CN115130728A - Method and device for predicting train wheel diameter calibration time - Google Patents
Method and device for predicting train wheel diameter calibration time Download PDFInfo
- Publication number
- CN115130728A CN115130728A CN202210616156.7A CN202210616156A CN115130728A CN 115130728 A CN115130728 A CN 115130728A CN 202210616156 A CN202210616156 A CN 202210616156A CN 115130728 A CN115130728 A CN 115130728A
- Authority
- CN
- China
- Prior art keywords
- wheel diameter
- time
- train
- value
- sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
- B61K9/00—Railway vehicle profile gauges; Detecting or indicating overheating of components; Apparatus on locomotives or cars to indicate bad track sections; General design of track recording vehicles
- B61K9/12—Measuring or surveying wheel-rims
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Operations Research (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Mechanical Engineering (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention provides a method and a device for predicting train wheel diameter calibration time, wherein the method comprises the following steps: inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model; determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence; the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label. The vehicle diameter calibration time determined by the method provides reference for the operator to recalibrate the wheel diameter in advance, and recalibration is performed in time in a targeted manner, so that the train control accuracy is improved, and the train maintenance workload is reduced.
Description
Technical Field
The invention relates to the technical field of rail transit, in particular to a method and a device for predicting train wheel diameter calibration time.
Background
In the urban rail transit signal system, for newly built vehicles, the wheel diameter value is uniformly 840mm, and the signal system uses the set wheel diameter value to participate in the calculation of the train running distance and the train speed.
When the train runs on the track for a long time, the wheels are worn or scratched, and the diameter of the train wheel is changed. However, the signal system still uses the configured wheel diameter value, and the wheel diameter value for calculation and the actual wheel diameter value have larger deviation, thereby influencing the positioning calculation of the signal system on the train. For example, the train arrives at a station and stops at a station, and the train door and the shield door are not aligned, so that passengers can get on or off the train. And the accurate control of the speed of the train by a signal system can be influenced, and the traction brake control index is influenced. Therefore, after the train is used for a period of time, the operator needs to measure the wheel diameter of the train again and recalibrate the wheel diameter value.
At present, according to whether a train runs or not, static detection and dynamic detection are mainly used for measuring the wheel diameter value of the train. The static detection is that under the non-running state of the train, subway technicians realize the detection work of the wheel diameter according to the operation rules. The dynamic detection is that the wheel diameter value of the current train is calculated according to the fixed transponder interval and the pulse number output by the speed sensor through two accurate positioning transponders when the train is in the running state.
The dynamic detection method is high in detection speed and automation degree, does not occupy the train turnover time, but is complex in communication structure of the calibration system, poor in signal transmission real-time performance and limited in calculation precision. Although the static detection method has high measurement precision, the real state of the wheel in operation cannot be known accurately in time, and the situation that the wheel is still continuously used due to serious abrasion cannot be avoided. If there are no turns caused by occasional scratches, the track operation will generally turn all the wheels once and measure the wheel diameter values again after a long period of time or when the vehicle is running for more than several kilometers.
Disclosure of Invention
The invention provides a method and a device for predicting train wheel diameter calibration time, which are used for solving the defect that a static detection method in the prior art cannot calibrate the train wheel diameter in time and realizing the real-time prediction of the train wheel diameter calibration time.
The invention provides a method for predicting train wheel diameter calibration time, which comprises the following steps:
inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model;
determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence;
the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
According to the train wheel diameter calibration time prediction method provided by the invention, before the step of inputting a first wheel diameter time series of a train before the current time into an ARIMA model and obtaining a second wheel diameter time series of the train after the current time, which is output by the ARIMA model, the method further comprises the following steps:
acquiring a difference value between accumulated pulse numbers output by a speed sensor on the train when the train passes through a first transponder and a second transponder each time before the current moment;
acquiring an actual installation distance between the first transponder and the second transponder;
acquiring a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance;
constructing the first wheel diameter time series from the first wheel diameter values.
According to the train wheel diameter calibration time prediction method provided by the invention, the step of obtaining the first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance comprises the following steps:
acquiring a difference value between the accumulated travelling distances of the train when the train passes through the first transponder and the second transponder every time;
and under the condition that the deviation between the difference between the actual installation distance and the accumulated walking distance is smaller than a second preset threshold value, acquiring a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance.
According to the train wheel diameter calibration time prediction method provided by the invention, a first wheel diameter value of the train is calculated according to the difference between the accumulated pulse numbers and the actual installation distance by the following formula:
wherein, S is an actual installation distance between the first transponder and the second transponder, Δ N is a difference between the accumulated pulse numbers, D is a first wheel diameter value of the train, N is a pulse number output by a speed sensor when a wheel of the train rotates for one circle, and α is a preset coefficient.
According to the train wheel diameter calibration time prediction method provided by the invention, before the step of inputting a first wheel diameter time series of a train before the current time into an ARIMA model and obtaining a second wheel diameter time series of the train after the current time, which is output by the ARIMA model, the method further comprises the following steps:
under the condition that the sample wheel diameter time sequence is detected to be unstable, carrying out differential processing on the sample wheel diameter time sequence until the sample wheel diameter time sequence is stable;
taking the number of the differential processing as a differential number value in the ARIMA model;
determining an autoregressive parameter value and a moving average parameter value in the ARIMA model based on an information criterion function method;
inputting the time series of sample wheel diameters into the ARIMA model, and calculating a residual sequence between a subsequent wheel diameter predicted value sequence and a subsequent wheel diameter observed value sequence output by the ARIMA model;
and under the condition that the residual error sequence is not detected to be white noise, adjusting an autoregressive parameter value and a moving average parameter value in the ARIMA model to enable the residual error sequence to be white noise.
According to the train wheel diameter calibration time prediction method provided by the invention, the time for performing wheel diameter calibration on the train is determined according to the time corresponding to the second wheel diameter value which is smaller than the first preset threshold value in the second wheel diameter time sequence, and the method comprises the following steps:
and determining the time for calibrating the wheel diameter of the train according to the earliest time in the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence.
The invention also provides a device for predicting the train wheel diameter calibration time, which comprises:
inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model;
determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence;
the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
The present invention also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor executes the program to implement the train wheel diameter calibration time prediction method as described in any one of the above.
The present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the train wheel diameter alignment time prediction method as described in any one of the above.
The present invention also provides a computer program product comprising a computer program which, when executed by a processor, implements a train wheel diameter calibration time prediction method as described in any one of the above.
According to the method and the device for predicting the train wheel diameter calibration time, the historical train wheel diameter time sequence of the train is input into the ARIMA model, the future development trend of the train wheel diameter value is predicted, the time when the train wheel diameter value reaches the ATP tolerance threshold value is determined according to the future development trend, the time determined for different situations of different trains is different, reference is provided for operators to recalibrate the wheel diameter in advance, timely recalibration is performed in a targeted manner, the train control accuracy is improved, and the train maintenance workload is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for predicting train wheel diameter alignment time according to the present invention;
FIG. 2 is a schematic diagram of a data transmission flow of a vehicle-mounted platform in the method for predicting train wheel diameter calibration time according to the present invention;
FIG. 3 is a schematic diagram of a process for constructing an ARIMA model in the method for predicting train wheel diameter calibration time provided by the invention;
FIG. 4 is a schematic structural diagram of a train wheel diameter calibration time prediction device provided by the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The train wheel diameter calibration time prediction method of the present invention is described below with reference to fig. 1, and includes: step 101, inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model;
the embodiment uses an ARIMA model to predict the change trend of train wheel diameter values of a time series.
The autoregressive model ar (p) predicts itself by using the historical time data of the variable itself. In particular to a method for determining the time sequence of the digital television by utilizing the correlation between the early-stage numerical value and the later-stage numerical value of the time sequence,establishing an independent variable regression equation containing values of early and late stages, i.e. t time series value x t From x 1 ，x 2 ，...，x t-1 Establishing equation acquisition, wherein the equation is as follows:
x t ＝α 1 x t-1 +α 2 x t-2 +...+α p x t-p +u t
a k denotes the k-th time-series autocorrelation coefficient, u k For the k-th time series to be white noise, p is the autoregressive order, and the time series data must have stationarity.
The moving average model ma (q) focuses on the accumulation of error terms in the autoregressive model, which is a weighted sum of white noise over time t, subject to a moving average equation of order q:
x t ＝β 1 u t-1 +β 2 u t-2 +...+β q u t-q +u t
β k represents the k-th time-series moving average coefficient, u k The moving average method can effectively eliminate random fluctuation in prediction for the white noise of the kth time series.
Combining the AR (p) model and the MA (q) model to obtain an autoregressive moving average model ARMA (p, q). The ARMA model integrates the advantages of the AR model and the MA model to form a more powerful model, and the formula is as follows:
x t ＝α 1 x t-1 +α 2 x t-2 +...+α p x t-p +u t +β 1 u t-1 +β 2 u t-2 +...+β q u t-q
the ARMA model is used for analyzing a stationary time sequence, and the diameter value of the wheel of the actual train is influenced by various factors, is not stationary, shows the superposition of period and trend changes, and needs to carry out stabilization processing on an original observation sequence. ARIMA (p, q, d) can be used to process non-stationary time series by means of difference, d being the number of differences. Compared with an ARMA model, the ARIMA model is characterized in that unstable data are subjected to d-time difference to form stable time sequence data, and then the ARMA model is adopted for prediction.
Assuming that the current time is the (n + 1) th time, the first wheel diameter time series { D) before the current time 1 ，…，D n }。D 1 A first wheel diameter value, D, for a first moment of the train n The first wheel diameter value of the train at the nth time.
The first wheel diameter value is a historical wheel diameter value of the train, and may be obtained by a static method or a dynamic method, and this embodiment is not particularly limited.
Predicting the later development trend of the diameter value of the first wheel by using the constructed ARIMA model to obtain a subsequent second wheel diameter time sequence { D } of the (n + 1) th moment n+1 ，…，D n+k }。
102, determining time for calibrating the wheel diameter of the train according to the time corresponding to a second wheel diameter value smaller than a first preset threshold value in the second wheel diameter time sequence;
each second wheel diameter value D in the second wheel diameter time series n+1 ，…，D n+k And comparing the diameter value with a first preset threshold value, and selecting a second wheel diameter value smaller than the first preset threshold value. The diameter value of the second wheel is smaller than the first preset threshold value, which shows that the wheel abrasion degree of the train is large, and the positioning calculation and the speed control of the signal system to the train are influenced.
Alternatively, in a case where the second wheel diameter value is smaller than the first preset threshold value a plurality of consecutive times, a certain time before the earliest time among times corresponding to the first occurrence of the second wheel diameter value that is smaller than the first preset threshold value a plurality of consecutive times is taken as the latest time for the train to perform the wheel diameter calibration. And judging how long the train still needs the larger wheel wear degree according to the wheel diameter calibration time, and calibrating.
And warning or displaying the wheel diameter calibration before the time of the wheel diameter calibration, and reminding a train operator of recalibrating the vehicle diameter in time so as to achieve the purpose of prediction.
The ARIMA model is obtained by training with a sample wheel diameter time sequence as a sample and a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
Let sample wheel diameter time series be { D' 1 ，…，D' n And then the time series of sample wheel diameters is input into the ARIMA model to obtain a sequence of subsequent wheel diameter predicted values of { D' n+1 ，…，D' n+k And the observed value sequence of the subsequent wheel diameters corresponding to the sample wheel diameter time sequence is { D " n+1 ，…，D” n+k }。
Use { D' 1 ，…，D' n And { D " n+1 ，…，D” n+k And (5) training the ARIMA model to obtain parameters after the optimization of the ARIMA model.
According to the method, the historical wheel diameter time sequence of the train is input into the ARIMA model, the future development trend of the train wheel diameter value is predicted, the time when the train wheel diameter value reaches the ATP tolerance threshold value is determined according to the future development trend, the time determined for different conditions of different trains is different, reference is provided for operators to recalibrate the wheel diameter in advance, timely recalibration is conducted in a targeted mode, the train control accuracy is improved, and the train maintenance workload is reduced.
On the basis of the foregoing embodiment, in this embodiment, before the step of inputting the first wheel diameter time series of the train before the current time into the ARIMA model and obtaining the second wheel diameter time series of the train after the current time output by the ARIMA model, the method further includes: acquiring a difference value between accumulated pulse numbers output by a speed sensor on the train when the train passes through a first transponder and a second transponder each time before the current moment;
in a track line, the installation error of the accurate positioning responder device is not more than 2cm, and the installation error of the non-accurate positioning responder device is not more than 1 m. Therefore, data of the train passing through the non-accurate positioning transponder is not suitable for calculating the diameter value of the wheel. Optionally, the first transponder a and the second transponder B are precision positioning transponders.
The first responder A and the second responder B are arranged beside a track where a train runs, and the train passes through the first responder A and the second responder B successively. When the train passes through the first responder A, receiving a message sent by the first responder A; the embodiment of receiving a message sent by the second transponder B when passing the second transponder B is not limited to the content of the sent message.
The speed sensor is used for measuring the running speed of the train and acquiring the accumulated pulse number N output by the speed sensor when the train successively receives the messages sent by the first responder A and the second responder B 1 And N 2 . Calculating the pulse variation Δ N ═ N 2 -N 1 。
Acquiring an actual installation distance between the first transponder and the second transponder;
and acquiring the actual installation distance S between the first transponder A and the second transponder B from the electronic map, wherein the distance is also the actual running distance of the train between the first transponder A and the second transponder B. May be obtained from the distance of the first transponder and the second transponder in the electronic map.
Acquiring a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance;
a first wheel diameter value is calculated based on Δ N and S. For example, the train running speed is calculated from Δ N and S, and the first wheel diameter value of the train is calculated from the speed. And storing the first wheel diameter value and the output time of the accumulated pulse number corresponding to the first wheel diameter value into a database.
Constructing the first wheel diameter time series from the first wheel diameter values.
And sequencing the first wheel diameter values calculated when the train passes through the two sensors for multiple times according to the sequence of the output moments of the accumulated pulse numbers corresponding to the first wheel diameter values to obtain a first wheel diameter time sequence.
Optionally, after the vehicle-mounted device recording apparatus calculates the first wheel diameter time series, the vehicle-mounted device recording apparatus sends the first wheel diameter time series of the vehicle-mounted platform to a data server on the ground by using a vehicle-ground wireless transmission network, and then the data server forwards the first wheel diameter time series to the operation and maintenance subsystem, and stores the first wheel diameter time series in the database.
The vehicle-mounted equipment recording device can also directly send the difference between the accumulated pulse numbers and the actual installation distance to a data server on the ground, and then the data server forwards the difference and the actual installation distance to the operation and maintenance subsystem. And calculating the first wheel diameter time sequence by the operation and maintenance subsystem according to the received data. The data transmission flow of the vehicle-mounted platform is shown in fig. 2.
On the basis of the foregoing embodiment, in this embodiment, the calculating a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance includes: acquiring a difference value between the accumulated travelling distances of the train when the train passes through the first transponder and the second transponder every time;
when the train passes through the first transponder A, namely when a message sent by the first transponder A is received, the accumulated running distance S of the train calculated by the ATP 1 . Then, when the train passes through the second transponder B, namely when the message sent by the second transponder B is received, the accumulated running distance S of the train calculated by the ATP 2 Calculating the traveling distance delta S between the first transponder A and the second transponder B as S 2 -S 1 。
And under the condition that the deviation between the difference between the actual installation distance and the accumulated walking distance is smaller than a second preset threshold value, acquiring a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance.
And if the difference value between the delta S and the S is smaller than a second preset threshold value, the first responder A and the second responder B are consistent with the configuration data position of the electronic map and are installed at the appointed position according to the signal system engineering data map.
And if the difference value between the delta S and the S is larger than or equal to a second preset threshold value, carrying out alarm prompt on operators.
On the basis of the above embodiment, the present embodiment obtains the first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance by the following formula:
wherein, S is an actual installation distance between the first transponder and the second transponder, Δ N is a difference between the accumulated pulse numbers, D is a first wheel diameter value of the train, N is a pulse number output by a speed sensor when a wheel of the train rotates for one circle, and α is a preset coefficient.
Alternatively, α is a coefficient converted into length units, and if it is millimeters, it is 10.
In addition to the above embodiments, as shown in fig. 3, before the step of inputting a first wheel diameter time series of a train before a current time into an ARIMA model and obtaining a second wheel diameter time series of the train after the current time output by the ARIMA model, the present embodiment further includes: under the condition that the sample wheel diameter time sequence is detected to be unstable, carrying out differential processing on the sample wheel diameter time sequence until the sample wheel diameter time sequence is stable; taking the number of differential processing as a differential number value in the ARIMA model;
in modeling the ARIMA model, a sample wheel diameter time series is first introduced. Obtaining a time sequence { D) of sample wheel diameters calculated each time a train with the number N passes through the first transponder A and the second transponder B from a database 1 ，…，D n }。
Optionally, the sample wheel diameter time series { D }is first aligned 1 ，…，D n Preprocessing is carried out, and an amplitude limiting filtering method is adopted. If the calculated first wheel diameter value D is not equal to the first wheel diameter value n And the last calculated first wheel diameter value D n-1 Deviation of (2) exceeds a certain threshold value, D n And (4) invalidation. The impulse interference caused by accidental factors can be eliminated through the amplitude limiting filtering method.
Then, the stability of the sample wheel diameter time sequence is judged, and the stability of the sample wheel diameter time sequence is judged by using an ADF (extended dicky filler) unit root test method. If the sample wheel diameter time series is stable, no unit root exists; and if the time sequence is not stable, the unit root exists, and the time sequence of the sample wheel diameter is subjected to differential processing until the time sequence of the sample wheel diameter after differential processing is stable. The difference times d value of the ARIMA model can be determined through difference processing.
Determining an autoregressive parameter value and a moving average parameter value in the ARIMA model based on an information criterion function method;
then, model scaling is carried out, and an autoregressive parameter value p and a moving average parameter value q in the ARIMA model are determined. The order of the model is determined using an Information Criterion function, optionally using a BIC (Bayesian Information Criterion) or AIC (Akaike Information Criterion) Criterion. The BCI criterion is calculated according to the following formula:
BIC＝kln(n)-2ln(L)；
wherein k is the number of parameters in the model, n is the sample capacity, and L is the likelihood function value. The BIC criterion considers the sample capacity relative to the AIC criterion, and when the sample capacity is large, the phenomenon of overfitting caused by overhigh model complexity due to overhigh model progress can be effectively prevented.
And (3) obtaining all BIC information quantities within a certain value range of p and q (the approximate range of the p and q values can be determined by applying an autocorrelation function ACF and a partial autocorrelation function PACF and performing model order by tailing and truncation), and then taking the values of the model orders p and q in which the BIC information quantities reach the minimum.
Inputting the time series of sample wheel diameters into the ARIMA model, and calculating a residual sequence between a subsequent wheel diameter predicted value sequence and a subsequent wheel diameter observed value sequence output by the ARIMA model;
and obtaining a residual sequence according to the difference between the observed value of the variable and the predicted value obtained by the estimated regression equation.
And under the condition that the residual error sequence is not detected to be white noise, adjusting an autoregressive parameter value and a moving average parameter value in the ARIMA model to enable the residual error sequence to be white noise.
The reasonability of the model is determined by checking the autocorrelation of the residual sequence, and the randomness of the residual sequence can be checked by an autocorrelation function method. Alternatively, it is determined whether the residual sequence is white noise using an Ljung-Box test, i.e., a white noise test. If the residual error sequence is white noise, the useful information in the residual error sequence is extracted, the rest information is random fluctuation and cannot be predicted and used, no information can be extracted continuously, and therefore the model can be determined. If the residual sequence is not white noise, useful information still exists in the residual, and the ARIMA model parameters need to be further adjusted to extract the useful information.
On the basis of the foregoing embodiments, in this embodiment, the determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold in the second wheel diameter time series includes: and determining the time for calibrating the wheel diameter of the train according to the earliest time in the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence.
For example, 2 hours before the earliest time is set as the latest time for performing wheel diameter calibration on the train.
The following describes a train wheel diameter calibration time prediction apparatus provided by the present invention, and the train wheel diameter calibration time prediction apparatus described below and the train wheel diameter calibration time prediction method described above may be referred to in correspondence with each other.
As shown in fig. 4, the apparatus comprises a prediction module 401 and a determination module 402, wherein:
the prediction module 401 is configured to input a first wheel diameter time series of a train before a current time into an ARIMA model, and obtain a second wheel diameter time series of the train after the current time, which is output by the ARIMA model;
the determining module 402 is configured to determine a time for performing wheel diameter calibration on the train according to a time corresponding to a second wheel diameter value smaller than a first preset threshold in the second wheel diameter time sequence;
the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
According to the method, the historical wheel diameter time sequence of the train is input into the ARIMA model, the future development trend of the train wheel diameter value is predicted, the time when the train wheel diameter value reaches the ATP tolerance threshold value is determined according to the future development trend, the time when the train wheel diameter value reaches the ATP tolerance threshold value is determined to be different for different conditions of different trains, reference is provided for operators to recalibrate the wheel diameter in advance, timely recalibration is conducted in a targeted mode, the train control accuracy is improved, and the train maintenance workload is reduced.
Fig. 5 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 5: a processor (processor)510, a communication Interface (Communications Interface)520, a memory (memory)530, and a communication bus 540, wherein the processor 510, the communication Interface 520, and the memory 530 communicate with each other via the communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a train wheel diameter calibration time prediction method comprising: inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model; determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than a first preset threshold value in the second wheel diameter time sequence; the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
Furthermore, the logic instructions in the memory 530 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing the train wheel diameter calibration time prediction method provided by the above methods, the method comprising: inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model; determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence; the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a train wheel diameter calibration time prediction method provided by the above methods, the method comprising: inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model; determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence; the ARIMA model is obtained by training with a sample wheel diameter time sequence as a sample and a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment may be implemented by software plus a necessary general hardware platform, and may also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A train wheel diameter calibration time prediction method is characterized by comprising the following steps:
inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model, and obtaining a second wheel diameter time sequence of the train after the current time, which is output by the ARIMA model;
determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence;
the ARIMA model is obtained by training by taking the sample wheel diameter time sequence as a sample and taking a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
2. The train wheel diameter calibration time prediction method of claim 1, wherein prior to the step of inputting a first time series of wheel diameters of a train before a current time into an ARIMA model and obtaining a second time series of wheel diameters of the train after the current time output by the ARIMA model, further comprising:
acquiring a difference value between accumulated pulse numbers output by a speed sensor on the train when the train passes through a first transponder and a second transponder before the current moment each time;
acquiring an actual installation distance between the first transponder and the second transponder;
acquiring a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance;
constructing the first wheel diameter time series from the first wheel diameter values.
3. The train wheel diameter calibration time prediction method of claim 2, wherein the obtaining a first wheel diameter value of the train based on the difference between the accumulated number of pulses and the actual installation distance comprises:
acquiring a difference value between the accumulated travelling distances of the train when the train passes through the first transponder and the second transponder every time;
and under the condition that the deviation between the difference between the actual installation distance and the accumulated walking distance is smaller than a second preset threshold value, acquiring a first wheel diameter value of the train according to the difference between the accumulated pulse numbers and the actual installation distance.
4. The train wheel diameter calibration time prediction method according to claim 2, wherein the first wheel diameter value of the train is obtained from the difference between the accumulated pulse numbers and the actual installation distance by the following formula:
wherein, S is an actual installation distance between the first transponder and the second transponder, Δ N is a difference between the accumulated pulse numbers, D is a first wheel diameter value of the train, N is a pulse number output by a speed sensor when a wheel of the train rotates for one circle, and α is a preset coefficient.
5. The train wheel diameter calibration time prediction method of any one of claims 1 to 4, wherein prior to the step of inputting a first time series of wheel diameters of the train before the current time into the ARIMA model and obtaining a second time series of wheel diameters of the train after the current time output by the ARIMA model, further comprising:
under the condition that the sample wheel diameter time sequence is detected to be unstable, carrying out differential processing on the sample wheel diameter time sequence until the sample wheel diameter time sequence is stable;
taking the number of the differential processing as a differential number value in the ARIMA model;
determining an autoregressive parameter value and a moving average parameter value in the ARIMA model based on an information criterion function method;
inputting the sample wheel diameter time sequence into the ARIMA model, and acquiring a residual sequence between a subsequent wheel diameter predicted value sequence and a subsequent wheel diameter observed value sequence output by the ARIMA model;
and under the condition that the residual error sequence is not detected to be white noise, adjusting an autoregressive parameter value and a moving average parameter value in the ARIMA model to enable the residual error sequence to be white noise.
6. The method for predicting train wheel diameter calibration time according to any one of claims 1 to 4, wherein the determining the time for performing wheel diameter calibration on the train according to the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time series includes:
and determining the time for calibrating the wheel diameter of the train according to the earliest time in the time corresponding to the second wheel diameter value smaller than the first preset threshold value in the second wheel diameter time sequence.
7. A train wheel diameter calibration time prediction apparatus, comprising:
the prediction module is used for inputting a first wheel diameter time sequence of a train before the current time into an ARIMA model to obtain a second wheel diameter time sequence of the train after the current time, wherein the second wheel diameter time sequence is output by the ARIMA model;
the determining module is used for determining the time for calibrating the wheel diameter of the train according to the time corresponding to the second wheel diameter value smaller than a first preset threshold value in the second wheel diameter time sequence;
the ARIMA model is obtained by training with a sample wheel diameter time sequence as a sample and a subsequent wheel diameter observation value sequence corresponding to the sample wheel diameter time sequence as a label.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the train wheel diameter calibration time prediction method of any one of claims 1 to 6.
9. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the train wheel diameter calibration time prediction method according to any one of claims 1 to 6.
10. A computer program product comprising a computer program, wherein the computer program when executed by a processor implements a train wheel diameter calibration time prediction method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210616156.7A CN115130728A (en) | 2022-05-31 | 2022-05-31 | Method and device for predicting train wheel diameter calibration time |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210616156.7A CN115130728A (en) | 2022-05-31 | 2022-05-31 | Method and device for predicting train wheel diameter calibration time |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115130728A true CN115130728A (en) | 2022-09-30 |
Family
ID=83378269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210616156.7A Pending CN115130728A (en) | 2022-05-31 | 2022-05-31 | Method and device for predicting train wheel diameter calibration time |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115130728A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117302293A (en) * | 2023-08-14 | 2023-12-29 | 北京城建智控科技股份有限公司 | Wheel diameter value prediction method and device, electronic equipment and storage medium |
-
2022
- 2022-05-31 CN CN202210616156.7A patent/CN115130728A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117302293A (en) * | 2023-08-14 | 2023-12-29 | 北京城建智控科技股份有限公司 | Wheel diameter value prediction method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Real-time defect detection of high-speed train wheels by using Bayesian forecasting and dynamic model | |
| CN109782325B (en) | Train speed estimation method based on particle filtering and multi-sensor information fusion | |
| EP3783356A1 (en) | Train component crack damage monitoring method and system | |
| CN110533229B (en) | Track maintenance time prediction method and device | |
| JP6313929B2 (en) | Method and system for monitoring structures | |
| CN109000940B (en) | Abnormal axle temperature diagnosis method and system for rolling stock | |
| CN110361180B (en) | Intelligent train pantograph service performance dynamic monitoring and evaluating method and system | |
| CN114996654B (en) | Road surface flatness detection method and device, electronic equipment and medium | |
| CN110533239B (en) | Smart city air quality high-precision measurement method | |
| JP2019112049A (en) | Method for recognizing driving style of driver of land vehicle and corresponding device | |
| CN113247041B (en) | Wheel set axle box device state monitoring method based on digital twin model | |
| Lin-Hai et al. | Fault diagnosis for track circuit using AOK-TFRs and AGA | |
| CN115130728A (en) | Method and device for predicting train wheel diameter calibration time | |
| CN115050214B (en) | AIS data-based ship collision risk prediction method | |
| CN114093171B (en) | Traffic running state monitoring method and device based on multi-source data fusion | |
| Muñoz et al. | Estimation of lateral track irregularity using a Kalman filter. Experimental validation | |
| Cunillera et al. | Real-time train motion parameter estimation using an unscented Kalman filter | |
| CN115758289A (en) | Rail corrugation identification method based on multitask learning neural network | |
| Zheng et al. | A novel prediction model for car body vibration acceleration based on correlation analysis and neural networks | |
| CN112348039B (en) | Training method of driving behavior analysis model, driving behavior analysis method and equipment | |
| CN112722003B (en) | Method and equipment for monitoring train derailment risk | |
| CN104457644A (en) | Detecting method and device for non-pulse abnormal data in track geometry inspection data | |
| CN116186877A (en) | Method and device for detecting characteristics of a vehicle | |
| CN115829100A (en) | Train parking precision prediction method and device | |
| JP2000322682A (en) | System, device and method for predicting traffic flow |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |