CN114117687A - Method and system for building and predicting life prediction model of key parts of wheel set - Google Patents
Method and system for building and predicting life prediction model of key parts of wheel set Download PDFInfo
- Publication number
- CN114117687A CN114117687A CN202111489662.6A CN202111489662A CN114117687A CN 114117687 A CN114117687 A CN 114117687A CN 202111489662 A CN202111489662 A CN 202111489662A CN 114117687 A CN114117687 A CN 114117687A
- Authority
- CN
- China
- Prior art keywords
- wheel
- turning
- wheel diameter
- mileage
- data
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- 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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Vehicle Cleaning, Maintenance, Repair, Refitting, And Outriggers (AREA)
Abstract
The invention provides a method and a system for building and predicting a life prediction model of a key part of a wheel set, which comprises the following steps: determining the input parameters of the model as the environment temperature, the running speed and the running mileage, and the output parameters as the residual abrasion loss of the wheel diameter; acquiring the current environmental temperature, the driving speed and the driving mileage of the wheel pair in real time, and acquiring turning data of the wheel pair as sample data; preprocessing the sample data; and respectively constructing models according to the preprocessed sample data, and finally selecting the model with the highest verification sample accuracy as the optimal prediction model of the current wheel pair. And the residual service life of the wheel is estimated and predicted by combining the wheel geometric part abrasion data and the development trend with the mileage evaluation, and the prediction precision is high.
Description
Technical Field
The invention belongs to the technical field of prediction, and particularly relates to a method and a system for building and predicting a service life prediction model of a key part of a wheelset.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The wheel set is one of key parts of the motor train unit and plays an important role in the actual operation of the motor train unit. The wheel set plays a crucial role in the running of the motor train unit, the failure of the wheel set not only endangers the driving safety of the train, but also brings huge impact and damage to the railway lines and bridges, aggravates the fatigue damage degree of the vehicle structure and increases the maintenance cost. The traditional mode for checking the wheel set faults is static checking, namely, when a vehicle is maintained regularly, the wheel set is disassembled and then measured by a mechanical special ruler; and secondly, dynamic inspection, namely, a train inspector performs manual operation on a vehicle in use by touching the vehicle with a hammer and striking the vehicle with hands, so that the failure of the wheel set tread in use cannot be found in time, and missing inspection and false inspection sometimes occur. Therefore, it is important to detect the wheel pair state.
Both the wheel rim and tread are lost under normal wheel-rail matching and wheel wear conditions. Particularly, the abrasion of the wheel rim is gradually reduced along with the increase of the operation time, and the wheel rim becomes thinner, so that the running safety is threatened. Therefore, turning work is required to restore the thickness of the rim. Wheel turnarounds are a repair that restores rim thickness at the expense of a partial tread diameter. The reduction of the thickness of the wheel flange and the diameter of the tread caused by abrasion and the reduction of the diameter of the tread caused by turning are key factors influencing the service life of the wheel, and the abrasion and the turning influence on the tread of the wheel of the motor train unit are shown in a schematic diagram as shown in a figure 1 below.
The remaining life of the wheel set is determined primarily by the amount of its turn-turning per turn and the mileage of the vehicle, among other parameters. The remaining life is generally estimated mainly by the wheel diameter value to the limit date. The wheel diameter to limit is determined by the last time of turning and the remaining operation time, and the determination of the remaining operation time is key and is determined by the remaining operation kilometers, and then the remaining operation kilometers are determined by emphasis, which is mainly determined by the remaining wheel diameter wear amount after the last turning of the shaft and the average wear rate per ten thousand wheel diameters at the last turning. The more the abrasion loss is, the shorter and shorter the remaining service life of the wheel set is, so that the study on the evaluation of the remaining service life of the wheel set, namely the study on the abrasion loss is required to be focused.
Through analyzing the abrasion conditions of all parts of the wheel, through analyzing relevant factors such as vehicle load, a straight line curve, different lines and the like, the abrasion degree of the wheel in the whole day can be reflected by the abrasion levels of the four parts, namely the tread abrasion amount, the wheel rim abrasion amount, the wheel inner diameter difference value and the wheel hub difference value of the wheel, and a prediction model is established. And then, evaluating and predicting the residual life of the wheel by combining the mileage of the wheel through the wear data and the development trend of several parts of the wheel. Currently commonly used prediction algorithms include: a model-based method, a data-driven method, a fusion-type prediction method, and the like, as shown in fig. 2.
The distribution rule needs to be determined by a model-based method, so that the method is difficult to realize in actual rail transit operation, and the model needs to be a failure physical model which is a mechanism model generally obtained in an experimental environment, and is greatly influenced by various factors in actual environment operation, so that the final model prediction effect is greatly reduced.
From the perspective of mechanism research, a mechanism model needs an accurate mathematical formula principle to prove, and under the current technical conditions, some models cannot be established or the mechanism research does not break through, so that some faults occur in object monitoring.
At present, a health monitoring and expert support system (PHM) of the motor train unit is connected to a plurality of motor train units and is applied to a plurality of road departments in an open mode, more than one part of hidden trouble faults of the motor train unit operation are prevented every year in recent years, and important functions are played for daily operation monitoring and fault emergency disposal of the motor train unit.
In the process of popularization and application of the PHM system, a user puts higher requirements on the predictive maintenance of the operation endpoint of the PHM system in the actual application process, and needs to develop research on the predictive maintenance of the PHM system on the application and maintenance of key parts, put forward a reasonable predictive maintenance suggestion and guide the specification of a reasonable economic maintenance plan and maintenance task.
Therefore, the technical problems solved by the invention include:
how to analyze a large amount of data related to the wheel set reduces the subjectivity of manual participation, and the wheel set service life prediction result is more objective and accurate.
Disclosure of Invention
The invention provides a method for constructing a wheel set key component service life prediction model to solve the problems.
According to some embodiments, the invention adopts the following technical scheme:
in a first aspect, a method for building a life prediction model of a key component of a wheel set is disclosed, which comprises the following steps:
determining the input parameters of the model as the environment temperature, the running speed and the running mileage, and the output parameters as the residual abrasion loss of the wheel diameter;
acquiring the current environmental temperature, the driving speed and the driving mileage of the wheel pair in real time, and acquiring turning data of the wheel pair as sample data;
preprocessing the sample data;
and respectively constructing models according to the preprocessed sample data, and finally selecting the model with the highest verification sample accuracy as the optimal prediction model of the current wheel pair.
According to a further technical scheme, the tuning data is stored in a database so as to be periodically transmitted back to a tuning detail table corresponding to the database from each service station offline.
According to the further technical scheme, the sample data is preprocessed at least by removing dirty data in the sample data, completing missing data and standardizing the data;
preferably, abnormal values with certain errors or overrun in the speed, temperature and mileage returning process and incomplete turning data uploading are removed;
preferably, the data are subjected to complete turning according to the transmission frequency of speed, temperature and mileage based on a difference algorithm;
preferably, the temperature, speed, mileage, and wheel diameter values are normalized so that the data is under a uniform metric.
In a second aspect, a method for predicting the life of a key component of a wheel set is disclosed, which comprises the following steps:
obtaining an optimal prediction model of the wheel set based on the method;
the running speed, the ambient temperature and the driving mileage of the wheel pair are obtained in real time and preprocessed, a corresponding prediction model is selected, the residual wheel diameter abrasion is output, the real-time prediction of the residual wheel diameter abrasion is realized, and the residual life of the wheel pair is estimated in real time based on the real-time predicted residual wheel diameter abrasion.
According to the further technical scheme, after an optimal prediction model of the wheel pair is obtained, the wheel diameter abrasion rate of each ten thousand kilometers is solved, then the number of remaining running kilometers and the number of remaining running days are calculated, and finally the wheel diameter value to the limit date is calculated.
According to a further technical scheme, when the wheel diameter abrasion rate of each ten thousand kilometers is solved, the wheel diameter abrasion rate of each ten thousand kilometers of the latest turning is equal to (the last turning rear wheel diameter value-the latest turning front wheel diameter value)/the operation mileage in the latest turning period;
the wheel diameter values refer to (the wheel diameter value of the left wheel + the wheel diameter value of the right wheel)/2 of the same wheel pair;
further preferably, the wear rates of all the turning times of the wheel are obtained, the maximum value and the minimum value of the wheel diameter wear rates of all the turning times are obtained, and finally the maximum, minimum and latest average values of the wear rates are obtained.
According to a further technical scheme, when the number of the remaining operating kilometers and the number of the remaining operating days are calculated, the remaining operating kilometers (average/maximum/minimum) is equal to the remaining wheel diameter wear amount after the last turning of the wheel shaft/(average wear rate/maximum wear rate/minimum wear rate per ten thousand kilometers of wheel diameter during the last turning);
further preferably, the remaining wear loss of the wheel diameter after the latest turning of the wheel shaft is the wheel diameter-wheel inner diameter after the latest turning;
average daily operating mileage in the latest turning period is (accumulated traveling mileage of the latest turning vehicle-accumulated traveling mileage of the last turning vehicle)/latest turning date-latest turning date;
further preferably, the remaining operation time is an average daily operation mileage per the last round-trip cycle.
According to a further technical scheme, when the wheel diameter value reaches the limit date, the wheel diameter value is estimated to be the limit date which is the latest turning time of the wheel plus the residual running time of the wheel, and three limit date are obtained, namely the farthest limit date, the latest limit date and the average limit date.
In a third aspect, a system for predicting life of critical components of a wheel set is disclosed, comprising:
a predictive model obtaining module configured to: obtaining an optimal prediction model of the wheel set based on the method;
a real-time residual life estimation module configured to: the running speed, the ambient temperature and the driving mileage of the wheel pair are obtained in real time and preprocessed, a corresponding prediction model is selected, the residual wheel diameter abrasion is output, the real-time prediction of the residual wheel diameter abrasion is realized, and the residual life of the wheel pair is estimated in real time based on the real-time predicted residual wheel diameter abrasion.
Compared with the prior art, the invention has the beneficial effects that:
due to the remaining life assessment of the wheel set, it is largely estimated by the wheel diameter value to the expiration date. Therefore, the evaluation of the remaining life of the wheel set comprises the specific processes of predicting the turning rear wheel diameter value, selecting the related factors such as mileage, temperature and speed, establishing a prediction model of the turning rear wheel diameter value, predicting the future turning rear wheel diameter value, and calculating the wheel diameter value to the nearest date, the average date and the farthest date according to the turning rear wheel diameter value and the average wear rate, the average value, the maximum value and the minimum value of all turning times per ten-thousand kilometers to realize the evaluation of the remaining life of the wheel set.
The machine learning algorithm utilizes a large amount of data to analyze the data, and excavates the intrinsic characteristic rules, so that the intrinsic implicit rules can be found from the data level, the subjectivity of manual participation is greatly reduced, and the calculation result can be more objective and accurate.
The invention can provide reasonable predictive suggestions and guide to appoint reasonable and economic maintenance plans and maintenance tasks according to the life prediction of the PHM-based application maintenance key components.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a schematic diagram showing the influence of abrasion and turning on the wheel tread of a motor train unit
FIG. 2 is a roadmap for fault detection and life prediction;
FIG. 3 is a schematic diagram of a prediction scheme;
FIG. 4 is a technical route diagram for evaluating the residual life of the wheel set according to an embodiment of the invention;
FIG. 5 is a calculation route map of a wheel diameter to a date limit according to an embodiment of the present invention.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The first embodiment is as follows:
in the present embodiment, the wheel set life prediction is exemplified, but it does not mean that the solution provided by the present invention is only applicable to the life prediction of the wheel set. The method can also be applied to the life prediction of other objects according to different prediction objects.
In this embodiment, a method for constructing a life prediction model of a key component of a wheel set is disclosed, which includes:
the method comprises the following steps: and analyzing the service scene and determining the input and output parameters of the model.
The remaining wear loss of the wheel diameter is related to the inner diameter of the wheel and the remaining wheel diameter of the wheel, and the inner diameter of the wheel is a fixed value, i.e. the related quantity of the remaining wheel diameter of the wheel needs to be found out. Because the wheel wear is caused by the continuous increase of the driving mileage and the difference of the driving speed and the environmental temperature, a prediction model between the speed, the temperature, the driving mileage and the wheel diameter residual wear amount is established, and the daily residual wear amount of the wheel can be monitored according to the daily vehicle running condition. Therefore, the input parameters of the model are determined to be temperature, speed and mileage, and the output parameters are the residual abrasion loss of the wheel diameter.
Step two: and (4) collecting and extracting data.
The current data temperature, speed, mileage and the like are stored in a big data platform Hbase and can be obtained in real time; the tuning data is stored in the mysql database, and is periodically transmitted from each service station to the tuning detail table corresponding to mysql offline. A large amount of modeling data needs to be extracted from different databases. And the modeling data is divided into a training set and a verification set according to a cross verification mode based on the modeling data.
Step three: and (4) preprocessing data.
The data preprocessing mainly comprises the removal of dirty data, completion of data missing, standardization processing of data and the like in the modeling data in each table.
3-1) dirty data removal:
certain errors or overrun abnormal values exist in the speed, temperature and mileage returning process; for the situation that the uploading of the tuning data is incomplete due to offline uploading, the data are deleted if the manual input is mistakenly filled, and the like, otherwise, the modeling accuracy is seriously influenced;
3-2) data missing completion:
the turning data is uploaded off-line, and only the turning front wheel diameter value and the turning rear wheel diameter value during turning are obtained, so that serious deficiency of modeling data occurs, and therefore, compensation is required to be performed according to the transmission frequency of speed, temperature and mileage based on a difference algorithm to construct an enough modeling sample;
3-3) data standardization treatment:
for temperature, speed, mileage, wheel diameter values and the like, the difference from a data layer is multiple orders of magnitude, optimization interference can be caused to the construction of a prediction model, and the model cannot be found to be optimal all the time. Therefore, the parameters are preprocessed so that the model parameters are under a uniform metric, such as a global normalization to [ 1, 1 ].
Step four: and (5) establishing and optimizing a model.
The modeling of the prediction object is a many-to-one prediction mode, so that a multi-dimensional prediction algorithm needs to be selected for model construction. However, the actual situation is different for each wheel, the environmental impact is also different, and the received impact is caused by multiple aspects. Therefore, model selection adopts the idea of an integrated algorithm, and the selection of a single prediction model has the following: and (3) constructing models respectively by using a random forest, a multi-dimensional linear regression, a neural network and the like, and finally selecting the model with the highest verification sample accuracy as the optimal prediction model of the round.
Step five: and (5) predicting the model in real time.
And calling a prediction model of the corresponding wheel position through real-time speed, temperature and driving mileage to predict the residual wear consumption of the wheel diameter in real time, and obtaining the residual wear consumption of the wheel diameter after the wheel is turned for the last time.
Example two:
based on the method of the first embodiment, a prediction model is obtained, and based on the output result of the model, namely the real-time wheel diameter residual wear loss, the service life of the key parts in the wheel pair is predicted, as shown in the attached fig. 4 and 5, the service life prediction process comprises the following steps:
(1) solving the wheel diameter abrasion rate per ten thousand kilometers:
the wheel diameter abrasion rate per ten thousand kilometers of the latest turning is (last turning rear wheel diameter value-latest turning front wheel diameter value)/operating mileage (ten kilometers) in the latest turning period.
The wheel diameter values refer to the wheel diameter value of the same wheel pair (the wheel diameter value of the left wheel + the wheel diameter value of the right wheel)/2.
Further, the wear rates of all turning times of the wheel are obtained, and the maximum value and the minimum value of the wheel diameter wear rates of all turning times are obtained. The final average of the maximum, minimum, and most recent wear rates is obtained.
(2) Calculating the number of the remaining running kilometers and the number of the remaining running days:
the remaining operating kilometers (average/maximum/minimum) are the remaining wear loss of the wheel diameter after the last turning of the shaft/(average wear rate/maximum wear rate/minimum wear rate per ten kilometers of wheel diameter at the last turning), and the remaining wear loss of the wheel diameter after the last turning of the shaft is the wheel diameter after the last turning-the wheel inner diameter.
The average daily driving distance in the last turning period is (the accumulated running distance of the last turning vehicle-the accumulated running distance of the last turning vehicle)/the date of the last turning (unit: day).
The remaining operating time (days) is the average daily operating mileage of the remaining operating kilometers per the last round-trip cycle.
(3) Calculating the wheel diameter value to the limit date:
the expected limit date of the wheel diameter value is the latest turning time of the wheel + the remaining running time (day) of the wheel, and of course, three limit-reaching dates are obtained, namely the farthest limit date, the latest limit date and the average limit date.
In the calculation process, the data is periodically transmitted back to a big data platform from each road bureau service station off line, is stored in a turning detail table corresponding to mysql data, and is extracted from the turning detail table through a Python program, and the method comprises the following steps: the system comprises an affiliation bureau, a service station, a vehicle type, a train number, a carriage number, a shaft position, a wheel position, a left wheel once, a right wheel turning rear wheel diameter value, a left wheel last, a right wheel turning front wheel diameter value, a repairing cycle running mileage last and the like.
From the perspective of the full life cycle of the wheel, in the initial use stage of the wheel, the wheel is not concerned much about the wheel reaching the limited date; however, as the service time of the wheel increases, the number of times of repair increases, which causes the wear of the wheel and the reduction of the wheel diameter, the date from the wheel diameter value to the limit needs to be paid attention to, the date from the longest to the limit can be selected in this period, the date from the average to the limit can be paid attention to as the service cycle of the wheel diameter increases, the date from the latest to the limit needs to be paid attention to, and the driving safety can be affected because of the greater potential safety hazard to the running of the wheel.
The invention utilizes a machine learning algorithm to evaluate the wheel pair residual life, establishes a wheel pair abrasion loss prediction model by analyzing relevant factors such as abrasion conditions of wheel parts, inputs wheel pair data acquired in real time into the wheel pair abrasion loss prediction model to realize the real-time prediction of the wheel pair abrasion loss, and evaluates the wheel pair residual life according to the abrasion loss prediction result, driving mileage and other relevant data.
Example three:
with the rapid development of a big data artificial intelligence technology, a large amount of historical data is analyzed and mined from the data perspective, rules hidden behind the data and implicit mechanism characteristics are found, powerful support is provided for the safe and economic operation monitoring of the rail transit motor train unit, and a new solution idea is provided for the PHM intelligent operation and maintenance system to provide early fault warning and residual life assessment.
The implementation example discloses a system for predicting the life of a critical part of a wheel set, which comprises:
a predictive model obtaining module configured to: obtaining an optimal prediction model of the wheel set based on the method;
a real-time residual life estimation module configured to: the running speed, the ambient temperature and the driving mileage of the wheel pair are obtained in real time and preprocessed, a corresponding prediction model is selected, the residual wheel diameter abrasion is output, the real-time prediction of the residual wheel diameter abrasion is realized, and the residual life of the wheel pair is estimated in real time based on the real-time predicted residual wheel diameter abrasion.
The method adopts a prediction method based on data driving and integrates a mechanism model to realize the residual life evaluation of the wheel set.
Due to the remaining life assessment of the wheel set, it is largely estimated by the wheel diameter value to the expiration date. Therefore, the specific process of evaluating the remaining life of the wheel set comprises the steps of predicting the turning rear wheel diameter value, selecting factors such as mileage, temperature and speed related to the turning rear wheel diameter value, establishing a prediction model of the turning rear wheel diameter value, predicting the future turning rear wheel diameter value, and calculating to obtain the date from the turning rear wheel diameter value to the nearest limit, the average date and the farthest date according to the turning rear wheel diameter value and the average wear rate, the average value, the maximum value and the minimum value of all turning times per ten-thousand kilometers; and the evaluation on the residual life of the wheel set is realized.
According to the wheel diameter value after the last turning, the wheel diameter value to the limited date or interval can be calculated; the round diameter to the limit date or interval is further predicted by increasing the mileage to predict the round diameter value of the round after the next round-turning period.
The method comprises the steps of evaluating the residual life by using a machine learning wheel, reflecting the whole-day loss degree of the wheel by adopting four parts of the wheel tread wear loss, the wheel rim wear loss, the wheel inner diameter difference and the wheel hub difference through analyzing relevant factors such as the wear condition of each part of the wheel, establishing a prediction model, and evaluating and predicting the residual life of the wheel by combining the wheel geometric part wear data and the development trend with the driving mileage; an accurate prediction result can be obtained.
In some embodiments, referring to fig. 3, the prediction method based on data driving of the present invention, i.e. the regression prediction algorithm based on machine learning, is shown. Current more popular machine learning prediction algorithms include linear regression, random forests, long short term memory networks (LSTM), and the like.
In some implementation examples, the method for solving the problem of the predictive repair of the PHM key parts of the motor train unit by using the machine learning algorithm can have various solutions, and the optimal machine learning algorithm can be selected to solve the problem of the predictive repair of the PHM key parts of the motor train unit according to the comprehensive consideration of multiple factors such as actual service scenes, data characteristic data quantity and the like. The method can be fused with a business mechanism model based on a machine learning algorithm, solves some complex business problems together, finally realizes fault early warning and residual life assessment of key components and the like, and provides auxiliary support for safe operation and economic operation of the motor train unit.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A method for building a life prediction model of a key part of a wheel set is characterized by comprising the following steps:
determining the input parameters of the model as the environment temperature, the running speed and the running mileage, and the output parameters as the residual abrasion loss of the wheel diameter;
acquiring the current environmental temperature, the driving speed and the driving mileage of the wheel pair in real time, and acquiring turning data of the wheel pair as sample data;
preprocessing the sample data;
and respectively constructing models according to the preprocessed sample data, and finally selecting the model with the highest verification sample accuracy as the optimal prediction model of the current wheel pair.
2. The method for constructing the wheel set key component life prediction model according to claim 1, wherein the tuning data is stored in a database so as to be periodically transmitted back to a corresponding tuning detail table of the database from each service station offline.
3. The method for constructing the life prediction model of the key components of the wheel set according to claim 1, wherein the sample data is preprocessed at least by removing dirty data from the sample data, completing missing data, and normalizing the data;
preferably, abnormal values with certain errors or overrun in the speed, temperature and mileage returning process and incomplete turning data uploading are removed;
preferably, the data are subjected to complete turning according to the transmission frequency of speed, temperature and mileage based on a difference algorithm;
preferably, the temperature, speed, mileage, and wheel diameter values are normalized so that the data is under a uniform metric.
4. A method for predicting the service life of key parts of a wheel set is characterized by comprising the following steps:
obtaining an optimal predictive model of the wheel-set based on the method of any of the preceding claims 1-3;
the running speed, the ambient temperature and the driving mileage of the wheel pair are obtained in real time and preprocessed, a corresponding prediction model is selected, the residual wheel diameter abrasion is output, the real-time prediction of the residual wheel diameter abrasion is realized, and the residual life of the wheel pair is estimated in real time based on the real-time predicted residual wheel diameter abrasion.
5. The method for predicting the service life of the key parts of the wheel pair as claimed in claim 4, wherein after an optimal prediction model of the wheel pair is obtained, the wheel diameter abrasion rate per ten thousand kilometers is solved, then the number of remaining running kilometers and the number of remaining running days are calculated, and finally the wheel diameter value to the date limit is calculated.
6. The method for predicting the service life of the wheel pair key component as claimed in claim 4, wherein when the wheel diameter wear rate per ten thousand kilometers is solved, the wheel diameter wear rate per ten thousand kilometers of the latest turning is equal to (the last turning rear wheel diameter value-the latest turning front wheel diameter value)/the operating mileage in the latest turning period;
the wheel diameter values above all refer to (left wheel diameter value + right wheel diameter value)/2 of the same wheel pair.
7. A method for predicting the lifetime of a wheel pair critical component as claimed in claim 6, wherein the wear rate of all turning times of the wheel is obtained, the maximum and minimum values of the wheel diameter wear rate of all turning times are obtained, and finally the maximum, minimum and most recent average values of the wear rate are obtained.
8. The method for predicting the service life of a wheel pair key component as claimed in claim 4, wherein when the remaining number of operating kilometers and the remaining number of operating days are calculated, the remaining operating kilometers (average/maximum/minimum) is the remaining wheel diameter wear amount after the most recent turning of the wheel axle/(average wear rate/maximum wear rate/minimum wear rate per kilometer of wheel diameter at the most recent turning);
further preferably, the remaining wear loss of the wheel diameter after the latest turning of the wheel shaft is the wheel diameter-wheel inner diameter after the latest turning;
average daily operating mileage in the latest turning period is (accumulated traveling mileage of the latest turning vehicle-accumulated traveling mileage of the last turning vehicle)/latest turning date-latest turning date;
further preferably, the remaining operation time is an average daily operation mileage per the last round-trip cycle.
9. A method as claimed in claim 4, wherein when calculating the deadline for the wheel diameter, the wheel diameter is predicted to be the latest turn-on time of the wheel + the remaining operating time of the wheel, and three deadline dates are obtained, namely the farthest deadline, the latest deadline and the average deadline.
10. A system for predicting the life of a critical component of a wheel set, comprising:
a predictive model obtaining module configured to: obtaining an optimal predictive model of the wheel-set based on the method of any of the preceding claims 1-3;
a real-time residual life estimation module configured to: the running speed, the ambient temperature and the driving mileage of the wheel pair are obtained in real time and preprocessed, a corresponding prediction model is selected, the residual wheel diameter abrasion is output, the real-time prediction of the residual wheel diameter abrasion is realized, and the residual life of the wheel pair is estimated in real time based on the real-time predicted residual wheel diameter abrasion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111489662.6A CN114117687A (en) | 2021-12-08 | 2021-12-08 | Method and system for building and predicting life prediction model of key parts of wheel set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111489662.6A CN114117687A (en) | 2021-12-08 | 2021-12-08 | Method and system for building and predicting life prediction model of key parts of wheel set |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114117687A true CN114117687A (en) | 2022-03-01 |
Family
ID=80367366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111489662.6A Pending CN114117687A (en) | 2021-12-08 | 2021-12-08 | Method and system for building and predicting life prediction model of key parts of wheel set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114117687A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925933A (en) * | 2022-06-16 | 2022-08-19 | 北京交通大学 | Method for realizing environmental vibration control by optimizing wheel rail abrasion intervention time |
| 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 |
| CN117302293B (en) * | 2023-08-14 | 2024-06-07 | 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 |
-
2021
- 2021-12-08 CN CN202111489662.6A patent/CN114117687A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925933A (en) * | 2022-06-16 | 2022-08-19 | 北京交通大学 | Method for realizing environmental vibration control by optimizing wheel rail abrasion intervention time |
| 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 |
| CN117302293B (en) * | 2023-08-14 | 2024-06-07 | 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 |
|---|---|---|
| CN112580153B (en) | Health state management system and method for vehicle running gear monitoring component | |
| CN103617110B (en) | Server device condition maintenance system | |
| Li et al. | Prediction of railcar remaining useful life by multiple data source fusion | |
| CN111114519B (en) | Railway vehicle brake fault prediction method and health management system | |
| CN110376003A (en) | Intelligent whole train service life prediction technique and its system based on BIM | |
| CN110222436B (en) | Method and device for evaluating health state of train parts and storage medium | |
| WO2014110099A2 (en) | Railway track geometry defect modeling for predicting deterioration, derailment risk, and optimal repair | |
| JP2022537937A (en) | Monitoring, Predicting, and Maintaining the Condition of Railroad Track Elements Using Digital Twins | |
| CN111401712A (en) | Urban rail transit risk assessment system and method | |
| CN113177650A (en) | Predictive maintenance method and device for wagon compartment | |
| CN114862275B (en) | Storage logistics system reliability assessment method and system based on digital twin model | |
| Braga et al. | Multivariate statistical aggregation and dimensionality reduction techniques to improve monitoring and maintenance in railways: The wheelset component | |
| Furch | A model for predicting motor vehicle life cycle cost and its verification | |
| CN114117687A (en) | Method and system for building and predicting life prediction model of key parts of wheel set | |
| Appoh et al. | Hybrid adaptive model to optimise components replacement strategy: A case study of railway brake blocks failure analysis | |
| CN110262460B (en) | Concrete piston fault prediction method for extracting features by combining clustering idea | |
| CN113313529A (en) | Finished oil sales amount prediction method based on time regression sequence | |
| Apezetxea et al. | Experimental validation of a fast wheel wear prediction model | |
| Shi et al. | Centralized maintenance time prediction algorithm for freight train wheels based on remaining useful life prediction | |
| CN115081784A (en) | Fault prediction and health management system for locomotive wheels | |
| CN115601005A (en) | Intelligent auxiliary decision-making method and device for railway engineering equipment technical improvement and major repair establishment | |
| CN114179858A (en) | Wheel turning method and device based on wheel health state | |
| RU2724915C1 (en) | Monitoring of worn element state | |
| Tang et al. | Machine Learning Based Estimation of Residual Useful Life of High-Speed Train Wheels Based on Vehicle-Mounted Vibration Sensor Data | |
| Etschmaier | Transit bus maintenance management in the United States |
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 |