CN106406296B - It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method - Google Patents
It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method Download PDFInfo
- Publication number
- CN106406296B CN106406296B CN201611154208.4A CN201611154208A CN106406296B CN 106406296 B CN106406296 B CN 106406296B CN 201611154208 A CN201611154208 A CN 201611154208A CN 106406296 B CN106406296 B CN 106406296B
- Authority
- CN
- China
- Prior art keywords
- fault diagnosis
- data
- fault
- cloud
- model
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0221—Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24048—Remote test, monitoring, diagnostic
Abstract
The present invention provides a kind of based on vehicle-mounted and cloud train fault diagnostic system and method, is related to high-speed rail big data fault diagnosis technology field.The system includes vehicle-mounted fault diagnosis subsystem and cloud fault diagnosis subsystem, vehicle-mounted fault diagnosis subsystem is in conjunction with the fault diagnosis subsystem of cloud, it is monitored using driving process of the cloud service to train, cloud data are constantly updated simultaneously, fault diagnosis and fault prediction model is continued to optimize and is updated, and diagnosis process can be carried out visualization display.The present invention can make fault diagnosis and fault prediction model detail more and more perfect, precision is continuously improved, the single real-time fault diagnosis module control of performance beyond tradition, the high-precision requirement highly reliable in the process of moving of train-installed fault diagnosis subsystem is met, the safety of train in the process of moving is improved.
Description
Technical field
The present invention relates to high-speed rail big data fault diagnosis technology fields more particularly to a kind of based on vehicle-mounted and cloud train
Fault diagnosis system and method.
Background technique
Currently, the High-speed Railway Network in China is quickly grown, this plays it economic effect as early as possible has positive effect.Due to
Train running speed is fast, in order to better grasp the operating status of train, diagnoses train fault in time, needs highly reliable train
Fault diagnosis control system, train status monitoring and fault diagnosis system are the important composition portions of railway operation safety security system
Point.
The diagnosis of existing train fault is mainly controlled using onboard system, using the conventional method based on diagnostic rule,
It can obtain and detect the deviation of anticipatory behavior on the basis of state variable and examined with the help of failure and phenomenon of the failure relevant knowledge
Disconnected abort situation.Onboard diagnostic system monitors the relevant element of all subsystems and function.Since bullet train data are with bright
Big data feature is shown, the value of the bullet train failure diagnosis information contained in the big data of bullet train operation and maintenance is huge
Greatly, and data have the characteristics that more type, strong real-time, data volume are big, processing speed is fast, so that existing fault diagnosis system is difficult
To adapt to the above feature.The modeling of data-driven and fault diagnosis obtain many progress in industry in recent years and apply, and are
Bullet train failure, diagnosis provide new approach and means.
Currently, having the research of some method for diagnosing faults based on data model, such as Patent No.
A kind of 201610143119.3 high-speed rail train control on board equipment method for diagnosing faults, this method pass through to mobile unit fault data
Analysis and feature extraction are carried out, decision information table is established in extraction, and is established Bayes's fault diagnosis network and carried out fault diagnosis.It should
Method still has many deficiencies: 1., which are based only upon onboard system, carries out fault diagnosis, and diagnosis efficiency is low;2. model algorithm is single, and
And computation model cannot be improved and be optimized according to train operation situation, train-installed fault diagnosis system diagnostic reliability
It is low.
In conclusion traditional train fault diagnosis is based only upon onboard system, and since train data grows at top speed, and it is vehicle-mounted
System computing capacity and hardware resource are limited, and high-speed rail train is difficult to complete in the process of moving to substantial amounts, Rapid Accumulation
Large-scale data carries out analysis modeling.Simultaneously in view of the special space environment of bullet train, big rule can not be also disposed ON TRAINS
Mould computing cluster calculates mould to carry out mass rapid operation to establish high-precision train Diagnose System Model in real time
Type cannot be improved and be optimized according to train operation situation, cause train-installed fault diagnosis subsystem fault diagnosis rate low,
The highly reliable monitoring requirement of the high-precision not being able to satisfy under bullet train complex situations.Therefore, it is badly in need of studying newly based on cloud
Bullet train big data fault diagnosis system.
Summary of the invention
In view of the drawbacks of the prior art, the present invention provides a kind of based on vehicle-mounted and cloud train fault diagnostic system and side
Method is monitored train using cloud service in the process of moving, while cloud data are constantly updated, and model detail is more and more completeer
Kind, precision is continuously improved, and the single real-time fault diagnosis module control of performance beyond tradition meets train-installed fault diagnosis
System high-precision requirement highly reliable in the process of moving, improves the safety of train in the process of moving.
On the one hand, the present invention provides a kind of based on vehicle-mounted and cloud train fault diagnostic system, which includes vehicle-mounted
Fault diagnosis subsystem and cloud fault diagnosis subsystem.
The vehicle-mounted fault diagnosis subsystem includes vehicle carried data collecting module, real-time fault diagnosis module and cloud diagnosis mould
Block;The cloud fault diagnosis subsystem includes cloud data acquisition module, data memory module, data processing module, failure
Diagnosis and prediction module and data visualization module.
The vehicle carried data collecting module is for being acquired the operation data of train.
The real-time fault diagnosis module is for carrying out real-time fault diagnosis and prediction to train during train operation.
The cloud diagnostic module for call fault diagnosis service provided by the fault diagnosis subsystem of cloud to train into
Row assist type fault diagnosis, i.e., on the one hand to the diagnostic result of vehicle-mounted real-time fault diagnosis module and cloud fault diagnosis subsystem
Fault diagnosis result compare, and comparing result is shown;On the other hand when comparing result shows the vehicle-mounted event
There is when failing to report or misrepresenting deliberately of significant trouble in barrier diagnostic subsystem, and by setting data transport priority, preferential transmitting diagnosis refers to
Order and data utilize cloud diagnostic module to carry out quick auxiliary diagnosis.
The cloud data acquisition module, on the one hand for acquiring the column transmitted in train travelling process by train network
The real time data of vehicle operation, on the other hand for acquiring the historical data of train operation after train operation.
The data memory module is used to data acquisition module data collected in cloud carrying out data cleansing, data turn
It changes and data compression, and data is stored in corresponding data storage system by different types of data;The data cleansing includes data
It fills a vacancy, data are replaced and data format specifications;The data conversion includes data fractionation, data sorting, data deduplication sum number
According to verifying;The data compression is used for compress, to save memory space.
The data processing module includes Computational frame submodule, inquiry submodule, data statistics submodule and algorithms library
Submodule;
The Computational frame submodule includes real-time streaming Computational frame and non real-time batch processing Computational frame, it is described in real time
Streaming computing frame is used for the calculating of real-time stream, and the non real-time batch processing Computational frame is for calculating non real-time history number
According to;
The inquiry submodule for inquire train operation in real time and historical data;
The data statistics submodule is used to carry out statistical disposition to the historical data of train operation;
The algorithms library submodule is used to manage the algorithm of data processing.
The fault diagnosis and fault prediction module includes model foundation submodule, model evaluation submodule, model management submodule
Block and fault diagnosis and fault prediction Attendant sub-module;
The model foundation submodule is used to construct fault diagnosis and fault prediction model using train history data;
The model evaluation submodule is for assessing the diagnosis and prediction effect of fault diagnosis and fault prediction model;
The model management submodule is for all historical failures diagnosis constructed by administrative model setting up submodule and in advance
Survey model;
The fault diagnosis and fault prediction Attendant sub-module is for providing and monitoring fault diagnosis and fault prediction cloud service.
The data visualization module be used for show Various types of data processing operation as a result, include query result, statistics knot
Fruit, calculated result and fault diagnosis result.
On the other hand, the present invention also provides a kind of based on vehicle-mounted and cloud train fault diagnostic method, and this method passes through
It is above-mentioned based on vehicle-mounted and cloud train fault realizing of the diagnosis system, the specific steps are as follows:
Historical data in step 1, vehicle carried data collecting module is uploaded to cloud data acquisition module, cloud by network
Data acquisition module is acquired train data;
Step 2 after carrying out data cleansing and conversion to the initial data that uploads in the data acquisition module of cloud, then carries out
Data storage, specifically includes the following steps:
Step 2.1, using ETL (Extract-Transform-Load) tool to uploading in the data acquisition module of cloud
Initial data carry out data fill a vacancy, data replacement and data format specifications data cleansing operation and data split, number
According to the data transformation operations of sequence, data deduplication and data verification;
Step 2.2 stores the data after cleaning, conversion to corresponding data-storage system, specifically:
Step 2.2.1, structural data is saved in database;
Step 2.2.2, unstructured data is saved in file system;
Step 3 assesses the original fault diagnosis and fault prediction model in cloud using the history data of train, and
New fault diagnosis and fault prediction model is constructed, specifically includes the following steps:
Step 3.1, model evaluation submodule are newly uploaded to the historical data of cloud data acquisition module using train to cloud
The end original fault diagnosis and fault prediction model of fault diagnosis subsystem is assessed, if having wrong report or failing to report situation, executes step
Rapid 3.2, otherwise directly execute step 5;
Step 3.2, the historical data and cloud data-storage system that cloud data acquisition module is newly uploaded to using train
In original historical data establish new fault diagnosis and fault prediction model, be optimal models by the model specification, and by the model
Store model management submodule;
The cloud fault diagnosis service interface that step 3.3, building are remotely accessed for vehicle-mounted fault diagnosis subsystem, executes step
Rapid 4;
Before step 4, train operation, to the fault diagnosis in the real-time fault diagnosis module of vehicle-mounted fault diagnosis subsystem
It is updated with prediction model, specifically includes the following steps:
Step 4.1, before train operation, cloud fault diagnosis subsystem to vehicle-mounted fault diagnosis subsystem real-time therefore
Barrier diagnostic module in fault diagnosis and fault prediction model test, judge the model whether be fault diagnosis and fault prediction effect most
Excellent model;If the model is not the optimal model of fault diagnosis and fault prediction effect, 4.2 are thened follow the steps, vehicle mounted failure is examined
Fault diagnosis and fault prediction model in the real-time fault diagnosis module of disconnected subsystem is updated, if the model be fault diagnosis with
The optimal model of prediction effect, thens follow the steps 5;
Step 4.2 judges whether to need to update that the fault diagnosis and fault prediction model in real-time fault diagnosis module is whole
The model can be made to be optimal, if it is not, i.e. only need to be to the correlation of the fault diagnosis and fault prediction model in real-time fault diagnosis module
Parameter, which is updated, can be such that the model is optimal, and then follow the steps 4.2.1, if so, thening follow the steps 4.2.2;
Step 4.2.1, the Fault diagnosis and forecast in the real-time fault diagnosis module of train-installed fault diagnosis subsystem
The model model optimal compared to the diagnosis and prediction effect in the fault diagnosis and fault prediction module of cloud fault diagnosis subsystem,
Partial parameters only need to be changed, then change the relevant parameter of model;
Step 4.2.2, the fault diagnosis and fault prediction in the real-time fault diagnosis module of train-installed fault diagnosis subsystem
The model model optimal compared to diagnosis and prediction effect in the fault diagnosis and fault prediction module of cloud fault diagnosis subsystem needs
The optimal model of effect is integrally downloaded into vehicle-mounted fault diagnosis subsystem, if the hardware of vehicle-mounted fault diagnosis subsystem calculates
Resource can effectively support the operation for the model that fault diagnosis and fault prediction effect is optimal constructed by the fault diagnosis subsystem of cloud,
Then follow the steps 4.2.2.1;If the hardware computing resource of vehicle-mounted fault diagnosis subsystem can not support cloud fault diagnosis subsystem
The operation of the constructed newest fault diagnosis and fault prediction model of system, thens follow the steps 4.2.2.2;
Step 4.2.2.1, by the complete of optimal fault diagnosis and fault prediction model constructed by the fault diagnosis subsystem of cloud
Model is directly downloaded in train-installed fault diagnosis subsystem;
Step 4.2.2.2, to the newest fault diagnosis and fault prediction model of complexity constructed by the fault diagnosis subsystem of cloud
Fault diagnosis model carries out reduction, the fault diagnosis and fault prediction model after constructing reduction, and downloads it to train-installed failure
In diagnostic subsystem;
Step 5, vehicle-mounted fault diagnosis subsystem carry out real-time fault diagnosis to train, specifically includes the following steps:
Step 5.1, vehicle carried data collecting module are by real-time Data Transmission collected to real-time fault diagnosis module and cloud
Diagnostic module;
Step 5.2, real-time fault diagnosis module carry out real time fail to the train in traveling using the data acquired in real time
Diagnosis, is transferred to cloud diagnostic module for fault diagnosis result;
Step 5.3, cloud diagnostic module are using real time data by calling failure provided by the fault diagnosis subsystem of cloud
Diagnosis and prediction service interface carries out fault diagnosis to train, method particularly includes:
Step 5.3.1, cloud data acquisition module is carried out in train travelling process by the real time data that network uploads
Acquisition;
Step 5.3.2, data cleansing is carried out to the real time data acquired in step 5.3.1 using ETL tool and data turns
After changing operation, corresponding data-storage system is arrived in storage;
Step 5.3.3, cloud fault diagnosis subsystem responds remote fault diagnosis service request, specific response process are as follows:
Step 5.3.3.1, the fault diagnosis and fault prediction model of cloud fault diagnosis subsystem is clear to carrying out in step 5.3.2
Train operating data after washing and converting carries out analysis and fault diagnosis;
Step 5.3.3.2, fault diagnosis result is returned to the cloud diagnostic module of vehicle-mounted fault diagnosis subsystem;
Step 5.4, cloud diagnostic module will by call the service of cloud fault diagnosis and fault prediction be formed by diagnostic result with
Real-time fault diagnosis module in step 5.2 in vehicle-mounted fault diagnosis subsystem is formed by diagnostic result and is compared, and goes forward side by side
Row prompt;If the real-time fault diagnosis module in Comparative result discovery vehicle-mounted fault diagnosis subsystem is failed to report for significant trouble
Or misrepresent deliberately, then follow the steps 5.4.1;If Comparative result has not found to fail to report or misrepresent deliberately, directly execution step 5.4.4;
Step 5.4.1, vehicle-mounted fault diagnosis subsystem is promoted by cloud diagnostic module and calls cloud fault diagnosis and fault prediction
Service carries out fault diagnosis and fault prediction and the priority with cloud communication;
Step 5.4.2, vehicle-mounted fault diagnosis subsystem is improved by cloud diagnostic module and calls cloud fault diagnosis and fault prediction
Service carries out fault diagnosis and fault prediction and communicates occupied bandwidth with cloud, preferential to transmit diagnostic instruction and data;
Step 5.4.3, cloud diagnostic module calls service provided by the fault diagnosis subsystem of cloud to carry out quickly auxiliary and examines
It is disconnected;
Step 5.4.4, cloud diagnostic module is prompted fault diagnosis result in time;
The historical data of this train is uploaded into cloud data acquisition module after step 6, train operation.
As shown from the above technical solution, the beneficial effects of the present invention are: the present invention provide it is a kind of based on vehicle-mounted and cloud
Train fault diagnostic system and method, train is monitored in the process of moving using cloud service, while cloud data are not
Disconnected to update, fault diagnosis and fault prediction model detail is more and more perfect, and precision is continuously improved, the single real time fail of performance beyond tradition
Diagnostic module control, meets the high-precision requirement highly reliable in the process of moving of train-installed fault diagnosis subsystem, improves
The safety of train in the process of moving.Specific effect is: train-installed fault diagnosis subsystem and cloud failure are examined
Disconnected subsystem combines, the dual high reliability that ensure that train-installed fault diagnosis subsystem;Vehicle-mounted fault diagnosis subsystem will
Data are deposited into the data-storage system of cloud, construct unified fault data management system, realize the height to fault data
Effect management;The monitoring of fault diagnosis effect is carried out to the train in traveling by the cloud diagnostic module on train, it is ensured that vehicle-mounted
The high reliability of real-time fault diagnosis;Cloud utilizes fault diagnosis and fault prediction mould constructed by the constantly operation data of accumulation update
Type constantly improve, and precision is continuously improved, the single vehicle-mounted real-time fault diagnosis system of fault diagnosis and fault prediction performance beyond tradition.
Detailed description of the invention
Fig. 1 is provided in an embodiment of the present invention a kind of based on vehicle-mounted and cloud train fault diagnostic system structural block diagram;
Fig. 2 is a kind of functional knot based on vehicle-mounted and cloud train fault diagnostic system provided in an embodiment of the present invention
Composition;
Fig. 3 is a kind of overall procedure based on vehicle-mounted and cloud train fault diagnostic method provided in an embodiment of the present invention
Figure;
Fig. 4 is provided in an embodiment of the present invention a kind of driven based on the train fault diagnostic method in vehicle-mounted and cloud in train
Preceding flow chart;
Fig. 5 be it is provided in an embodiment of the present invention it is a kind of based on the train fault diagnostic method in vehicle-mounted and cloud in train driving
Flow chart in the process.
Specific embodiment
With reference to the accompanying drawings and examples, specific embodiments of the present invention will be described in further detail.Implement below
Example is not intended to limit the scope of the invention for illustrating the present invention.
The present embodiment is a kind of based on vehicle-mounted and cloud train fault diagnostic system, such as Fig. 1 by taking axis reviewing knowledge already acquired barrier diagnosis as an example
It is shown, including vehicle-mounted fault diagnosis subsystem and cloud fault diagnosis subsystem.
Vehicle-mounted fault diagnosis subsystem includes vehicle carried data collecting module, real-time fault diagnosis module and cloud diagnostic module.
Vehicle carried data collecting module is used to be acquired the operation data of train, including several for acquiring different data
Sensor.
Real-time fault diagnosis module is for carrying out real-time fault diagnosis and prediction to train during train operation.
Cloud diagnostic module is auxiliary for calling fault diagnosis service provided by the fault diagnosis subsystem of cloud to carry out train
Help formula fault diagnosis, the i.e. on the one hand event to the diagnostic result of vehicle-mounted real-time fault diagnosis module and cloud fault diagnosis subsystem
Barrier diagnostic result compares, and comparing result is shown;On the other hand when comparing result shows that the vehicle mounted failure is examined
There is when failing to report or misrepresenting deliberately of significant trouble in disconnected subsystem, by setting data transport priority, it is preferential transmit diagnostic instruction and
Data carry out quick auxiliary diagnosis using cloud diagnostic module.
Cloud fault diagnosis subsystem includes cloud data acquisition module, data memory module, data processing module, failure
Diagnosis and prediction module and data visualization module.
On the one hand cloud data acquisition module is transported for acquiring in train travelling process by the train that train network is transmitted
Capable real time data, on the other hand for acquiring the historical data of train operation after train operation.
Data memory module be used to carry out cloud data acquisition module data collected data cleansing, data conversion and
Data compression, and data are stored in corresponding data storage system by different types of data;Data cleansing include data fill a vacancy, data
Replacement and data format specifications;Data conversion includes data fractionation, data sorting, data deduplication and data verification;Data pressure
Contracting is used for compress, to save memory space.
Data processing module includes Computational frame submodule, inquiry submodule, data statistics submodule and algorithms library submodule
Block;Computational frame submodule includes real-time streaming Computational frame and non real-time batch processing Computational frame, real-time streaming Computational frame
For the calculating of real-time stream, non real-time batch processing Computational frame is for calculating non real-time historical data;Submodule is inquired to use
In inquiry train operation in real time and historical data;Data statistics submodule is used to carry out Statistics Division to the historical data of train operation
Reason;Algorithms library submodule is used to manage the algorithm of data processing.
Fault diagnosis and fault prediction module include model foundation submodule, model evaluation submodule, model management submodule and
Fault diagnosis and fault prediction Attendant sub-module;Model foundation submodule be used for using train history data building fault diagnosis with
Prediction model;Model evaluation submodule is for assessing the diagnosis and prediction effect of fault diagnosis and fault prediction model;Model
Submodule is managed for all historical failure diagnosis and prediction models constructed by administrative model setting up submodule;Fault diagnosis with
Prediction Attendant sub-module is for providing and monitoring fault diagnosis and fault prediction cloud service.
Data visualization module by show Various types of data processing operation as a result, include query result, statistical result, based on
Calculate result and fault modeling and prediction result.
A kind of functional structure such as Fig. 2 institute based on vehicle-mounted and cloud train fault diagnostic system provided in this embodiment
Show.
Using a kind of above-mentioned side for carrying out fault diagnosis to axis temperature based on vehicle-mounted and cloud train fault diagnostic system
Method, as shown in figure 3, the specific method is as follows.
Step 1, the axis temperature in the vehicle carried data collecting module of a certain column high-speed rail train and other historical datas pass through network
It is uploaded to cloud data acquisition module, cloud data acquisition module is acquired the train data, these historical datas include
Axis temperature, train sensing data, train operation status information, train number, working line, the driver's number, letter of train operation
Cease classification coding, train operation log etc..
Step 2 after carrying out data cleansing and conversion to the initial data that uploads in the data acquisition module of cloud, then carries out
Data storage, specifically includes the following steps:
Step 2.1, using ETL tool Kettle (a kind of ETL tool of open source) to uploading to cloud data acquisition module
In initial data carry out data fill a vacancy, data replacement and data format specifications data cleansing processing, and to data carry out
Data fractionation, data sorting, data deduplication and data verification data conversion treatment;
Step 2.2 stores the data after cleaning, conversion to corresponding data-storage system, specifically:
Step 2.2.1, structural data is saved in database, specifically included: by train axle temperature data, mode of operation number
According to equal storages in memory database Redis (Key-Value database), data copy write-in local file is backed up,
Storage is into non-relational database HBase after data in Redis are compressed using lossless compression algorithm;High speed is arranged
The storages such as train number, working line, driver's number, the information category coding of vehicle are into relevant database Mysql;
Step 2.2.2, unstructured data is saved in file system, specially by the operation log text of train operator
The storage of the unstructured datas such as part, failure message information is distributed to HDFS (Hadoop Distributed File System)
In file management system.
Step 3 assesses the original fault diagnosis and fault prediction model in cloud using the history data of train, and
New fault diagnosis and fault prediction model is constructed, specifically includes the following steps:
Step 3.1, model evaluation submodule are newly uploaded to the axis temperature historical data of cloud data acquisition module using train
Fault diagnosis subsystem original fault diagnosis and fault prediction model in cloud is assessed, if assessment result determination has wrong report or leakage
Situation is reported, thens follow the steps 3.2, otherwise directly executes step 5;Specific appraisal procedure are as follows:
Step 3.1.1, model evaluation submodule obtains high-speed rail train axle temperature historical data from database, and normalization is (a kind of
Simplify calculate mode, will have the expression formula of dimension, by transformation, turn to nondimensional expression formula, become scalar) processing after structure
Build the test set sample of fault diagnosis and fault prediction model;
Step 3.1.2, the T of test set sample data is calculated2(Hotelling T2Statistic) and square prediction error
(Square Predicted Error, SPE) statistic and its corresponding control limit (i.e. the limitation range of statistical indicator), use
Test set sample data is assessed and is verified to original fault diagnosis and fault prediction model, if having wrong report or failing to report situation, is needed
Fault diagnosis subsystem original fault diagnosis and fault prediction model in cloud is updated, execute step 3.2, otherwise do not update
Model, directly execution step 5;
Step 3.2, the axis temperature historical data that cloud data acquisition module is newly uploaded to using train and cloud data are deposited
Original axis temperature historical data establishes new fault diagnosis and fault prediction model in storage system, is optimal models by the model specification,
And the model is stored to model management submodule;In the present embodiment, the method for establishing new fault diagnosis and fault prediction model are as follows:
High-speed rail train axle temperature is directed to using PCA (Principal Component Analysis, principal component analysis) method
Historical data establishes new Fault diagnosis and forecast model, at database acquisition high-speed rail train axle temperature historical data, normalization
The training set and test set sample that model is constructed after reason, are trained model and verify.
The cloud fault diagnosis service interface that step 3.3, building are remotely accessed for vehicle-mounted fault diagnosis subsystem, executes step
Rapid 4.
Before step 4, train operation, the axis reviewing knowledge already acquired in the real-time fault diagnosis module of vehicle-mounted fault diagnosis subsystem is hindered
Diagnosis and prediction model is updated, specifically includes the following steps:
Step 4.1, before train operation, cloud fault diagnosis subsystem to vehicle-mounted fault diagnosis subsystem real-time therefore
Axis temperature fault diagnosis and fault prediction model in barrier diagnostic module is tested, and judges whether the model is fault diagnosis and fault prediction effect
The optimal model of fruit;If the model is not the optimal model of fault diagnosis and fault prediction effect, 4.2 are thened follow the steps, to vehicle-mounted event
The axis temperature fault diagnosis and fault prediction model hindered in the real-time fault diagnosis module of diagnostic subsystem is updated, if the model is event
Hinder the optimal model of diagnosis and prediction effect, thens follow the steps 5;
Step 4.2 judges whether to need to update that the axis temperature fault diagnosis and fault prediction model in real-time fault diagnosis module is whole
Body just can be such that the model is optimal, if it is not, i.e. only need to be to the axis temperature fault diagnosis and fault prediction mould in real-time fault diagnosis module
The relevant parameter of type, which is updated, can be such that the model is optimal, and then follow the steps 4.2.1, if so, thening follow the steps
4.2.2;
Step 4.2.1, axis temperature fault diagnosis in the real-time fault diagnosis module of train-installed fault diagnosis subsystem and
Prediction model is most compared to the diagnosis and prediction effect in the axis temperature fault diagnosis and fault prediction module of cloud fault diagnosis subsystem
Excellent model only need to change partial parameters, then change the axis temperature fault diagnosis and fault prediction model in vehicle-mounted fault diagnosis subsystem
The relevant parameter T of algorithm2Control with SPE statistic limits;
Step 4.2.2, axis temperature fault diagnosis in the real-time fault diagnosis module of train-installed fault diagnosis subsystem with
Prediction model is optimal compared to fault diagnosis and fault prediction effect in the fault diagnosis and fault prediction module of cloud fault diagnosis subsystem
Axis temperature fault diagnosis and fault prediction model, need the optimal model of effect integrally downloading to vehicle-mounted fault diagnosis subsystem;If
The hardware computing resource of vehicle-mounted fault diagnosis subsystem can effectively support axis reviewing knowledge already acquired constructed by the fault diagnosis subsystem of cloud
The operation for hindering the optimal model of diagnosis and prediction effect, thens follow the steps 4.2.2.1;If the hardware of vehicle-mounted fault diagnosis subsystem
Computing resource can not support the operation of newest axis temperature fault diagnosis and fault prediction model constructed by the fault diagnosis subsystem of cloud,
Then follow the steps 4.2.2.2;
Step 4.2.2.1, by axis temperature fault diagnosis and fault prediction model optimal constructed by the fault diagnosis subsystem of cloud
Complete model with Docker, (Docker is the engine of an open source, it may be convenient to be that any application creates a light weight
Grade, transplantable, self-centered container) form be directly downloaded to the real time fail of train-installed fault diagnosis subsystem
In diagnostic module;
Step 4.2.2.2, to the newest axis temperature fault diagnosis and fault prediction of complexity constructed by the fault diagnosis subsystem of cloud
Symbolic fault diagnosis model carries out reduction, the axis temperature fault diagnosis and fault prediction model after constructing reduction, and by the axis after the reduction
Warm fault diagnosis and fault prediction model downloads to the real-time fault diagnosis of train-installed fault diagnosis subsystem in the form of Docker
In module.
Above-mentioned step is the update weight of the fault diagnosis and fault prediction model carried out before train is driven according to historical data
Process is built, process is as shown in Figure 4.
Real-time fault diagnosis module in step 5, train-installed fault diagnosis subsystem utilizes train shaft temperature sensor institute
The real-time axis temperature data of acquisition carry out real-time fault diagnosis to the train in traveling, as shown in figure 5, specifically includes the following steps:
Step 5.1, vehicle carried data collecting module give axis temperature real-time Data Transmission collected to real-time fault diagnosis module
With cloud diagnostic module;
Step 5.2, real-time fault diagnosis module carry out the train in traveling using the axis temperature data acquired in real time real-time
Fault diagnosis result is transferred to cloud diagnostic module by axis temperature fault diagnosis;
Step 5.3, cloud diagnostic module are by axis temperature fault diagnosis provided by calling cloud fault diagnosis subsystem and in advance
API (Application Programming Interface, the application programming interface) service for surveying service is real-time to train
Axis temperature data carry out fault diagnosis, but the process postpones with the regular hour, method particularly includes:
Step 5.3.1, cloud data acquisition module in train travelling process pass through GSM-R (GSM-R digital mobile communication
System is based on public wireless communication system GSM platform, exclusively for the digital wireless communication system for meeting railway applications and developing
System) network upload real-time axis temperature data be acquired;
Step 5.3.2, using ETL tool to the real-time axis temperature data that are acquired in step 5.3.1 carry out data cleansing (including
Data are filled a vacancy, data are replaced and data format specifications) and data conversion (including data fractionation, data sorting, data deduplication and
Data verification) operation after, storage arrive corresponding data-storage system;
Step 5.3.3, cloud fault diagnosis subsystem responds remote fault diagnosis service request, specific response process are as follows:
Step 5.3.3.1, the axis temperature fault diagnosis and fault prediction model of cloud fault diagnosis subsystem in step 5.3.2 into
Train operation axis temperature data after row cleaning and conversion carry out analysis and fault diagnosis;
Step 5.3.3.2, axis temperature fault diagnosis result is returned to the cloud diagnostic module of vehicle-mounted fault diagnosis subsystem;
Step 5.4, cloud diagnostic module will be examined by calling fault diagnosis and fault prediction service in cloud to be formed by axis reviewing knowledge already acquired barrier
Real-time fault diagnosis module in disconnected result and step 5.2 in vehicle-mounted fault diagnosis subsystem is formed by axis temperature fault diagnosis knot
Fruit is compared, and is prompted;If Comparative result finds the real-time fault diagnosis module needle in vehicle-mounted fault diagnosis subsystem
Great axis reviewing knowledge already acquired barrier is failed to report or misrepresented deliberately, 5.4.1 is thened follow the steps;If Comparative result has not found to fail to report or misrepresent deliberately,
Directly execute step 5.4.4;
Step 5.4.1, vehicle-mounted fault diagnosis subsystem is promoted by cloud diagnostic module and calls cloud fault diagnosis and fault prediction
Service carries out fault diagnosis and fault prediction and the priority with cloud communication;
Step 5.4.2, vehicle-mounted fault diagnosis subsystem is improved by cloud diagnostic module and calls cloud fault diagnosis and fault prediction
Service carries out fault diagnosis and fault prediction and communicates occupied bandwidth with cloud, preferential to transmit diagnostic instruction and data;
Step 5.4.3, cloud diagnostic module calls service provided by the fault diagnosis subsystem of cloud to carry out quickly auxiliary and examines
It is disconnected;
Step 5.4.4, cloud diagnostic module is prompted fault diagnosis result in time.
The historical data of this train is uploaded into cloud data acquisition module after step 6, train operation.
It gets back after this failure diagnostic process new operation data, returns again to and execute step 1.Train operating data
It constantly updates, the data of update are used to optimize or rebuild fault diagnosis and fault prediction model, and optimization or the model rebuild are used to pair again
Train carries out fault diagnosis, and with the continuous operation of system, the process of this method recycles always execution.
Due to the hardware resource limitation of train-installed fault diagnosis subsystem, high-speed rail train generates in the process of moving
Data volume is big, and the operational capability and model accuracy of real-time fault diagnosis module are limited, in addition computation model cannot be according to train
Operating condition is improved and is optimized, and causes train-installed fault diagnosis subsystem reliability not high, is not able to satisfy bullet train
The highly reliable monitoring requirement of high-precision under complex situations.Provided by the invention is diagnosed based on vehicle-mounted and cloud train fault
System and method, provide a kind of vehicle-mounted-cloud frame, the failure in the train-installed fault diagnosis subsystem of effective guarantee
The highly reliable high-precision problem of diagnosis and prediction model, distinctive cloud service can supervise train in the process of moving
Control, while cloud data are constantly updated, model detail is more and more perfect, and precision is continuously improved, and performance beyond tradition is single in real time
Fault diagnosis module control, improves the safety of train in the process of moving.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify to technical solution documented by previous embodiment, or some or all of the technical features are equal
Replacement;And these are modified or replaceed, model defined by the claims in the present invention that it does not separate the essence of the corresponding technical solution
It encloses.
Claims (2)
1. a kind of based on vehicle-mounted and cloud train fault diagnostic system, it is characterised in that: the system includes vehicle-mounted fault diagnosis
Subsystem and cloud fault diagnosis subsystem;
The vehicle-mounted fault diagnosis subsystem includes vehicle carried data collecting module, real-time fault diagnosis module and cloud diagnostic module;
The cloud fault diagnosis subsystem includes cloud data acquisition module, data memory module, data processing module, fault diagnosis
With prediction module and data visualization module;
The vehicle carried data collecting module is for being acquired the operation data of train;
The real-time fault diagnosis module is for carrying out real-time fault diagnosis and prediction to train during train operation;
The cloud diagnostic module is auxiliary for calling fault diagnosis service provided by the fault diagnosis subsystem of cloud to carry out train
Help formula fault diagnosis, the i.e. on the one hand event to the diagnostic result of vehicle-mounted real-time fault diagnosis module and cloud fault diagnosis subsystem
Barrier diagnostic result compares, and comparing result is shown;On the other hand when comparing result shows that the vehicle mounted failure is examined
There is when failing to report or misrepresenting deliberately of significant trouble in disconnected subsystem, by setting data transport priority, it is preferential transmit diagnostic instruction and
Data carry out quick auxiliary diagnosis using cloud diagnostic module;
On the one hand the cloud data acquisition module is transported for acquiring in train travelling process by the train that train network is transmitted
Capable real time data, on the other hand for acquiring the historical data of train operation after train operation;
The data memory module be used to carry out cloud data acquisition module data collected data cleansing, data conversion and
Data compression, and data are stored in corresponding data storage system by different types of data;The data cleansing include data fill a vacancy,
Data replacement and data format specifications;The data conversion includes that data fractionation, data sorting, data deduplication and data are tested
Card;The data compression is used for compress, to save memory space;
The data processing module includes Computational frame submodule, inquiry submodule, data statistics submodule and algorithms library submodule
Block;
The Computational frame submodule includes real-time streaming Computational frame and non real-time batch processing Computational frame, the real-time streaming
Computational frame is used for the calculating of real-time stream, and the non real-time batch processing Computational frame is for calculating non real-time historical data;
The inquiry submodule for inquire train operation in real time and historical data;
The data statistics submodule is used to carry out statistical disposition to the historical data of train operation;
The algorithms library submodule is used to manage the algorithm of data processing;
The fault diagnosis and fault prediction module include model foundation submodule, model evaluation submodule, model management submodule and
Fault diagnosis and fault prediction Attendant sub-module;
The model foundation submodule is used to construct fault diagnosis and fault prediction model using train history data;
The model evaluation submodule is for assessing the diagnosis and prediction effect of fault diagnosis and fault prediction model;
The model evaluation submodule is newly uploaded to the historical data of cloud data acquisition module first with train to cloud event
The barrier original fault diagnosis and fault prediction model of diagnostic subsystem is assessed;If having wrong report or failing to report situation, the model is built
Vertical submodule is newly uploaded to original in the historical data and cloud data-storage system of cloud data acquisition module using train
Historical data establish new fault diagnosis and fault prediction model, be optimal models by the model specification, and the model storage arrived
Then model management submodule constructs the cloud fault diagnosis service interface for the remote access of vehicle-mounted fault diagnosis subsystem;?
Before train operation, the fault diagnosis and fault prediction model in the real-time fault diagnosis module of vehicle-mounted fault diagnosis subsystem is carried out
It updates, judges whether the model is the optimal model of fault diagnosis and fault prediction effect;If the model is not fault diagnosis and fault prediction
The optimal model of effect then judges whether to need to update that the fault diagnosis and fault prediction model in real-time fault diagnosis module is whole
The model can be made to be optimal, if it is not, i.e. only need to be to the correlation of the fault diagnosis and fault prediction model in real-time fault diagnosis module
Parameter, which is updated, can be such that the model is optimal, then changes the relevant parameter of model, if so, needing effect is optimal
Model integrally downloads to vehicle-mounted fault diagnosis subsystem, if the hardware computing resource of vehicle-mounted fault diagnosis subsystem can be propped up effectively
The operation for the model that fault diagnosis and fault prediction effect is optimal constructed by the fault diagnosis subsystem of cloud is held, then is examined cloud failure
The complete model of optimal fault diagnosis and fault prediction model constructed by disconnected subsystem is directly downloaded to train-installed fault diagnosis
In system, if the hardware computing resource of vehicle-mounted fault diagnosis subsystem can not be supported constructed by the fault diagnosis subsystem of cloud most
The operation of new fault diagnosis and fault prediction model, then to the newest fault diagnosis of complexity constructed by the fault diagnosis subsystem of cloud and
Prediction model fault diagnosis model carries out reduction, the fault diagnosis and fault prediction model after constructing reduction, and downloads it to train
In vehicle-mounted fault diagnosis subsystem;If the model is the optimal model of fault diagnosis and fault prediction effect, vehicle-mounted fault diagnosis
System carries out real-time fault diagnosis to train;If the assessment judging result started is not report or fail to report situation by mistake, vehicle-mounted event
Hinder diagnostic subsystem and real-time fault diagnosis is carried out to train;
The model management submodule is for all historical failure diagnosis and prediction moulds constructed by administrative model setting up submodule
Type;
The fault diagnosis and fault prediction Attendant sub-module is for providing and monitoring fault diagnosis and fault prediction cloud service;
The data visualization module by show Various types of data processing operation as a result, include query result, statistical result, based on
Calculate result and fault diagnosis result.
2. a kind of based on vehicle-mounted and cloud train fault diagnostic method, by it is described in claim 1 it is a kind of based on vehicle-mounted and
The train fault realizing of the diagnosis system in cloud, it is characterised in that: specific step is as follows for this method:
Historical data in step 1, vehicle carried data collecting module is uploaded to cloud data acquisition module, cloud data by network
Acquisition module is acquired train data;
Step 2 after carrying out data cleansing and conversion to the initial data that uploads in the data acquisition module of cloud, then carries out data
Storage, specifically includes the following steps:
Step 2.1, using ETL (Extract-Transform-Load) tool to the original uploaded in the data acquisition module of cloud
Beginning data progress data are filled a vacancy, data are replaced and the data cleansing of data format specifications operates and data are split, data row
Sequence, data deduplication and the data transformation operations of data verification;
Step 2.2 stores the data after cleaning, conversion to corresponding data-storage system, specifically:
Step 2.2.1, structural data is saved in database;
Step 2.2.2, unstructured data is saved in file system;
Step 3 is assessed the original fault diagnosis and fault prediction model in cloud using the history data of train, and is constructed
New fault diagnosis and fault prediction model, specifically includes the following steps:
Step 3.1, model evaluation submodule are newly uploaded to the historical data of cloud data acquisition module using train to cloud event
The barrier original fault diagnosis and fault prediction model of diagnostic subsystem is assessed, if having wrong report or failing to report situation, is thened follow the steps
3.2, otherwise directly execute step 5;
Step 3.2 is newly uploaded in the historical data and cloud data-storage system of cloud data acquisition module using train
Original historical data establishes new fault diagnosis and fault prediction model, is optimal models by the model specification, and the model is deposited
Store up model management submodule;
The cloud fault diagnosis service interface that step 3.3, building are remotely accessed for vehicle-mounted fault diagnosis subsystem, executes step 4;
Before step 4, train operation, to the fault diagnosis in the real-time fault diagnosis module of vehicle-mounted fault diagnosis subsystem and in advance
Model is surveyed to be updated, specifically includes the following steps:
Step 4.1, before train operation, cloud fault diagnosis subsystem examines the real time fail of vehicle-mounted fault diagnosis subsystem
Fault diagnosis and fault prediction model in disconnected module is tested, and judges whether the model is that fault diagnosis and fault prediction effect is optimal
Model;If the model is not the optimal model of fault diagnosis and fault prediction effect, 4.2 are thened follow the steps, to vehicle-mounted fault diagnosis
Fault diagnosis and fault prediction model in the real-time fault diagnosis module of system is updated, if the model is fault diagnosis and fault prediction
The optimal model of effect, thens follow the steps 5;
Step 4.2 judges whether to need to update that the fault diagnosis and fault prediction model in real-time fault diagnosis module integrally can just make
The model is optimal, if it is not, i.e. only need to be to the relevant parameter of the fault diagnosis and fault prediction model in real-time fault diagnosis module
Being updated can be such that the model is optimal, and then follow the steps 4.2.1, if so, thening follow the steps 4.2.2;
Step 4.2.1, the Fault diagnosis and forecast model in the real-time fault diagnosis module of train-installed fault diagnosis subsystem
The model optimal compared to the diagnosis and prediction effect in the fault diagnosis and fault prediction module of cloud fault diagnosis subsystem only needs
Partial parameters are changed, then change the relevant parameter of model;
Step 4.2.2, the fault diagnosis and fault prediction model in the real-time fault diagnosis module of train-installed fault diagnosis subsystem
The model optimal compared to diagnosis and prediction effect in the fault diagnosis and fault prediction module of cloud fault diagnosis subsystem, need by
The optimal model of effect integrally downloads to vehicle-mounted fault diagnosis subsystem, if the hardware computing resource of vehicle-mounted fault diagnosis subsystem
The operation that can effectively support the model that fault diagnosis and fault prediction effect is optimal constructed by the fault diagnosis subsystem of cloud, then hold
Row step 4.2.2.1;If the hardware computing resource of vehicle-mounted fault diagnosis subsystem can not support cloud fault diagnosis subsystem institute
The operation of the newest fault diagnosis and fault prediction model of building, thens follow the steps 4.2.2.2;
Step 4.2.2.1, by the complete model of optimal fault diagnosis and fault prediction model constructed by the fault diagnosis subsystem of cloud
It is directly downloaded in train-installed fault diagnosis subsystem;
Step 4.2.2.2, to the newest fault diagnosis and fault prediction model failure of complexity constructed by the fault diagnosis subsystem of cloud
Diagnostic model carries out reduction, the fault diagnosis and fault prediction model after constructing reduction, and downloads it to train-installed fault diagnosis
In subsystem;
Step 5, vehicle-mounted fault diagnosis subsystem carry out real-time fault diagnosis to train, specifically includes the following steps:
Step 5.1, vehicle carried data collecting module diagnose real-time Data Transmission collected to real-time fault diagnosis module and cloud
Module;
Step 5.2, real-time fault diagnosis module carry out real-time fault diagnosis to the train in traveling using the data acquired in real time,
Fault diagnosis result is transferred to cloud diagnostic module;
Step 5.3, cloud diagnostic module are using real time data by calling fault diagnosis provided by the fault diagnosis subsystem of cloud
Fault diagnosis is carried out to train with prediction service interface, method particularly includes:
Step 5.3.1, cloud data acquisition module is acquired in train travelling process by the real time data that network uploads;
Step 5.3.2, data cleansing is carried out to the real time data acquired in step 5.3.1 using ETL tool and data conversion is grasped
After work, corresponding data-storage system is arrived in storage;
Step 5.3.3, cloud fault diagnosis subsystem responds remote fault diagnosis service request, specific response process are as follows:
Step 5.3.3.1, the fault diagnosis and fault prediction model of cloud fault diagnosis subsystem to carried out in step 5.3.2 cleaning and
Train operating data after conversion carries out analysis and fault diagnosis;
Step 5.3.3.2, fault diagnosis result is returned to the cloud diagnostic module of vehicle-mounted fault diagnosis subsystem;
Step 5.4, cloud diagnostic module will be by calling fault diagnosis and fault prediction service in cloud to be formed by diagnostic result and step
Real-time fault diagnosis module in 5.2 in vehicle-mounted fault diagnosis subsystem is formed by diagnostic result and is compared, and is mentioned
Show;If the real-time fault diagnosis module in Comparative result discovery vehicle-mounted fault diagnosis subsystem is failed to report for significant trouble or mistake
Report, thens follow the steps 5.4.1;If Comparative result has not found to fail to report or misrepresent deliberately, directly execution step 5.4.4;
Step 5.4.1, vehicle-mounted fault diagnosis subsystem is promoted by cloud diagnostic module and calls the service of cloud fault diagnosis and fault prediction
Carry out fault diagnosis and fault prediction and the priority with cloud communication;
Step 5.4.2, vehicle-mounted fault diagnosis subsystem is improved by cloud diagnostic module and calls the service of cloud fault diagnosis and fault prediction
It carries out fault diagnosis and fault prediction and communicates occupied bandwidth with cloud, it is preferential to transmit diagnostic instruction and data;
Step 5.4.3, cloud diagnostic module calls service provided by the fault diagnosis subsystem of cloud to carry out quick auxiliary diagnosis;
Step 5.4.4, cloud diagnostic module is prompted fault diagnosis result in time;
The historical data of this train is uploaded into cloud data acquisition module after step 6, train operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611154208.4A CN106406296B (en) | 2016-12-14 | 2016-12-14 | It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611154208.4A CN106406296B (en) | 2016-12-14 | 2016-12-14 | It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106406296A CN106406296A (en) | 2017-02-15 |
CN106406296B true CN106406296B (en) | 2019-04-23 |
Family
ID=58087610
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611154208.4A Active CN106406296B (en) | 2016-12-14 | 2016-12-14 | It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106406296B (en) |
Families Citing this family (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106953847B (en) * | 2017-02-27 | 2020-10-16 | 江苏徐工信息技术股份有限公司 | Method for cross-platform real-time processing of big data based on thrift |
CN107016057A (en) * | 2017-02-28 | 2017-08-04 | 北京铁路局 | Row control vehicle-mounted ATP equipment integral intelligent O&M method and system |
CN107025274A (en) * | 2017-03-21 | 2017-08-08 | 华中科技大学 | Equipment health status intelligent perception system and method based on Hadoop |
CN106970607B (en) * | 2017-03-31 | 2020-10-27 | 株洲中车时代电气股份有限公司 | Testing method and system for converter control system |
CN107697107B (en) * | 2017-09-12 | 2020-03-10 | 中国中车股份有限公司 | Ground intelligent processing system and method for train set fault prediction and health management |
CN107589695B (en) * | 2017-09-12 | 2020-03-10 | 中国中车股份有限公司 | Train set fault prediction and health management system |
CN107679159B (en) * | 2017-09-28 | 2021-06-11 | 百度在线网络技术(北京)有限公司 | Method and device for generating fault diagnosis question response, server and storage medium |
JP6848791B2 (en) * | 2017-09-28 | 2021-03-24 | 株式会社デンソー | Vehicle diagnostic equipment, vehicle diagnostic system and vehicle diagnostic program |
CN109696316B (en) * | 2017-10-20 | 2021-05-28 | 株洲中车时代电气股份有限公司 | Train remote monitoring system |
CN109917772B (en) * | 2017-12-13 | 2020-09-08 | 北京航空航天大学 | PHM rapid prototyping system for remotely evaluating equipment state on line |
CN109948169B (en) * | 2017-12-20 | 2021-05-18 | 中国中车股份有限公司 | Fault prediction and health management system for railway wagon |
CN108038214A (en) * | 2017-12-21 | 2018-05-15 | 重庆脉实智能制造有限公司 | The collection of railway Overhaul Yard section device data, storage, the method and system of analysis and application |
CN110163225A (en) * | 2018-02-11 | 2019-08-23 | 顺丰科技有限公司 | It is a kind of that package detection and recognition methods, apparatus and system are mixed based on cloud platform |
CN110874093B (en) * | 2018-09-03 | 2023-02-21 | 上汽通用汽车有限公司 | Method and system for predicting life of brake lining of vehicle and storage medium |
CN109141898A (en) * | 2018-09-13 | 2019-01-04 | 湖北谊立舜达动力科技有限公司 | A kind of Diagnosis of Diesel Motor system based on Internet of Things |
CN109765448B (en) * | 2019-02-01 | 2021-07-13 | 唐智科技湖南发展有限公司 | Distributed fault diagnosis method, device and system |
CN109915218B (en) * | 2019-03-07 | 2021-09-03 | 东方电气自动控制工程有限公司 | DEH system electrohydraulic conversion part fault diagnosis system |
CN110018425B (en) * | 2019-04-10 | 2021-01-29 | 北京理工大学 | Power battery fault diagnosis method and system |
CN111830927A (en) * | 2019-04-23 | 2020-10-27 | 中车大连电力牵引研发中心有限公司 | Vehicle fault monitoring method and device and vehicle-mounted diagnosis equipment |
CN110428066B (en) * | 2019-07-25 | 2023-04-07 | 中南大学 | Intelligent train electrical device online state evaluation and operation and maintenance system and method thereof |
CN110361207B (en) * | 2019-07-25 | 2021-03-16 | 中南大学 | Intelligent train running gear online state prediction system and method thereof |
CN113085952A (en) * | 2020-01-08 | 2021-07-09 | 株洲中车时代电气股份有限公司 | Train monitoring system |
CN111352116A (en) * | 2020-03-31 | 2020-06-30 | 湖北阿桑奇汽车电子科技有限公司 | Ultrasonic obstacle positioning simulation device for vehicle |
CN111516727A (en) * | 2020-05-12 | 2020-08-11 | 上海数深智能科技有限公司 | High-speed rail defect abnormity intelligent diagnosis and detection system and method based on double vibration measurement sensors |
CN112307335A (en) * | 2020-10-29 | 2021-02-02 | 中国第一汽车股份有限公司 | Vehicle service information pushing method, device and equipment and vehicle |
CN114676844A (en) | 2020-12-24 | 2022-06-28 | 北京百度网讯科技有限公司 | Method, apparatus, device, computer-readable storage medium for automatic driving |
CN113204227B (en) * | 2021-04-26 | 2023-04-18 | 江苏徐工工程机械研究院有限公司 | Cloud collaborative fault diagnosis system and method for layered modular engineering machinery |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102529903A (en) * | 2010-12-31 | 2012-07-04 | 上海博泰悦臻电子设备制造有限公司 | Comprehensive vehicle failure detecting system |
CN103338261A (en) * | 2013-07-04 | 2013-10-02 | 北京泰乐德信息技术有限公司 | Storage and processing method and system of rail transit monitoring data |
CN104133467A (en) * | 2014-07-30 | 2014-11-05 | 浪潮集团有限公司 | OBDS long-distance fault diagnosis and recovery system based on cloud computation |
CN104185309A (en) * | 2014-08-12 | 2014-12-03 | 深圳市元征科技股份有限公司 | On-board wireless local area network equipment |
CN105045256A (en) * | 2015-07-08 | 2015-11-11 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on data comparative analysis |
CN105487532A (en) * | 2016-02-19 | 2016-04-13 | 上海果路交通科技有限公司 | Vehicle-mounted diagnostic data sharing terminal system |
CN107589695A (en) * | 2017-09-12 | 2018-01-16 | 中国中车股份有限公司 | A kind of train groups prognostic and health management system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10140109B2 (en) * | 2014-02-25 | 2018-11-27 | Ford Global Technologies, Llc | Silent in-vehicle software updates |
-
2016
- 2016-12-14 CN CN201611154208.4A patent/CN106406296B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102529903A (en) * | 2010-12-31 | 2012-07-04 | 上海博泰悦臻电子设备制造有限公司 | Comprehensive vehicle failure detecting system |
CN103338261A (en) * | 2013-07-04 | 2013-10-02 | 北京泰乐德信息技术有限公司 | Storage and processing method and system of rail transit monitoring data |
CN104133467A (en) * | 2014-07-30 | 2014-11-05 | 浪潮集团有限公司 | OBDS long-distance fault diagnosis and recovery system based on cloud computation |
CN104185309A (en) * | 2014-08-12 | 2014-12-03 | 深圳市元征科技股份有限公司 | On-board wireless local area network equipment |
CN105045256A (en) * | 2015-07-08 | 2015-11-11 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on data comparative analysis |
CN105487532A (en) * | 2016-02-19 | 2016-04-13 | 上海果路交通科技有限公司 | Vehicle-mounted diagnostic data sharing terminal system |
CN107589695A (en) * | 2017-09-12 | 2018-01-16 | 中国中车股份有限公司 | A kind of train groups prognostic and health management system |
Non-Patent Citations (1)
Title |
---|
"动车组维修决策系统数据仓库的研究与应用";赵虹;《万方硕士论文数据库》;20100316;参见摘要,正文第4-5页,第17-18页,第42-51页 |
Also Published As
Publication number | Publication date |
---|---|
CN106406296A (en) | 2017-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106406296B (en) | It is a kind of based on vehicle-mounted and cloud train fault diagnostic system and method | |
JP7202847B2 (en) | Real-time streaming analytics for flight data processing | |
CN107589695B (en) | Train set fault prediction and health management system | |
Killeen et al. | IoT-based predictive maintenance for fleet management | |
CN104102773B (en) | A kind of equipment fault early-warning and state monitoring method | |
CN113465920B (en) | Cloud, fog and edge end cooperative bearing state monitoring and management method and system | |
US20130268501A1 (en) | System and method for monitoring distributed asset data | |
CN110806743A (en) | Equipment fault detection and early warning system and method based on artificial intelligence | |
US20160313216A1 (en) | Fuel gauge visualization of iot based predictive maintenance system using multi-classification based machine learning | |
CN108597057A (en) | A kind of unmanned plane failure predication diagnostic system and method based on noise deep learning | |
JP2019509565A (en) | Handling prediction models based on asset location | |
US20110314331A1 (en) | Automated test and repair method and apparatus applicable to complex, distributed systems | |
CN112926257A (en) | Reciprocating natural gas compressor fault diagnosis system and diagnosis method | |
US20230184627A1 (en) | System and method for distributed networked test of electric vehicles, storage medium and terminal device | |
CN105775943A (en) | Data driven elevator part early warning system and method | |
CN116107282B (en) | Industrial robot predictive maintenance system based on enterprise application integration | |
CN111651530A (en) | Intelligent port monitoring system | |
CN110723169A (en) | Train information processing method, device and system | |
CN116186946B (en) | Hydraulic system fault diagnosis method and system based on diagnosis model | |
CN116628633A (en) | IGBT real-time monitoring and service life prediction evaluation method | |
CN117172751A (en) | Construction method of intelligent operation and maintenance information analysis model | |
CN109507986A (en) | Long-distance monitoring method for oil field operation equipment | |
CN110058576A (en) | Equipment fault prognostics and health management method based on big data | |
CN112541647A (en) | Risk early warning method and early warning system for oil refining mobile equipment | |
CN103955182A (en) | Safe operation monitoring and instructing method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |