (MSE), determination coefficient (R²), etc.; after the training is completed, save the final prediction model parameters and prepare for the prediction of real-time data.

[0056] Step S24, input the pre-processed real-time data into the trained prediction model to obtain the track deformation prediction value at the current moment;

[0057] The newly collected real-time data is subjected to the same preprocessing steps as the historical data to ensure that the format and scale are consistent. The feature engineering method in step S21 is then applied to extract the corresponding features from the real-time data. The preprocessed real-time data and its features are then input into the trained prediction model until the prediction model outputs the predicted value of the track deformation at the current moment. Finally, the prediction results are stored in the cloud server and prepared for subsequent steps.

[0058] Step S3, obtaining the actual measured value of the track deformation at the current moment, and comparing it with the predicted value of the track deformation at the current moment, to obtain the residual value of the track deformation at the current moment, and using the residual value of the track deformation at the current moment in combination with the threshold to identify whether there is a track deformation abnormal point at the current moment;

[0059] Step S3a1, obtaining the real-time data and performing calculations to obtain the actual measurement value of the track deformation at the current moment;

[0060] The real-time data collected by the above multiple sensors are integrated, and the median method is selected to calculate the actual measurement value of the track deformation, that is, the median of the real-time data collected by all sensors is taken as the actual measurement value of the track deformation. The advantage of this method is that it is highly robust to outliers because the median is not affected by extreme values;

[0061] Specifically, in order to comprehensively obtain the actual measured value of track deformation at the current moment from the real-time data collected by multiple sensors, the median method is used. The following is the detailed calculation method, principle and algorithm:

[0062] The principles of the median method include: median definition and advantages;

[0063] The definition of median: The median is the value in the middle of a set of data after it is sorted. If the number of data is odd, the median is the number in the middle; if it is even, the median is the average of the two numbers in the middle;

[0064] Advantages: The median is robust to outliers because it is not affected by extreme values. Even if some sensors have abnormal readings, the median can still provide a relatively stable and reliable actual measurement value.

[0065] The following is the specific calculation method including data collection, sorting data and median calculation;

[0066] First, data collection is performed: Assume that n sensors are deployed at key locations on the track, and the readings of each sensor at the current moment are , ; Then sort the above data and arrange the readings in ascending order, recorded as , ; Then the median is calculated. If n is an odd number, the corresponding calculation formula is: ; If n is an even number, the corresponding calculation formula is: ; the obtained That is ; Through the above detailed implementation method, it is clearly demonstrated how to use the median method to calculate the actual measurement value of the track deformation at the current moment from the real-time data collected by multiple sensors. This method is not only simple and easy, but also has strong robustness to outliers and can provide a stable and reliable measurement result.

[0067] Step S3a2, calling the track deformation prediction value at the current moment obtained in step S24, and analyzing the influence of the track deformation prediction value at the current moment on the current track deformation state, assigning a score to the track deformation prediction value at the current moment based on the analysis result, and assigning a weight to the track deformation prediction value at the current moment based on the score;

[0068] The track deformation prediction value at the current moment generated in step S24 is called from the cloud server to analyze the influence of the track deformation prediction value at the current moment on the current track deformation state. This should be evaluated through similar situations in historical data or expert knowledge. For example, when the track deformation prediction value at the current moment shows a large change trend, it indicates that there is a significant risk of track deformation, so the influence of the prediction value is relatively high. Based on the above analysis results, a score is assigned to the track deformation prediction value at the current moment. The score ranges from 0 to 1, where 1 indicates the highest influence. According to the score, a corresponding weight is assigned to the track deformation prediction value at the current moment. The weight can be a direct mapping of the score or a value converted by a certain function.

[0069] Step S3a3, combining the score and weight of the track deformation prediction value at the current moment to obtain the comprehensive weight of the track deformation prediction value at the current moment;

[0070] Step S3a4, obtaining the predicted mean value of the track deformation at the current moment according to the weight of the predicted value of the track deformation at the current moment, the comprehensive weight of the predicted value of the track deformation at the current moment, and the predicted value of the track deformation at the current moment;

[0071] Among them, the calculation formula for obtaining the predicted mean value of track deformation at the current moment is:

[0072] ;

[0073] in, Represents the current The predicted mean value of track deformation at the time; Represents the current The predicted value of track deformation at the time; Represents the current The weight of the track deformation prediction value at the moment; Represents the current The comprehensive weight of the track deformation prediction value at the moment;

[0074] Step S3a5, obtaining the track deformation residual value at the current moment according to the predicted mean value of the track deformation at the current moment and the actual measured value of the track deformation at the current moment;

[0075] Among them, the calculation formula for the track deformation residual value at the current moment is:

[0076] ;

[0077] in, Represents the current Track deformation residual value at time; Represents the current The predicted mean value of track deformation at the time; Represents the current The actual measured value of track deformation at the time;

[0078] Step S3b1, obtaining multiple groups of track deformation residual values, including the track deformation residual value at the current time t, the track deformation residual value at the current time t+1, and the track deformation residual value at the current time t+2, and calculating the mean of the multiple groups of track deformation residual values, and then obtaining the standard deviation of the multiple groups of track deformation residual values ​​through the mean value;

[0079] In the above, it is worth noting that the multiple groups of orbit deformation residual values ​​include but are not limited to the orbit deformation residual value at the current time t, the orbit deformation residual value at the current time t+1, and the orbit deformation residual value at the current time t+2.

[0080] Step S3b2, obtaining the lower threshold limit and the upper threshold limit of the track deformation residual value through the means and standard deviations of multiple groups of track deformation residual values;

[0081] Among them, the formula group for obtaining the lower threshold and upper threshold of the track deformation residual value is:

[0082] ;

[0083] ;

[0084] in, Represents the lower threshold of the track deformation residual value; Represents the upper threshold of the track deformation residual value; Represents the mean of multiple groups of track deformation residual values; represents the standard deviation of multiple sets of track deformation residual values; k represents a constant;

[0085] It is worth noting that when choosing a suitable constant k, the value is usually 2 or 3 to cover approximately 95% or 99.7% of the data. This value should be adjusted based on the specific application scenario and experience with historical data.

[0086] Step S3b3: if the track deformation residual value at the current moment is less than the lower threshold value of the track deformation residual value or greater than the upper threshold value of the track deformation residual value, it is identified that there is a track deformation abnormal point at the current moment; if the track deformation residual value at the current moment is greater than or equal to the lower threshold value of the track deformation residual value and less than or equal to the upper threshold value of the track deformation residual value, it is identified that there is no track deformation abnormal point at the current moment; the track deformation abnormal point at the current moment is identified by using the track deformation residual value at the current moment in combination with the threshold value;

[0087] Step S4: If it is identified that there is an abnormal point of track deformation at the current moment, the alarm mechanism is triggered and a detailed abnormality report is generated. If it is identified that there is no abnormal point of track deformation at the current moment, there is no need to trigger the alarm mechanism.

[0088] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0089] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. Intelligent monitoring method for track deformation of terminal equipment, characterized in that: The following steps are involved: Step S1, collecting real-time track data through sensors deployed at key positions of the track, and transmitting the real-time data to a cloud server through wireless communication, and then the cloud server pre-processes the collected real-time data; Step S2: using the pre-processed real-time data to obtain the track deformation prediction value at the current moment; specifically including: Step S21, calling the track history data stored in the cloud server, performing preprocessing, and performing feature engineering on the preprocessed track history data to extract key features of track deformation; Step S22, the pre-processed track history data and key features are collected into a data set, and the data set is divided according to a preset ratio, and any part of the divided data set is obtained as a sample set; Step S23: construct a prediction model, and use the sample set to train the prediction model to obtain a trained prediction model; Step S24, input the pre-processed real-time data into the trained prediction model to obtain the track deformation prediction value at the current moment; Step S3, obtaining the actual measured value of the track deformation at the current moment, and comparing it with the predicted value of the track deformation at the current moment, to obtain the residual value of the track deformation at the current moment, and using the residual value of the track deformation at the current moment in combination with the threshold to identify whether there is a track deformation abnormal point at the current moment; Step S4: If it is identified that there is an abnormal point of track deformation at the current moment, the alarm mechanism is triggered and a detailed abnormality report is generated. If it is identified that there is no abnormal point of track deformation at the current moment, there is no need to trigger the alarm mechanism.

2. The intelligent monitoring method for track deformation of terminal equipment according to claim 1 is characterized in that: The actual measured value of the track deformation at the current moment is obtained, and compared with the predicted value of the track deformation at the current moment to obtain the residual value of the track deformation at the current moment, including: Step S3a1, calculating by using the acquired real-time data to obtain the actual measured value of the track deformation at the current moment; Step S3a2, calling the track deformation prediction value at the current moment obtained in step S24, and analyzing the influence of the track deformation prediction value at the current moment on the current track deformation state, assigning a score to the track deformation prediction value at the current moment based on the analysis result, and assigning a weight to the track deformation prediction value at the current moment based on the score; Step S3a3, combining the score and weight of the track deformation prediction value at the current moment to obtain the comprehensive weight of the track deformation prediction value at the current moment; Step S3a4, obtaining the predicted mean value of the track deformation at the current moment according to the weight of the predicted value of the track deformation at the current moment, the comprehensive weight of the predicted value of the track deformation at the current moment, and the predicted value of the track deformation at the current moment; Step S3a5: Obtain the track deformation residual value at the current moment according to the predicted mean value of the track deformation at the current moment and the actual measured value of the track deformation at the current moment.

3. The intelligent monitoring method for track deformation of terminal equipment according to claim 1 is characterized in that: The method of using the track deformation residual value at the current moment in combination with a threshold to identify whether there is a track deformation abnormal point at the current moment includes: Step S3b1, obtaining multiple groups of track deformation residual values, calculating the mean of the multiple groups of track deformation residual values, and then obtaining the standard deviation of the multiple groups of track deformation residual values ​​through the mean value; Step S3b2, obtaining the lower threshold limit and the upper threshold limit of the track deformation residual value through the means and standard deviations of multiple groups of track deformation residual values; Step S3b3: If the track deformation residual value at the current moment is less than the lower threshold value of the track deformation residual value or greater than the upper threshold value of the track deformation residual value, it is identified that there is a track deformation abnormal point at the current moment; if the track deformation residual value at the current moment is greater than or equal to the lower threshold value of the track deformation residual value and less than or equal to the upper threshold value of the track deformation residual value, it is identified that there is no track deformation abnormal point at the current moment.

4. The intelligent monitoring method for track deformation of terminal equipment according to claim 2 is characterized in that: The calculation formula for obtaining the predicted mean value of track deformation at the current moment is: ； in, Represents the current The predicted mean value of track deformation at the time; Represents the current The predicted value of track deformation at the time; Represents the current The weight of the track deformation prediction value at the moment; Represents the current The comprehensive weight of the track deformation prediction value at the moment.

5. The intelligent monitoring method for track deformation of terminal equipment according to claim 2 is characterized in that: The calculation formula for obtaining the track deformation residual value at the current moment is: ； in, Represents the current Track deformation residual value at time; Represents the current The predicted mean value of track deformation at the time; Represents the current The actual measured value of the track deformation at the time.

6. The intelligent monitoring method for track deformation of terminal equipment according to claim 3 is characterized in that: The formula group for obtaining the lower threshold and upper threshold of the track deformation residual value is: ； ； in, Represents the lower threshold of the track deformation residual value; Represents the upper threshold of the track deformation residual value; Represents the mean of multiple groups of track deformation residual values; Represents the standard deviation of multiple groups of track deformation residual values; k represents a constant.

Images