Railway deformation monitoring method based on Beidou satellite positioning technology

Abstract

The present application relates to the field of satellite positioning technology, and specifically to a method for monitoring deformation along a railway based on Beidou satellite positioning technology, the method comprising: using Beidou satellite positioning technology to obtain the vertical monitoring value of any monitoring station along the railway in each monitoring result, and obtaining the temperature data of the mortar layer; based on the changes in each vertical monitoring value and multiple vertical monitoring values ​​of its temporal neighbors, and the fluctuations after detrending, respectively determine the settlement trend characteristic value and abnormal fluctuation characteristic value of each vertical monitoring value; and then determine the settlement characteristic value of each vertical monitoring value; combine the discrete degree and average level of each settlement characteristic value and multiple settlement characteristic values ​​of its temporal neighbors, and determine the deformation characteristic value of each settlement characteristic value; and use a neural network to obtain the deformation monitoring result of the railway at any of the monitoring stations. The present application aims to improve the accuracy of deformation monitoring along the railway.

Description

Technical Field

[0001] The present application relates to the field of satellite positioning technology, and in particular to a method for monitoring deformation along a railway based on Beidou satellite positioning technology. Background Art

[0002] At present, high-speed railways have become one of the important means of transportation for people due to their advantages such as high speed and good comfort. However, due to the complex environment along the railway, it is often necessary to fully understand the safety situation along the railway. For example, along the railway in the mining area, due to the long-term mining of mineral resources, the area along the railway has settlement problems, which can easily lead to deformation along the railway in the mining area, thus causing greater safety hazards in the operation of high-speed railways. Therefore, monitoring the deformation along the railway can timely discover the safety hazards of the railway and formulate targeted emergency response measures for accidents, which is of great practical significance for ensuring the safe operation of the railway.

[0003] Nowadays, Beidou satellite positioning technology is mostly used to obtain Beidou positioning data in the environment along the railway. However, Beidou positioning data in the environment along the railway is easily affected by environmental noise. The existing technology usually performs noise reduction processing on Beidou positioning data to eliminate the interference of environmental noise, thereby improving the accuracy of deformation monitoring along the railway. However, due to the strong correlation between the settlement characteristics of the environmental area along the railway and the deformation along the railway, the existing method of noise reduction processing of Beidou positioning data can easily reduce the settlement characteristics of the environmental area along the railway, making it impossible to effectively improve the accuracy of deformation monitoring along the railway, resulting in poor accuracy of deformation monitoring along the railway. Summary of the invention

[0004] In view of the above, it is necessary to provide a railway deformation monitoring method based on Beidou satellite positioning technology, which can improve the accuracy of railway deformation monitoring compared with traditional railway deformation monitoring methods.

[0005] The railway deformation monitoring method based on Beidou satellite positioning technology in this application adopts the following technical solutions:

[0006] An embodiment of the present application provides a method for monitoring deformation along a railway based on Beidou satellite positioning technology, the method comprising the following steps:

[0007] Use Beidou satellite positioning technology to obtain the vertical monitoring value of each monitoring result at any monitoring station along the railway within a preset time range, and obtain the mortar layer temperature data at the corresponding time of each monitoring result;

[0008] Based on the changes of each vertical monitoring value and multiple vertical monitoring values ​​that are close to it in time, determine the settlement trend characteristic value of each vertical monitoring value;

[0009] Based on the fluctuation of each vertical monitoring value and multiple vertical monitoring values ​​that are temporally adjacent to each other after detrending, the abnormal fluctuation characteristic value of each vertical monitoring value is determined;

[0010] Determining the settlement characteristic value of each vertical monitoring value based on the settlement trend characteristic value and the abnormal fluctuation characteristic value;

[0011] Determine the deformation characteristic value of each settlement characteristic value by combining the discrete degree and average level of each settlement characteristic value and multiple settlement characteristic values ​​that are close to each other in time;

[0012] The neural network is trained based on all vertical monitoring values, all mortar layer temperature data and the deformation characteristic values ​​within the preset time range, and based on all vertical monitoring values ​​and all mortar layer temperature data of the day, the deformation monitoring result of the railway at any monitoring station is obtained using the trained neural network.

[0013] In one embodiment, the process of determining the settlement trend characteristic value is as follows:

[0014] Arrange all vertical monitoring values ​​within the preset time range in time sequence to form a vertical monitoring sequence;

[0015] A window of a preset size is constructed with each vertical monitoring value in the vertical monitoring sequence as the center, and all vertical monitoring values ​​in the window are arranged in time sequence to form an instantaneous monitoring sequence;

[0016] The expression of the settlement trend characteristic value of each vertical monitoring value in the vertical monitoring sequence is:

[0017] ; In the formula, is the settlement trend characteristic value of the ith vertical monitoring value in the vertical monitoring sequence; , , They are respectively the number of negative values, the number of positive values, and the sum of all elements in the first-order difference sequence of the instantaneous monitoring sequence of the i-th vertical monitoring value in the vertical monitoring sequence.

[0018] In one embodiment, the process of determining the abnormal fluctuation characteristic value is as follows:

[0019] Detrended fluctuation analysis method is used to detrend each instantaneous monitoring sequence to obtain the detrended monitoring sequence of each vertical monitoring value;

[0020] The abnormal fluctuation characteristic value is positively correlated with the discreteness and the range of all elements in the detrended monitoring sequence.

[0021] In one embodiment, the process of determining the sedimentation characteristic value is:

[0022] Determining the credible weight of the settlement trend characteristic of each vertical monitoring value in the vertical monitoring sequence based on the abnormal fluctuation characteristic value;

[0023] The settlement characteristic value of each vertical monitoring value in the vertical monitoring sequence is obtained by combining the credible weight and the settlement trend characteristic value.

[0024] In one embodiment, the credible weight is the difference between 1 and the normalized value of the abnormal fluctuation characteristic value.

[0025] In one embodiment, the process of determining the deformation characteristic value is as follows:

[0026] Arranging the sedimentation characteristic values ​​in time sequence to form a sedimentation characteristic sequence;

[0027] For the settlement feature sequence, the same acquisition method as the instantaneous monitoring sequence of each vertical monitoring value in the vertical monitoring sequence is adopted to acquire the settlement feature local sequence of each settlement feature value in the settlement feature sequence;

[0028] The deformation characteristic value is determined by the discrete degree and the mean value of all the settlement characteristic values ​​in the settlement characteristic local sequence of each settlement characteristic value.

[0029] In one embodiment, the expression of the deformation characteristic value is:

[0030] ; In the formula, is the deformation eigenvalue of the ith settlement eigenvalue in the settlement eigenvalue sequence; , are the discrete degree and mean of all the settlement characteristic values ​​in the settlement characteristic local sequence of the ith settlement characteristic value in the settlement characteristic sequence, respectively; is the normalization function.

[0031] In one embodiment, the training set for training the neural network includes: a temperature monitoring sequence composed of all mortar layer temperature data within the preset time range arranged in time sequence, and a vertical monitoring sequence.

[0032] In one embodiment, a sequence of the deformation feature values ​​arranged in time sequence is used as a label sequence of a training set.

[0033] In one embodiment, the deformation monitoring result of the railway at any one of the monitoring stations is: a label sequence with deformation characteristics along the railway obtained by using a trained neural network.

[0034] This application has at least the following beneficial effects:

[0035] Since the existing method of noise reduction processing of Beidou positioning data easily reduces the settlement characteristics of the environmental area along the railway, it is impossible to effectively improve the accuracy of deformation monitoring along the railway, resulting in poor accuracy of deformation monitoring along the railway. Therefore, this application determines the settlement trend characteristic value of each vertical monitoring value based on the changes in each vertical monitoring value and multiple vertical monitoring values ​​that are close to it in time; characterizes the trend characteristics of regional settlement near the railway, which is conducive to the subsequent accurate extraction of deformation characteristics along the railway;

[0036] Further, based on the fluctuation of each vertical monitoring value and multiple vertical monitoring values ​​that are temporally adjacent after detrending, the abnormal fluctuation characteristic value of each vertical monitoring value is determined; based on the settlement trend characteristic value and the abnormal fluctuation characteristic value, the settlement characteristic value of each vertical monitoring value is determined; according to the abnormal fluctuation characteristics of the vertical monitoring values, the credible weight of the settlement trend characteristic is obtained, and the settlement trend characteristic value is weighted to eliminate the adverse interference caused by the external environment interference, so that the extracted settlement characteristics are more real and effective, and the interference effect of noise on the accuracy of deformation monitoring along the railway is avoided;

[0037] Furthermore, the deformation characteristic value of each settlement characteristic value is determined by combining the discrete degree and average level of each settlement characteristic value and multiple settlement characteristic values ​​of its temporal neighbors; the complexity of regional settlement characteristics along the railway is analyzed to improve the accuracy of deformation monitoring along the railway;

[0038] Furthermore, the neural network model is used to monitor the deformation along the railway based on the deformation characteristic value. This application avoids the problem of reducing the settlement characteristics of the environmental area along the railway, and accurately eliminates the interference of the external environment, which can effectively improve the accuracy of deformation monitoring along the railway. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A flowchart of the steps of the railway deformation monitoring method based on Beidou satellite positioning technology provided in this application;

[0041] Figure 2 It is a schematic diagram of the change curve of the vertical monitoring value;

[0042] Figure 3 This is a schematic diagram of detrended volatility analysis;

[0043] Figure 4 Schematic diagram of the process of obtaining deformation eigenvalues. DETAILED DESCRIPTION

[0044] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. It should be understood that, unless otherwise specified, " / " means or.

[0046] It should also be noted that the terms "first" and "second" in the present application are used to distinguish similar objects rather than to describe a specific order or sequence.

[0047] The specific scheme of the railway deformation monitoring method based on Beidou satellite positioning technology provided by this application is described in detail below with reference to the accompanying drawings.

[0048] An embodiment of the present application provides a method for monitoring deformation along a railway based on Beidou satellite positioning technology. Specifically, the following method for monitoring deformation along a railway based on Beidou satellite positioning technology is provided. Figure 1 , the method comprises the following steps:

[0049] Step 1: Use Beidou satellite positioning technology to obtain the vertical monitoring value of each monitoring result at any monitoring station along the railway within a preset time range, and obtain the mortar layer temperature data at the corresponding moment of each monitoring result.

[0050] The purpose of this application is to accurately extract the settlement characteristics of the environmental area along the railway, avoid reducing the settlement characteristics of the environmental area along the railway while accurately eliminating the interference of environmental noise, thereby effectively improving the accuracy of deformation monitoring along the railway.

[0051] Build the Beidou monitoring and collection system architecture, specifically: deploy Beidou base stations and Beidou monitoring stations along the railway in a local area, where the Beidou base station is deployed within 2km of the high-speed railway, and dual Beidou base stations are set up in the open sky, with 2 base stations; at the same time, deploy Beidou monitoring stations on the trackside of the railway, with 700m between two adjacent Beidou monitoring stations, with 3 Beidou monitoring stations; the Beidou base station and Beidou monitoring station are respectively connected to the server in communication, and the server is connected to the monitoring terminal in communication.

[0052] The dual Beidou base stations use a dual-base station short baseline carrier phase double difference solution with a baseline length of 2km. After the solution results are subjected to network adjustment solution, high-precision Beidou monitoring results are obtained. Among them, the dual-base station short baseline carrier phase double difference solution and network adjustment solution are both well-known technologies and will not be repeated in this application.

[0053] The Beidou monitoring results include monitoring values ​​in the north-south, east-west and vertical directions, wherein vertical refers to the direction perpendicular to the ground. The vertical monitoring values ​​of all Beidou monitoring results within a preset time range are obtained through the monitoring terminal, and all vertical monitoring values ​​of any monitoring station are arranged in time sequence to form a vertical monitoring sequence of any monitoring station.

[0054] In this embodiment, the preset time range is the ten consecutive days before the current day. The value of the preset time range is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0055] While obtaining the vertical monitoring value of any one of the monitoring stations, the temperature of the mortar layer near any one of the monitoring stations is collected through a temperature sensor. At the corresponding moment of each vertical monitoring value of any one of the monitoring stations, the mortar layer temperature data is collected and arranged in time sequence to form a temperature monitoring sequence of any one of the monitoring stations.

[0056] Step 2: Determine the settlement trend characteristic value of each vertical monitoring value based on the change of each vertical monitoring value and multiple vertical monitoring values ​​that are close to it in time.

[0057] Normally, due to the complex environment along the railway, changes in the temperature of the mortar layer can cause obvious regional settlement characteristics, resulting in deformation along the railway, causing serious safety hazards to the high-speed railway foundation. In the deformation monitoring along the railway using Beidou satellite positioning technology, the quality of Beidou monitoring results is affected by high occlusion and strong electromagnetic interference, which can easily lead to abnormal monitoring data and large errors, making the subsequent deformation monitoring along the railway less accurate. Therefore, in order to effectively improve the accuracy of deformation monitoring along the railway, it is necessary to analyze the abnormalities of Beidou positioning monitoring data.

[0058] In order to analyze the instantaneous characteristics of the monitoring data, windows of preset sizes are constructed with each vertical monitoring value in the vertical monitoring sequence as the center. All the vertical monitoring values ​​in each window are arranged in time series to form an instantaneous monitoring sequence for each vertical monitoring value in the vertical monitoring sequence. In particular, when there is insufficient data in the window, the data filling method is used to fill the missing values.

[0059] In this embodiment, the size of the window is The size of the window is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0060] In this embodiment, the mean filling method is used to fill the missing values. As other implementation methods, on the basis of being able to fill the missing values, the implementer can use other existing technologies to fill the missing values, such as the median filling method, interpolation method, etc., and this application does not make any special restrictions.

[0061] Normally, when regional subsidence occurs near the railway line in a local area, the vertical monitoring data will show obvious trend changes. When the vertical monitoring data at a certain moment shows obvious fluctuations, it cannot reflect the regional subsidence characteristics. The more obvious the abnormal fluctuation characteristics of the monitoring data in this direction, the more likely it is caused by external interference. The schematic diagram of the vertical monitoring value change curve is as follows: Figure 2 As shown in the figure, the horizontal axis is the monitoring sequence number, that is, the temporal order of the vertical monitoring values, and the vertical axis is the vertical monitoring values. The change curve ① of the vertical monitoring values ​​has obvious fluctuations, indicating that it is greatly affected by external interference, resulting in large abnormal fluctuations in the vertical monitoring values; the change curve ② of the vertical monitoring values ​​has obvious stability characteristics, indicating that the distribution of vertical monitoring values ​​changes when there is no regional settlement; the change curve ③ of the vertical monitoring values ​​has obvious trend changes, indicating that the distribution of vertical monitoring values ​​changes when regional settlement occurs.

[0062] Based on the above analysis, the settlement trend characteristic value of each vertical monitoring value is determined based on the change trend of all vertical monitoring values ​​in the instantaneous monitoring sequence of each vertical monitoring value. The expression is:

[0063] ; In the formula, is the settlement trend characteristic value of the ith vertical monitoring value in the vertical monitoring sequence; , , They are respectively the number of negative values, the number of positive values, and the sum of all elements in the first-order difference sequence of the instantaneous monitoring sequence of the i-th vertical monitoring value in the vertical monitoring sequence; wherein, the calculation of the first-order difference sequence is a well-known technology and will not be repeated in this application.

[0064] It should be noted that: the greater the difference between the number of negative values ​​and the number of positive values ​​in the first-order difference sequence of the instantaneous monitoring sequence, the more obvious the trend change of the monitoring data in the vertical direction, including the growth trend change and the decline trend change of the monitoring data. At the same time, the greater the absolute value of the sum of all elements in the first-order difference sequence of the instantaneous monitoring sequence, the greater the degree of change of the growth trend or decline trend of the monitoring data. To a certain extent, the greater the trend characteristics of regional settlement near the railway line, the greater the settlement trend characteristic value.

[0065] Step 3: Determine the abnormal fluctuation characteristic value of each vertical monitoring value based on the fluctuation of each vertical monitoring value and multiple vertical monitoring values ​​that are temporally adjacent to each other after detrending.

[0066] Due to the complex environment along the railway, the quality of Beidou monitoring data is affected by high occlusion and strong electromagnetic interference, resulting in abnormal fluctuations in Beidou monitoring data. At this time, the quality of Beidou monitoring data becomes unreliable, making it impossible to accurately extract the regional settlement characteristics near the railway, and thus unable to effectively improve the accuracy of deformation monitoring along the railway.

[0067] In order to analyze the abnormal fluctuation characteristics of Beidou monitoring data, the instantaneous monitoring sequence of each vertical monitoring value in the vertical monitoring sequence is used as the input of the Detrended Fluctuation Analysis (DFA) method, and the detrended monitoring sequence of each vertical monitoring value is output. The DFA method is a well-known technology and will not be described in detail in this application. Figure 3 As shown, the figure includes a change curve of the original data before detrending and a fluctuation curve after detrending. On the change curve of the original data before detrending, the horizontal axis is the monitoring time, and the vertical axis is the normalized value of vertical monitoring, that is, the normalized value of the vertical monitoring value; on the fluctuation curve after detrending, the horizontal axis is the monitoring time, and the vertical axis is the fluctuation value of vertical monitoring, that is, the normalized value of the detrending result of the vertical monitoring value. In this embodiment, the vertical monitoring value is normalized and the detrending result of the vertical monitoring value is normalized by using the extreme difference normalization method.

[0068] Based on the discrete degree and range of all data in the detrended monitoring sequence of each vertical monitoring value, the abnormal fluctuation characteristic value of each vertical monitoring value is determined, and the expression is:

[0069] ; In the formula, is the abnormal fluctuation characteristic value of the i-th vertical monitoring value in the vertical monitoring sequence; , , They are respectively the discrete degree, maximum data and minimum data of all data in the detrended monitoring sequence of the i-th vertical monitoring value in the vertical monitoring sequence.

[0070] In this embodiment, the degree of discreteness of all data in the detrended monitoring sequence is the information entropy. As other implementation methods, on the basis of being able to measure the degree of uneven distribution of all data in the detrended monitoring sequence, the implementer may adopt other existing technologies for measurement, such as variance, standard deviation, coefficient of variation, etc., and this application does not impose any special restrictions.

[0071] It should be noted that the greater the discreteness and range of all data in the detrended monitoring sequence, the more it can reflect the degree of chaos of the data in the detrended monitoring sequence. At this time, the Beidou monitoring data is affected by external interference. The greater the abnormal fluctuation characteristics of the monitoring data, the greater the abnormal fluctuation characteristic value, that is, the less reliable the quality of the monitoring data.

[0072] Step 4: Determine the settlement characteristic value of each vertical monitoring value based on the settlement trend characteristic value and the abnormal fluctuation characteristic value.

[0073] In order to reduce the impact of external interference on Beidou monitoring data, and thus more accurately extract the regional subsidence characteristics near the railway line, the abnormal fluctuation characteristics of the monitoring data are used to construct the credible weights of the subsidence trend characteristics, thereby reducing the adverse interference caused by external environmental interference to the regional subsidence characteristics.

[0074] Based on the abnormal fluctuation characteristic value, the credible weight of the settlement trend characteristic of each vertical monitoring value in the vertical monitoring sequence is determined, and the expression is:

[0075] ; In the formula, is the credible weight of the settlement trend characteristic of the ith vertical monitoring value in the vertical monitoring sequence; is the normalization function; is the abnormal fluctuation characteristic value of the i-th vertical monitoring value in the vertical monitoring sequence.

[0076] In this embodiment, the Min-Max normalization method is used to normalize the abnormal fluctuation characteristic values. As other implementation methods, on the basis of being able to realize the normalization of the abnormal fluctuation characteristic values, the implementer may use other existing technologies to normalize the abnormal fluctuation characteristic values, such as the Z-Score normalization method, the decimal calibration normalization method, etc., and this application does not impose any special restrictions.

[0077] It should be noted that the smaller the abnormal fluctuation characteristic value of the i-th vertical monitoring value in the vertical monitoring sequence, to a certain extent, it means that the Beidou monitoring data is less affected by external interference at this time, and the higher the credibility of the monitoring data at this time, that is, the greater the credibility of the settlement trend characteristics, the greater the credibility weight.

[0078] Furthermore, by combining the credible weight and the settlement trend characteristic value, the settlement characteristic value of each vertical monitoring value in the vertical monitoring sequence is obtained, and the expression is:

[0079] ; In the formula, is the settlement characteristic value of the ith vertical monitoring value in the vertical monitoring sequence; is the credible weight of the settlement trend characteristic of the ith vertical monitoring value in the vertical monitoring sequence; is the settlement trend characteristic value of the i-th vertical monitoring value in the vertical monitoring sequence.

[0080] It should be noted that the larger the credible weight and the larger the settlement trend characteristic value, the more it can reflect the regional settlement characteristics in the area along the railway, and to a certain extent avoid the interference caused by external interference. At this time, the higher the credibility of the settlement characteristic value, the larger the settlement characteristic value.

[0081] Step 5: Determine the deformation characteristic value of each settlement characteristic value by combining the discreteness and average level of each settlement characteristic value and multiple settlement characteristic values ​​that are temporally adjacent to each other.

[0082] Generally, due to factors such as the dynamic load of trains along the railway and the temperature change of the mortar layer, the greater the regional settlement characteristics in the area along the railway and the greater the fluctuation of the regional settlement characteristics, the higher the complexity of the regional settlement characteristics along the railway, and the more likely it is to damage the structure of the area along the railway, causing deformation along the railway. Therefore, in order to more accurately extract the deformation characteristics along the railway, further analysis based on the settlement characteristic values ​​is required.

[0083] The settlement characteristic values ​​of all vertical monitoring values ​​in the vertical monitoring sequence are arranged in time sequence to form a settlement characteristic sequence of any monitoring station. For the settlement characteristic sequence, the same acquisition method as the instantaneous monitoring sequence of each vertical monitoring value in the vertical monitoring sequence is used to obtain the settlement characteristic local sequence of each settlement characteristic value in the settlement characteristic sequence.

[0084] Based on the above analysis, combined with the discrete degree and mean of all settlement characteristic values ​​in the local sequence of settlement characteristics of each settlement characteristic value, the deformation characteristic value of each settlement characteristic value is determined, and the expression is:

[0085] Compute the deformation eigenvalue for each element in the settlement eigensequence:

[0086] ; In the formula, is the deformation eigenvalue of the ith settlement eigenvalue in the settlement eigenvalue sequence; , are the discrete degree and mean of all the settlement characteristic values ​​in the settlement characteristic local sequence of the ith settlement characteristic value in the settlement characteristic sequence, respectively; is the normalization function.

[0087] In this embodiment, the degree of dispersion of all sedimentation feature values ​​within the local sequence of sedimentation features is the standard deviation. As other implementation methods, on the basis of being able to measure the uneven distribution of all sedimentation feature values ​​within the local sequence of sedimentation features, the implementer may adopt other existing technologies for measurement, such as variance, information entropy, coefficient of variation, etc., and this application does not impose any special restrictions.

[0088] In this embodiment, the Min-Max normalization method is used to Normalization is performed, as another implementation method, in which Based on the normalization process, implementers can use other existing technologies to Normalization processing is performed, such as Z-Score normalization method, decimal calibration normalization method, etc., and this application does not make any special restrictions.

[0089] It should be noted that the larger the mean of all settlement characteristic values ​​in the local sequence of settlement characteristics and the greater the degree of dispersion, to a certain extent, reflects that the regional settlement characteristics are larger and the fluctuation of regional settlement characteristics is larger. At this time, the complexity of regional settlement characteristics along the railway is higher. The higher the significance of deformation characteristics along the railway, the larger the deformation characteristic value. The schematic diagram of the process of obtaining deformation characteristic values ​​is shown in the figure. Figure 4 shown.

[0090] Step 6: Train the neural network based on all vertical monitoring values, all mortar layer temperature data and the deformation characteristic values ​​within the preset time range, and use the trained neural network to obtain the deformation monitoring results of the railway at any monitoring station based on all vertical monitoring values ​​and all mortar layer temperature data of the day.

[0091] Due to the influence of the temperature change of the mortar layer, the monitoring stations along the railway have regional settlement characteristics in the vertical direction, and deformation occurs along the railway. The present application uses a neural network prediction model to accurately monitor the deformation phenomenon along the railway, taking the vertical monitoring sequence and temperature monitoring sequence of any monitoring station within the preset time range as the training set, and taking the sequence composed of the deformation characteristic values ​​of all settlement characteristic values ​​in the settlement characteristic sequence as the label sequence of the training set, wherein the lengths of the vertical monitoring sequence, the temperature monitoring sequence and the label sequence are the same;

[0092] The training set and the label sequence of the training set are used as the input of the LSTM neural network model. The activation function uses the ReLU function and the optimizer uses the Adam optimizer. The loss function uses the mean square error function, and finally the trained LSTM neural network model is obtained.

[0093] The vertical monitoring sequence and temperature monitoring sequence of any monitoring station on that day are obtained through the monitoring terminal, and the vertical monitoring sequence and temperature monitoring sequence of any monitoring station on that day are used as the input of the trained LSTM neural network model. The trained LSTM neural network model is used to predict the deformation characteristics along the railway. The LSTM neural network model outputs a label sequence with the deformation characteristics along the railway, wherein the training and prediction of the neural network model are well-known technologies and will not be repeated in this application.

[0094] The tag sequence with deformation characteristics along the railway is used as the deformation monitoring result at any monitoring station along the railway. The deformation monitoring result along the railway can provide decision support for railway maintenance and safety management, and help relevant departments take timely measures to prevent possible structural damage and safety accidents.

[0095] In summary, the existing methods of noise reduction for Beidou positioning data can easily reduce the settlement characteristics of the environmental area along the railway, making it impossible to effectively improve the accuracy of deformation monitoring along the railway, resulting in poor accuracy of deformation monitoring along the railway. Therefore, this application determines the settlement trend characteristic value of each vertical monitoring value based on the changes in each vertical monitoring value and its multiple vertical monitoring values ​​that are close in time; characterizes the trend characteristics of regional settlement near the railway, which is conducive to the subsequent accurate extraction of deformation characteristics along the railway;

[0096] Further, based on the fluctuation of each vertical monitoring value and multiple vertical monitoring values ​​that are temporally adjacent after detrending, the abnormal fluctuation characteristic value of each vertical monitoring value is determined; based on the settlement trend characteristic value and the abnormal fluctuation characteristic value, the settlement characteristic value of each vertical monitoring value is determined; according to the abnormal fluctuation characteristics of the vertical monitoring values, the credible weight of the settlement trend characteristic is obtained, and the settlement trend characteristic value is weighted to eliminate the adverse interference caused by the external environment interference, so that the extracted settlement characteristics are more real and effective, and the interference effect of noise on the accuracy of deformation monitoring along the railway is avoided;

[0097] Furthermore, the deformation characteristic value of each settlement characteristic value is determined by combining the discrete degree and average level of each settlement characteristic value and multiple settlement characteristic values ​​of its temporal neighbors; the complexity of regional settlement characteristics along the railway is analyzed to improve the accuracy of deformation monitoring along the railway;

[0098] Furthermore, the neural network model is used to monitor the deformation along the railway based on the deformation characteristic value. This application avoids the problem of reducing the settlement characteristics of the environmental area along the railway, and accurately eliminates the interference of the external environment, which can effectively improve the accuracy of deformation monitoring along the railway.

[0099] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0100] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the basic features of the present application. Therefore, no matter from which point of view, the above embodiments of the present application should be regarded as exemplary and non-restrictive.

Claims

1. A method for monitoring deformation along a railway based on Beidou satellite positioning technology, characterized in that: The method comprises the following steps: Use Beidou satellite positioning technology to obtain the vertical monitoring value of each monitoring result at any monitoring station along the railway within a preset time range, and obtain the mortar layer temperature data at the corresponding time of each monitoring result; Based on the changes of each vertical monitoring value and multiple vertical monitoring values ​​that are close to it in time, determine the settlement trend characteristic value of each vertical monitoring value; Based on the fluctuation of each vertical monitoring value and multiple vertical monitoring values ​​that are temporally adjacent to each other after detrending, the abnormal fluctuation characteristic value of each vertical monitoring value is determined; Determining the settlement characteristic value of each vertical monitoring value based on the settlement trend characteristic value and the abnormal fluctuation characteristic value; Determine the deformation characteristic value of each settlement characteristic value by combining the discrete degree and average level of each settlement characteristic value and multiple settlement characteristic values ​​that are close to each other in time; The neural network is trained based on all vertical monitoring values, all mortar layer temperature data and the deformation characteristic values ​​within the preset time range, and based on all vertical monitoring values ​​and all mortar layer temperature data of the day, the deformation monitoring result of the railway at any monitoring station is obtained using the trained neural network.

2. The method for monitoring deformation along a railway based on Beidou satellite positioning technology according to claim 1, characterized in that: The process of determining the characteristic value of the subsidence trend is as follows: Arrange all vertical monitoring values ​​within the preset time range in time sequence to form a vertical monitoring sequence; A window of a preset size is constructed with each vertical monitoring value in the vertical monitoring sequence as the center, and all vertical monitoring values ​​in the window are arranged in time sequence to form an instantaneous monitoring sequence; The expression of the settlement trend characteristic value of each vertical monitoring value in the vertical monitoring sequence is: ; In the formula, is the settlement trend characteristic value of the ith vertical monitoring value in the vertical monitoring sequence; , , They are respectively the number of negative values, the number of positive values, and the sum of all elements in the first-order difference sequence of the instantaneous monitoring sequence of the i-th vertical monitoring value in the vertical monitoring sequence.

3. The method for monitoring deformation along a railway based on Beidou satellite positioning technology as claimed in claim 2, characterized in that: The process of determining the abnormal fluctuation characteristic value is as follows: Detrended fluctuation analysis method is used to detrend each instantaneous monitoring sequence to obtain the detrended monitoring sequence of each vertical monitoring value; The abnormal fluctuation characteristic value is positively correlated with the discreteness and the range of all elements in the detrended monitoring sequence.

4. The method for monitoring deformation along a railway based on Beidou satellite positioning technology according to claim 1, characterized in that: The process of determining the settlement characteristic value is as follows: Determining the credible weight of the settlement trend characteristic of each vertical monitoring value in the vertical monitoring sequence based on the abnormal fluctuation characteristic value; The settlement characteristic value of each vertical monitoring value in the vertical monitoring sequence is obtained by combining the credible weight and the settlement trend characteristic value.

5. The method for monitoring deformation along a railway based on Beidou satellite positioning technology as claimed in claim 4, characterized in that: The credible weight is the difference between 1 and the normalized value of the abnormal fluctuation characteristic value.

6. The method for monitoring deformation along a railway based on Beidou satellite positioning technology according to claim 1, characterized in that: The process of determining the deformation characteristic value is as follows: Arranging the sedimentation characteristic values ​​in time sequence to form a sedimentation characteristic sequence; For the settlement feature sequence, the same acquisition method as the instantaneous monitoring sequence of each vertical monitoring value in the vertical monitoring sequence is adopted to acquire the settlement feature local sequence of each settlement feature value in the settlement feature sequence; The deformation characteristic value is determined by the discrete degree and the mean value of all the settlement characteristic values ​​in the settlement characteristic local sequence of each settlement characteristic value.

7. The method for monitoring deformation along a railway based on Beidou satellite positioning technology as claimed in claim 6, characterized in that: The expression of the deformation characteristic value is: ; In the formula, is the deformation eigenvalue of the ith settlement eigenvalue in the settlement eigenvalue sequence; , are the discrete degree and mean of all the settlement characteristic values ​​in the settlement characteristic local sequence of the ith settlement characteristic value in the settlement characteristic sequence, respectively; is the normalization function.

8. The method for monitoring deformation along a railway based on Beidou satellite positioning technology as claimed in claim 2, characterized in that: The training set for training the neural network includes: a temperature monitoring sequence composed of all mortar layer temperature data within the preset time range arranged in time sequence, and a vertical monitoring sequence.

9. The method for monitoring deformation along a railway based on Beidou satellite positioning technology as claimed in claim 8, characterized in that: The sequence composed of the deformation feature values ​​arranged in time sequence is used as the label sequence of the training set.

10. The method for monitoring deformation along a railway based on Beidou satellite positioning technology according to claim 9, characterized in that: The deformation monitoring result of the railway at any of the monitoring stations is: a label sequence with deformation characteristics along the railway obtained by using a trained neural network.

Images