Train speed and mileage estimation method and device based on detection data of adder and passenger instrument

Abstract

The invention discloses a train speed and mileage estimation method and device based on add-ride instrument detection data, which can be used in the field of portable add-ride instruments, and the method comprises the following steps: acquiring train body longitudinal acceleration, train body head shaking angular velocity and curve superelevation data of a train collected by an add-ride instrument; integral calculation is conducted on the longitudinal acceleration of the train body, and the speed estimated value and the mileage estimated value of the train are determined; extracting a trend term of the head shaking angular velocity of the vehicle body, calculating an optimal mapping path between the trend term and a reference datum by adopting a derivative dynamic time warping algorithm by taking curve superelevation data as the reference datum, and determining a mileage deviation value corresponding to the head shaking angular velocity of the vehicle body under the optimal mapping path; and correcting the speed estimation value and the mileage estimation value according to the mileage deviation value, determining the corrected speed estimation value as the speed information of the train, and determining the corrected mileage estimation value as the mileage information of the train. According to the invention, the train speed information and mileage information can be accurately estimated, and the detection convenience of the passenger adding instrument is improved.

Description

Technical Field

[0001] The present invention relates to the field of portable ride-on inspection instruments, and particularly to a method and device for estimating train speed and mileage based on inspection data of ride-on inspection instruments. Background Art

[0002] This section aims to provide background or context for embodiments of the present invention. The descriptions herein are not admitted to be prior art merely because they are included in this section.

[0003] In the past, low-configuration ride-on inspection instruments were only equipped with three-axis acceleration sensors. The mileage information corresponding to abnormal vehicle vibrations (such as vehicle shaking and swaying) was determined by manually observing the line mileage posts. Affected by factors such as the position of mileage posts and viewing conditions, the mileage accuracy recorded manually was low and speed information could not be obtained. With the continuous development of Internet of Things technology, high-configuration ride-on inspection instruments are equipped with six-axis inertial sensors, microphones, cameras and Beidou positioning modules, which can simultaneously sense data such as vehicle running quality, interior noise and external environment of vehicle operation. The speed information and mileage information of the inspection data are obtained by connecting with the train or the Beidou positioning module, which limits the ride-on inspection instrument to be placed only at positions where it is convenient to obtain train speed information and mileage information, and the Beidou positioning module is vulnerable to interference from factors such as weather, tunnels and stations, resulting in the lack of train speed information and mileage information corresponding to inspection data in special sections. Summary of the Invention

[0004] Embodiments of the present invention provide a method for estimating train speed and mileage based on inspection data of a ride-on inspection instrument, which is used to accurately estimate train speed information and mileage information and improve the convenience of ride-on inspection instrument detection. The method includes:

[0005] Obtain the longitudinal acceleration of the train body, the yaw angular velocity of the train body and the curve superelevation data collected by the ride-on inspection instrument;

[0006] Perform integral calculation on the longitudinal acceleration of the train body to determine the speed estimate and mileage estimate of the train;

[0007] Extract the trend term of the yaw angular velocity of the train body, use the curve superelevation data as a reference benchmark, and calculate the best mapping path between the trend term and the reference benchmark by using the derivative dynamic time warping algorithm to determine the mileage deviation amount corresponding to the yaw angular velocity of the train body under the best mapping path;

[0008] Correct the speed estimate and mileage estimate according to the mileage deviation amount, determine the corrected speed estimate as the train speed information, and determine the corrected mileage estimate as the train mileage information.

[0009] Embodiments of the present invention also provide a device for estimating train speed and mileage based on inspection data of a ride-on inspection instrument, which is used to accurately estimate train speed information and mileage information and improve the convenience of ride-on inspection instrument detection. The device includes:

[0010] A data acquisition module, configured to obtain the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the riding instrument;

[0011] An integral calculation module, configured to perform integral calculation on the longitudinal acceleration of the train body to determine the speed estimation value and the mileage estimation value of the train;

[0012] A deviation calculation module, configured to extract the trend term of the yaw angular velocity of the train body, take the curve superelevation data as a reference benchmark, and use the derivative dynamic time warping algorithm to calculate the optimal mapping path between the trend term and the reference benchmark, and determine the mileage deviation amount corresponding to the yaw angular velocity of the train body under the optimal mapping path;

[0013] A correction module, configured to correct the speed estimation value and the mileage estimation value according to the mileage deviation amount, determine the corrected speed estimation value as the speed information of the train, and determine the corrected mileage estimation value as the mileage information of the train.

[0014] An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned method for estimating the speed and mileage of a train based on the detection data of a riding instrument is implemented.

[0015] An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method for estimating the speed and mileage of a train based on the detection data of a riding instrument is implemented.

[0016] An embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the above-mentioned method for estimating the speed and mileage of a train based on the detection data of a riding instrument is implemented.

[0017] In an embodiment of the present invention, the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the ride-on instrument are obtained; the longitudinal acceleration of the train body is integrated to calculate the speed estimate and mileage estimate of the train; the trend term of the yaw angular velocity of the train body is extracted, and taking the curve superelevation data as the reference benchmark, the derivative dynamic time warping algorithm is used to calculate the best mapping path between the trend term and the reference benchmark, and the mileage deviation amount corresponding to the yaw angular velocity of the train body under the best mapping path is determined; the speed estimate and mileage estimate are corrected according to the mileage deviation amount, the corrected speed estimate is determined as the speed information of the train, and the corrected mileage estimate is determined as the mileage information of the train. In this way, the ride-on instrument is used to calculate accurate train speed information and mileage information that meet the on-site verification requirements, greatly improving the convenience and intelligence of ride-on instrument detection. Considering the local "shape" information of the yaw angular velocity of the train body and the curve superelevation data, the alignment accuracy between the yaw angular velocity of the train body and the curve superelevation data is improved through the derivative dynamic time warping (DDTW) algorithm, and the mileage deviation amount of the yaw angular velocity data of the train body under the best mapping path is accurately calculated, so as to correct the train speed and mileage errors, accurately estimate the train speed information and mileage information, and improve the convenience of ride-on instrument detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:

[0019] Figure 1 is a flowchart of a method for estimating the speed and mileage of a train based on ride-on instrument detection data provided in an embodiment of the present invention;

[0020] Figure 2 is a schematic diagram of a data calculation model provided in an embodiment of the present invention;

[0021] Figure 3 is an example diagram of ride-on instrument detection data provided in an embodiment of the present invention;

[0022] Figure 4 is an example diagram of ledger superelevation and gradient information provided in an embodiment of the present invention;

[0023] Figure 5 is a schematic diagram showing the overall offset and local stretching or compression between the yaw angular velocity and the superelevation provided in an embodiment of the present invention;

[0024] Figure 6 is a schematic diagram of a spatial sequence to be aligned provided in an embodiment of the present invention;

[0025] Figure 7 Schematic diagram of DDTW matching path provided in the embodiment of the present invention;

[0026] Figure 8 Schematic diagram of initial correction of train speed and mileage provided in the embodiment of the present invention;

[0027] Figure 9 Example diagram of train speed and mileage correction provided in the embodiment of the present invention;

[0028] Figure 10 Schematic diagram of train speed and mileage estimation device based on inspection data of ride-on instrument provided in the embodiment of the present invention;

[0029] Figure 11 Block diagram of the computer device provided in the embodiment of the present invention. Detailed implementation manners

[0030] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer and more understandable, the following further elaborates on the embodiments of the present invention in conjunction with the accompanying drawings. Herein, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but do not limit the present invention.

[0031] The term "and / or" in this article merely describes an association relationship and indicates that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" in this article means any one of multiple or any combination of at least two of multiple. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set composed of A, B, and C.

[0032] In the description of this specification, the terms "include", "comprise", "have", "contain", etc. are all open-ended terms, that is, they are intended to include but not limited to. The descriptions with reference to terms such as "one embodiment", "one specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. The step sequences involved in each embodiment are used to schematically illustrate the implementation of the present application, and the step sequences are not limited and can be adjusted appropriately as needed.

[0033] Aiming at the problem that the traditional on-vehicle line inspection instrument cannot independently calculate train speed information and mileage information, the embodiment of the present invention provides a train speed and mileage estimation method based on inspection data of ride-on instrument,Figure 1 Schematic diagram of a train speed and mileage estimation method based on inspection data of a multiplication instrument, as Figure 1 shown, including:

[0034] Step 101: Obtain the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the multiplication instrument;

[0035] Step 102: Integrate the longitudinal acceleration of the train body to determine the speed estimate and mileage estimate of the train;

[0036] Step 103: Extract the trend term of the yaw angular velocity of the train body. Taking the curve superelevation data as the reference benchmark, use the derivative dynamic time warping algorithm to calculate the optimal mapping path between the trend term and the reference benchmark, and determine the mileage deviation amount corresponding to the yaw angular velocity of the train body under the optimal mapping path;

[0037] Step 104: Correct the speed estimate and mileage estimate according to the mileage deviation amount to determine the speed information and mileage information of the train.

[0038] The train speed and mileage estimation method based on inspection data of a multiplication instrument proposed in the embodiments of the present invention mainly involves two core aspects: (1) By means of the multiplication instrument, accurate train speed information and mileage information that meet the requirements of on-site review are calculated, and there is no need to obtain train speed information and mileage information through methods such as operating trains or Beidou positioning modules, which greatly improves the convenience and intelligence of multiplication instrument inspection. (2) The derivative dynamic time warping (DDTW) algorithm can consider the local "shape" information of the yaw angular velocity of the train body and the curve superelevation data, improve the alignment accuracy between the yaw angular velocity of the train body and the curve superelevation data, accurately calculate the mileage deviation amount of the yaw angular velocity data of the train body under the optimal mapping path, and use this to correct the train speed and mileage errors.

[0039] In specific implementation, based on the longitudinal acceleration of the car body, the yaw angular velocity of the car body, and the curve superelevation data in the line ledger detected by a single ride-on instrument, the longitudinal acceleration of the car body is integrated to calculate the train speed information and mileage information. The integration error caused by influencing factors such as sensor performance and line gradient is analyzed, resulting in an overall offset, local stretching, or compression of the yaw angular velocity of the car body in the mileage dimension. Using the correlation between the yaw angular velocity of the car body and the curve superelevation data in the line ledger, with the curve superelevation data as the reference benchmark, the Derivative Dynamic Time Warping (DDTW) algorithm is used to calculate the optimal mapping path between the yaw angular velocity of the car body and the curve superelevation data, and the mileage deviation of the yaw angular velocity data under the optimal mapping path is determined. Based on this, the initial calculation errors of the train speed information and mileage information are corrected, and the positioning accuracy of the detection data of the ride-on instrument is improved. Then, a field test is carried out, and the train speed information and mileage information independently estimated by a single ride-on instrument are compared with the comprehensive speed information and mileage information of a high-speed comprehensive inspection train (dynamic inspection car) to verify the accuracy of the detection data of the ride-on instrument sensor.

[0040] In one embodiment, obtaining the longitudinal acceleration of the car body, the yaw angular velocity of the car body, and the curve superelevation data of the train collected by the ride-on instrument includes:

[0041] Obtaining the longitudinal acceleration of the car body, the yaw angular velocity of the car body, and the curve superelevation data of the train synchronously collected by the ride-on instrument in the time dimension; the timestamps of the longitudinal acceleration of the car body, the yaw angular velocity of the car body, and the curve superelevation data are consistent.

[0042] In specific implementation, the ride-on instrument is built-in with a six-axis inertial sensor and a microphone, and can synchronously collect the three-way acceleration (longitudinal acceleration, lateral acceleration, and vertical acceleration), three-axis angular velocity (roll, pitch, and yaw angular velocity) of the train car body, and the in-car noise in time, a total of 7-channel data. Figure 2 This is a schematic diagram of the data calculation model provided in the embodiment of the present invention. The speed information and mileage information of multi-source detection data at any moment cannot be directly obtained, and need to be estimated through the train speed information and mileage calculation model to improve the on-site review efficiency of the train vibration and noise overrun sections. The architecture of this calculation model is as Figure 2 shown, Figure 2 The corresponding calculation process in it includes:

[0043] 1. Data input of the calculation model. The data sources for train speed and mileage estimation include the detection data of the ride-on instrument (intelligent ride-on instrument) and the ledger data, specifically the longitudinal acceleration and yaw angular velocity of the car body detected by the intelligent ride-on instrument, and the curve superelevation and mileage data used as the reference benchmark.

[0044] In the embodiment of the present invention, the intelligent riding instrument is built-in with a six-axis inertial sensor and a microphone sensor, which can synchronously collect the three-axis acceleration, three-axis angular velocity of the train car body and the in-car noise data in time, so that the time stamps of the sensing data of each channel are consistent. The relevant technical parameters are shown in Table 1. During the detection process of the train vibration and noise data, the intelligent riding instrument can be directly placed on the surface of the train car floor. When placing it, the coordinate system of the intelligent riding instrument (inertial sensor coordinate system) needs to be kept parallel to the car body coordinate system to reduce the measurement error introduced by the placement attitude angle of the intelligent riding instrument.

[0045] Table 1 Technical parameters of the intelligent riding instrument

[0046]

[0047] To verify the accuracy of the train speed and mileage estimation based on the detection data of the intelligent riding instrument, the intelligent riding instrument is placed on the dynamic inspection vehicle for data collection, and compared with the speed, mileage and car body acceleration data of the dynamic inspection vehicle. Figure 3 This is an example diagram of the detection data of the riding instrument provided in the embodiment of the present invention. Taking a certain high-speed rail line as an example, the detection data of the intelligent riding instrument is as Figure 3 shown. Figure 3 This is an example diagram of the detection data of the riding instrument provided in the embodiment of the present invention. There are 7 channels in total, and the detection duration is about 1 h. From top to bottom, they are the three-axis acceleration of the car body, the three-axis angular velocity of the car body and the in-car noise data. It can be seen that the dynamic inspection vehicle stops and stands still for a period of time at stations A, B, and C; the longitudinal acceleration of the car body records the acceleration and deceleration amplitudes during the train operation, which is used for integrating and calculating the train operation speed and mileage information; the yaw angular velocity data of the car body contains significant trend item characteristics when the train passes through the curve. Based on this characteristic and the curve superelevation data in the line ledger, the train mileage deviation amount is calculated.

[0048] During the process of train speed and mileage estimation and correction, the line ledger superelevation and slope information are introduced. On the one hand, the curve superelevation information is the reference benchmark for correcting the train speed and mileage errors. On the other hand, the overall trend of the line slope determines the deviation trend of the initial estimation of the train speed and mileage. Figure 4 This is an example diagram of the ledger superelevation and slope information provided in the embodiment of the present invention. The ledger superelevation and slope information of a certain high-speed rail line is as Figure 4 shown. The train running direction is from station A to B to C, which is the down direction of the line. The X-axis of the intelligent riding instrument coordinate system points to the train running direction. Through statistics, it can be known that the total length of the high-speed rail line to be detected is about 161.5 km, the curve section is about 74 km, accounting for about 45.8% of the total length, and the number of curves is 57; the cumulative value of the overall slope of the line is -83.26‰, showing a downhill trend, and the component of the gravitational acceleration is consistent with the train running direction, which will cause the initial integral values of the train speed and mileage to be too large.

[0049] In one embodiment, the longitudinal acceleration of the car body is integrated to determine the speed estimate and mileage estimate of the train, including:

[0050] Perform a first integration calculation on the longitudinal acceleration of the car body to determine the speed estimate of the train;

[0051] Perform a second integration calculation on the longitudinal acceleration of the car body to determine the mileage estimate of the train.

[0052] 2. Initial estimation of train speed information and mileage information. The longitudinal acceleration of the car body at any moment is expressed as Perform the first and second integrations on it to calculate the speed and mileage information of the data detected by the ride-on instrument respectively. The calculations are as follows:

[0053]

[0054] In the formula, are respectively the initial estimated values of the train speed and mileage at time t i ; v0 and s0 are the initial speed and mileage information of the train. If the train is in a stationary state at time t0, both values are 0.

[0055] Affected by factors such as sensor performance, line profile, and the attitude angle of the intelligent ride-on instrument, the accuracy of the train speed information and mileage information obtained by integrating the longitudinal acceleration of the car body is relatively low, and the train speed information and mileage information need to be corrected. At the same time, since the mileage interval of the dynamic inspection vehicle's track geometry dynamic detection data is 0.25 m, that is, the mileage interval of the reference benchmark superelevation channel data is 0.25 m. Before performing the train speed and mileage error correction, it is necessary to linearly interpolate the yaw angular velocity data of the intelligent ride-on instrument at equal intervals of 0.25 m according to the initial mileage estimate value

[0056] In one embodiment, extracting the trend term of the car body yaw angular velocity includes:

[0057] Perform moving average filtering on the car body yaw angular velocity to obtain the trend term of the car body yaw angular velocity.

[0058] In one embodiment, taking the curve superelevation data as the reference benchmark, using the derivative dynamic time warping algorithm to calculate the best mapping path between the trend term and the reference benchmark, further including:

[0059] Normalize the trend term and the curve superelevation data.

[0060] ​3. Correction of train speed information and mileage information. Affected by the superelevation of the curve, when the train passes through the curve section, there is an obvious trend term in the angular velocity of the car body shaking. The trend term of the angular velocity of the car body shaking is extracted by moving average filtering. The derivative dynamic time warping (DDTW) algorithm is used to calculate the mileage deviation Δs between the data of the trend term of the angular velocity of the car body shaking and the reference benchmark superelevation data, and the train speed and mileage errors are corrected.

[0061] Figure 5 It is a schematic diagram showing the overall offset, local stretching or compression between the angular velocity of car body shaking and the superelevation in the embodiment of the present invention. Among them, ω is the angular velocity of car body shaking, V is the running speed of the train, X is the mileage information, K is the curvature, and R is the curve radius. According to kinematics and test data, when the vehicle passes through the curve section, there is a trend term in the data of the angular velocity of the car body shaking related to the curve superelevation characteristics, such as Figure 5 shown. Affected by the numerical integration of the sensor error amount, there is an obvious mileage deviation between the data of the angular velocity of the car body shaking and the curve superelevation, and the mileage deviation amounts at the 4 key points of the curve section are also different, such as Figure 5 the deviation amounts Δx1, Δx2, Δx3 and Δx4 in it, resulting in overall offset, stretching or compression of the detection data and deviation from the mileage in the account. Therefore, with the help of the superelevation information in the account as a reference benchmark, the mileage deviation at the position of the key feature points is calculated by the DDTW algorithm, the speed value is corrected in reverse, and then the speed and mileage errors of the detection data of the intelligent riding instrument are corrected to meet the on-site accuracy requirements.

[0062] Figure 6 It is a schematic diagram of the space sequence to be aligned in the embodiment of the present invention. Taking the curve superelevation information in the line account as a reference benchmark, moving average filtering is performed on the angular velocity of the car body shaking, and the filtering result and the curve superelevation data are normalized to obtain the space sequence to be aligned of the angular velocity of car body shaking and the curve superelevation data as shown in Figure 6 shown.

[0063] The traditional DTW algorithm is widely used. This algorithm can achieve the alignment of amplitudes by bending the X-axis, only considering the data amplitudes and not considering the advanced features based on "shape". Here, the original data of the angular velocity of the car body shaking and the account superelevation are not used, and the local derivatives of the two data are used. By considering the first-order derivative of the sequence to obtain information about the shape, that is, the DDTW algorithm. For the data of the angular velocity of the car body shaking with random noise, it is smoothed (moving average filtering) before calculating the derivative. Using this feature information, DDTW can find the stretching or compression amount of the waveforms with similar shapes by finding the optimal planning path.

[0064] To avoid the difference in the magnitude order of the data to be matched and the phenomenon of magnitude drowning, all data sequences are preprocessed by amplitude normalization. Then, the spatial domain car body yaw angular velocity sequence and the superelevation sequence of the line ledger curve after amplitude normalization are respectively represented as spatial domain vectors and Calculate the sequence derivative according to Equation (2) to form data sequences considering "shape", which are respectively and The calculation is as follows.

[0065]

[0066] Calculate and The distance matrix M(i,j) under a 200m moving window is calculated. The Euclidean distance method is used to calculate the distance between each point under the 200m moving window:

[0067]

[0068] To find the best alignment path under the 200m moving window, the starting and ending points of the best path of the distance matrix M are M(1,1) to M(m,n), and the best path w k ∈M(i,j). The optimization problem of the above best path can be expressed as solving Equation (4):

[0069]

[0070] This optimization problem can be calculated row by row (or column by column) from [1,1] to [m,n] and fill the table of the distance cumulative matrix γ. After filling the entire table, the shortest distance γ(m,n) between the two sequences can be obtained. This distance is usually also called the DDTW distance. The previous point that reaches [i,j] through the shortest cumulative distance must be one of [i - 1,j - 1], [i,j - 1], [i - 1,j]. On the premise that the shortest cumulative distances corresponding to γ[i - 1,j - 1], γ[i,j - 1], γ[i - 1,j] have been calculated respectively, the recurrence relation about γ[i,j] can be obtained as follows:

[0071] γ[i,j] = M[i,j] + min(γ[i - 1,j - 1],γ[i - 1,j],γ[i,j - 1]) (5)

[0072] The best path w k is the shortest cumulative distance of γ[i,j]. The warping path that meets the above conditions may not be unique. The number of paths K can be introduced into Equation (4) to compensate for the fact that the dynamic programming paths may have different lengths. The dynamic programming path that minimizes the cost is calculated as follows:

[0073]

[0074] The optimal path sequence number k determined by formula (6) corresponds to and the matching pair [i, j], and the difference between the two is the sequence and the reference benchmark The deviation amount, then the mileage deviation sequence is calculated as follows:

[0075] Δs k = Δs i,j = (i - j)·ds (7)

[0076] In the formula, Δs k is the mileage deviation amount of the k-th and of the optimal path under the moving window, and Δs i,j is used subsequently, where ds is taken as 0.25 m.

[0077] Figure 7 This is the schematic diagram of the DDTW matching path provided in the embodiment of the present invention. Taking the DDTW optimal mapping path in a certain curved section as an example, the mileage deviation amount is obtained, as shown in Figure 7 , and the speed and mileage information of the intelligent riding instrument detection data are corrected.

[0078] In one embodiment, the speed estimation value and the mileage estimation value are corrected according to the mileage deviation amount, and the corrected speed estimation value is determined as the speed information of the train, and the corrected mileage estimation value is determined as the mileage information of the train, including:

[0079] Correct the speed estimation value and the mileage estimation value according to the mileage deviation amount;

[0080] When the mileage deviation of the corrected mileage estimation value satisfies the mileage deviation limit value, the corrected speed estimation value is determined as the speed information of the train, and the corrected mileage estimation value is determined as the mileage information of the train; the mileage deviation limit value is preset based on production requirements.

[0081] According to the mileage deviation amount Δs i,j correct the mileage information of the car body yaw angular velocity so that the mileage of the intelligent riding instrument detection data at the i-th point under the optimal path is aligned with the reference benchmark curve superelevation data:

[0082] S′ i = S i + Δs i,j (8)

[0083] In the formula, S i , S′ i are the mileage before and after correction of the i-th point of the intelligent riding instrument detection data respectively, and S iIt can be calculated from Equation (1).

[0084] Combined with Equation (1), (8) and Δs i,j Calculate the speed of the i-th point of the intelligent riding detector data, and the specific calculation is as follows:

[0085]

[0086] In the formula, v i , v′ i are the speeds before and after correction of the i-th point of the intelligent riding detector data respectively, and v i It can be calculated from Equation (1).

[0087] 4. Data output of the calculation model. Affected by the number of railway line curves, it is difficult to make the mileage deviation be 0 at any time. Combining the actual situation of on-site review, set the allowable mileage deviation limit to C to ensure that the maximum value of the mileage deviation Δs max < C. When the yaw angular velocity of the car body in the curve section and the mileage deviation of the main points of curve superelevation (straight-slow point, slow-circular point, circular-slow point, slow-straight point) meet the allowable limit C, the speed information and mileage information of the intelligent riding detector data with higher accuracy can be output.

[0088] Using Figure 3 The longitudinal acceleration of the car body is directly integrated to calculate the initial information of the train speed and mileage. To avoid introducing integration errors in the detection data during the stationary phase of the inspection vehicle, the longitudinal acceleration of the car body during the movement process of the inspection vehicle is automatically identified and intercepted, and the detection data during the stationary phase of the inspection vehicle is not considered in the integral calculation. Figure 8 This is the schematic diagram of the initial correction of the train speed and mileage provided in the embodiment of the present invention, including: (a) Removal of the longitudinal acceleration of the car body and its trend term; (b) Comparison between the initial estimated value of the train speed and the comprehensive speed of the inspection vehicle; (c) Comparison between the yaw angular velocity of the initial estimated value of the train mileage and the ledger superelevation and the superelevation information of the inspection vehicle; (d) Initial correction of the train running speed based on the station mileage information; (e) Comparison between the yaw angular velocity of the initial correction of the train mileage and the ledger superelevation and the superelevation information of the inspection vehicle. Affected by multiple factors, there is a zero-offset trend term in the longitudinal acceleration of the car body, and the trend term needs to be removed before integral calculation, as shown in (a) of Figure 8 shown in

[0089] Figure 8Figure (b) shows the comparison between the initial estimated train speed from the inspection data of the ride-along instrument and the comprehensive speed of the dynamic inspection vehicle in the mileage dimension. After the dynamic inspection vehicle has traveled 50 km, the deviation between the initial estimated train speed and the comprehensive speed of the dynamic inspection vehicle gradually expands. This deviation is related to the overall slope trend of the high-speed railway line in the downward direction and belongs to a positive deviation. The dynamic inspection vehicle is equipped with a high-precision integrated positioning system and a track geometry inspection system. The measured curve superelevation data is consistent with the amplitude and mileage information of the superelevation data in the line ledger. Affected by the error factors of the inspection data of the ride-along instrument, there are obvious deviations between the car body yaw angular velocity in the mileage dimension and the superelevation data of the ledger and the dynamic inspection vehicle, as shown in Figure 8 Figure (c).

[0090] Based on the preprocessed car body yaw angular velocity and curve superelevation data, the DDTW algorithm is used to calculate the mileage deviation between the two, and the maximum mileage deviation reaches 1.8 km. Figure 9 Figure for the train speed and mileage correction example provided in the embodiment of the present invention, as shown in Figure 9 Figure (a). Combining formula (9) to correct the train speed curve estimated by the ride-along instrument further improves the accuracy of the train speed estimate, as shown in Figure 9 Figure (b). To verify the correction effect of the train speed and mileage errors, the corrected car body yaw angular velocity data is compared with the curve superelevation and the superelevation data detected by the dynamic inspection vehicle. The mileage of the three is basically the same, and the mileage positioning error is less than 20 m, as shown in Figure 9 Figure (c). It can be seen that the train speed and mileage accuracy corrected based on the DDTW algorithm meet the on-site application requirements.

[0091] The embodiment of the present invention also provides a train speed and mileage estimation device based on the inspection data of the ride-along instrument, as described in the following embodiments. Since the principle of this device to solve problems is similar to the train speed and mileage estimation method based on the inspection data of the ride-along instrument, the implementation of this device can refer to the implementation of the train speed and mileage estimation method based on the inspection data of the ride-along instrument, and the repeated parts will not be elaborated.

[0092] Figure 10 Figure for the train speed and mileage estimation device based on the inspection data of the ride-along instrument provided in the embodiment of the present invention, as shown in Figure 10 shown, the device includes:

[0093] A data acquisition module 1001 for acquiring the longitudinal acceleration of the train body, the yaw angular velocity of the car body, and the curve superelevation data collected by the ride-along instrument;

[0094] An integral calculation module 1002 for integrating the longitudinal acceleration of the car body to determine the speed estimate and mileage estimate of the train;

[0095] The deviation calculation module 1003 is configured to extract the trend term of the car body yaw angular velocity, take the curve superelevation data as a reference benchmark, and use the derivative dynamic time warping algorithm to calculate the optimal mapping path between the trend term and the reference benchmark, and determine the mileage deviation amount corresponding to the car body yaw angular velocity under the optimal mapping path;

[0096] The correction module 1004 is configured to correct the speed estimate and the mileage estimate according to the mileage deviation amount, and determine the corrected speed estimate as the speed information of the train, and determine the corrected mileage estimate as the mileage information of the train.

[0097] In one embodiment, the data acquisition module 1001 is specifically configured to:

[0098] Obtain the longitudinal acceleration of the car body, the yaw angular velocity of the car body, and the curve superelevation data of the train synchronously collected by the ride meter in the time dimension; the timestamps of the longitudinal acceleration of the car body, the yaw angular velocity of the car body, and the curve superelevation data are consistent.

[0099] In one embodiment, the integral calculation module 1002 is specifically configured to:

[0100] Perform a first integral calculation on the longitudinal acceleration of the car body to determine the speed estimate of the train;

[0101] Perform a second integral calculation on the longitudinal acceleration of the car body to determine the mileage estimate of the train.

[0102] In one embodiment, the deviation calculation module 1003 is specifically configured to:

[0103] Perform a moving average filter on the yaw angular velocity of the car body to obtain the trend term of the yaw angular velocity of the car body.

[0104] In one embodiment, the deviation calculation module 1003 is further configured to:

[0105] Perform normalization processing on the trend term and the curve superelevation data.

[0106] In one embodiment, the correction module 1004 is specifically configured to:

[0107] Correct the speed estimate and the mileage estimate according to the mileage deviation amount;

[0108] When the mileage deviation of the corrected mileage estimate meets the mileage deviation limit value, determine the corrected speed estimate as the speed information of the train, and determine the corrected mileage estimate as the mileage information of the train; the mileage deviation limit value is preset based on production requirements.

[0109] Based on the foregoing inventive concept, as Figure 11As shown in the figure, the present invention also provides a computer device 1100, including a memory 1110, a processor 1120, and a computer program 1130 stored on the memory 1110 and executable on the processor 1120. When the processor 1120 executes the computer program 1130, the foregoing train speed and mileage estimation method based on the inspection data of the riding meter is implemented.

[0110] An embodiment of the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the foregoing train speed and mileage estimation method based on the inspection data of the riding meter is implemented.

[0111] An embodiment of the present invention also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the foregoing train speed and mileage estimation method based on the inspection data of the riding meter is implemented.

[0112] In summary, in the embodiments of the present invention, the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the riding meter are obtained; the longitudinal acceleration of the train body is integrated to calculate the speed estimate and mileage estimate of the train; the trend term of the yaw angular velocity of the train body is extracted, and with the curve superelevation data as the reference benchmark, the derivative dynamic time warping algorithm is used to calculate the best mapping path between the trend term and the reference benchmark, and the mileage deviation amount corresponding to the yaw angular velocity of the train body under the best mapping path is determined; the speed estimate and mileage estimate are corrected according to the mileage deviation amount, the corrected speed estimate is determined as the train speed information, and the corrected mileage estimate is determined as the train mileage information. In this way, accurate train speed information and mileage information that meet the on-site verification requirements are calculated with the help of the riding meter, greatly improving the convenience and intelligence of the riding meter detection. Considering the local "shape" information of the yaw angular velocity of the train body and the curve superelevation data, the alignment accuracy between the yaw angular velocity of the train body and the curve superelevation data is improved through the derivative dynamic time warping (DDTW) algorithm, and the mileage deviation amount of the yaw angular velocity data of the train body under the best mapping path is accurately calculated, so as to correct the train speed and mileage errors, accurately estimate the train speed information and mileage information, and improve the convenience of the riding meter detection.

[0113] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0114] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0115] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0117] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for estimating the speed and mileage of a train based on the detection data of a multiplying instrument, characterized in that, Including: Obtaining the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the riding instrument; Performing integral calculation on the longitudinal acceleration of the train body to determine the speed estimation value and the mileage estimation value of the train; Extracting the trend item of the yaw angular velocity of the train body, using the curve superelevation data as a reference benchmark, and calculating the optimal mapping path between the trend item and the reference benchmark by using the derivative dynamic time warping algorithm to determine the mileage deviation amount corresponding to the yaw angular velocity of the train body under the optimal mapping path; Correcting the speed estimation value and the mileage estimation value according to the mileage deviation amount, determining the corrected speed estimation value as the speed information of the train, and determining the corrected mileage estimation value as the mileage information of the train.

2. The method according to claim 1, characterized in that Obtaining the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the riding instrument, including: Obtaining the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data synchronously collected by the riding instrument in the time dimension; the timestamps of the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data are consistent.

3. The method according to claim 1, characterized in that, Performing integral calculation on the longitudinal acceleration of the train body to determine the speed estimation value and the mileage estimation value of the train, including: Performing a first integral calculation on the longitudinal acceleration of the train body to determine the speed estimation value of the train; Performing a second integral calculation on the longitudinal acceleration of the train body to determine the mileage estimation value of the train.

4. The method according to claim 1, characterized in that Extracting the trend item of the yaw angular velocity of the train body, including: Performing moving average filtering on the yaw angular velocity of the train body to obtain the trend item of the yaw angular velocity of the train body.

5. The method according to claim 4, characterized in that, Using the curve superelevation data as a reference benchmark and calculating the optimal mapping path between the trend item and the reference benchmark by using the derivative dynamic time warping algorithm, further including: Performing normalization processing on the trend item and the curve superelevation data.

6. The method according to claim 1, wherein Correcting the speed estimation value and the mileage estimation value according to the mileage deviation amount, determining the corrected speed estimation value as the speed information of the train, and determining the corrected mileage estimation value as the mileage information of the train, including: Correcting the speed estimation value and the mileage estimation value according to the mileage deviation amount; When the mileage deviation of the corrected mileage estimation value meets the mileage deviation limit value, determining the corrected speed estimation value as the speed information of the train, and determining the corrected mileage estimation value as the mileage information of the train; the mileage deviation limit value is preset based on production requirements.

7. A train speed and mileage estimation device based on the detection data of a multiplying instrument, characterized in that, Including: A data acquisition module for obtaining the longitudinal acceleration of the train body, the yaw angular velocity of the train body, and the curve superelevation data collected by the riding instrument; An integral calculation module for performing integral calculation on the longitudinal acceleration of the train body to determine the speed estimation value and the mileage estimation value of the train; A deviation calculation module for extracting the trend item of the yaw angular velocity of the train body, using the curve superelevation data as a reference benchmark, and calculating the optimal mapping path between the trend item and the reference benchmark by using the derivative dynamic time warping algorithm to determine the mileage deviation amount corresponding to the yaw angular velocity of the train body under the optimal mapping path; A correction module for correcting the speed estimation value and the mileage estimation value according to the mileage deviation amount, determining the corrected speed estimation value as the speed information of the train, and determining the corrected mileage estimation value as the mileage information of the train.

8. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the method according to any one of claims 1 to 6.

Images