CN103745229A - Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) - Google Patents
Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) Download PDFInfo
- Publication number
- CN103745229A CN103745229A CN201410009600.4A CN201410009600A CN103745229A CN 103745229 A CN103745229 A CN 103745229A CN 201410009600 A CN201410009600 A CN 201410009600A CN 103745229 A CN103745229 A CN 103745229A
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring data
- fault
- assembly
- monitoring
- 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
Abstract
The invention relates to a method and a system of fault diagnosis of a rail transit based on an SVM (Support Vector Machine). The method comprises the following steps of collecting historical monitoring data and real-time monitoring data of the rail transit and transmitting the historical monitoring data and the real-time monitoring data to a data analysis server to carry out preprocessing, feature selection, data vectoring and model training, computing, analyzing and classifying the real-time monitoring data according to a classification model which is obtained through the historical monitoring data, judging whether a fault exists and obtaining a reason for generating the fault. The system comprises a data collection component, a data storage component, a data preprocessing component, a feature selection component, a data vectoring component, a model training component and a real-time data analysis component. According to the method and the system, manual judgment and analysis of the fault in mass monitoring signals can be replaced by an automatic monitoring manner, so that a large amount of labor cost and the time for analyzing the reason of the fault can be reduced for providing a time guarantee for subsequent jobs, such as maintenance and rescue.
Description
Technical field
The invention provides a kind of track traffic method for diagnosing faults and system based on SVM, relate to railway signal data, railway communication data, railway knowledge data, system alarm data, machine learning, SVM(support vector machine), etc. technical field, in order to solve the data analysis problems of track traffic Monitoring Data.
Background technology
At present, track traffic (government railway, enterprise railway and urban track traffic) field, monitoring and maintenance product mainly contain three classes: CSM (centralized signal supervision system), each plant maintenance machine, communication network management system.In order to improve the modernization maintenance level of China railways signal system equipment, since the nineties, successively independent development the constantly centralized signal supervision CSM systems during upgrading such as TJWX-I type and TJWX-2000 type.Current most of station has all adopted computer monitoring system, realized the Real-Time Monitoring to signaling at stations equipment state, and by the main running status of inspecting and recording signalling arrangement, for telecommunication and signaling branch, grasping the current state of equipment and carry out crash analysis provides basic foundation, has brought into play vital role.And, to Urban Rail Transit Signal equipment, concentrate monitoring CSM system to be also widely deployed in city rail cluster/rolling stock section etc. and locate, for city rail O&M.In addition, follow the construction development of China Express Railway, the distinctive RBC system of high ferro, TSRS system, ATP system, be also faced with the demand of including centralized signal supervision system in, also be faced with and improve its monitoring capability, O&M ability, and the demand of equipment self-diagnosis ability.
Data mining analysis is the mathematical knowledge that utilizes statistical study, analyzes the data such as text, image, numerical value, finds default rule, the relation of data, sets up data model, for data are classified, the operation such as cluster, statistics.SVM is very ripe data classification algorithm, supports two class classification and multicategory classifications.The main thought of SVM is: it is to analyze for linear separability situation, for the situation of linearly inseparable, by using non-linear map that the sample of low-dimensional input space linearly inseparable is converted into high-dimensional feature space, make its linear separability, thereby make high-dimensional feature space adopt linear algorithm to carry out linear analysis to the nonlinear characteristic of sample, become possibility.
The mining analysis of track traffic Monitoring Data, has great importance for the technical failure of judgement and analysis track traffic.Mostly be at present by manually carry out judgement and the analysis of fault in the Monitoring Data of magnanimity, need a large amount of human costs and the time of failure reason analysis, thereby be difficult to ensure for the work such as follow-up maintenance, rescue provide the time, thereby need the more efficient track traffic Analysis on monitoring data of research and failure analysis methods.Along with the development of track traffic monitoring technology, increasing monitoring equipment is mounted use, and the kind of the Monitoring Data collecting and quantity are also more and more, and using algorithm to substitute to a certain extent manual analysis is an inevitable trend.
Summary of the invention
The object of the invention is to carry out data analysis for track traffic Monitoring Data, utilize SVM to classify to Monitoring Data, can show the operations such as failure cause classification.By the monitoring means of robotization, replace and manually in the Monitoring Data of magnanimity, carry out judgement and the analysis of fault, a large amount of human costs of saving and the time of failure reason analysis, for the work such as follow-up maintenance, rescue provide the time, ensure.
For achieving the above object, the technical solution used in the present invention is as follows:
A track traffic method for diagnosing faults based on SVM, its step comprises:
1) by Historical Monitoring data and the Real-time Monitoring Data of the traffic of purpose data classifying assembly acquisition trajectory, and by these data transmission in data analytics server;
2) data analytics server is stored all kinds of Monitoring Data, and it is carried out to pre-service with by its standardization;
3) reason that the concrete fault of data analytics server analysis and fault produce, carries out feature selecting to Monitoring Data, maps out the Monitoring Data relevant to failure problems;
4) data analytics server is carried out vectorization to characteristic, is converted into the vector space model data that can be processed by SVM;
5) data analytics server is carried out model training according to vector space model to Historical Monitoring data, produces corresponding Question Classification model;
6) data analytics server, according to the disaggregated model being obtained by Historical Monitoring data, is carried out computational analysis and classification to Real-time Monitoring Data, judges whether fault and must be out of order the reason producing.
Further, purpose data classifying assembly comprises that Historical Monitoring purpose data classifying and real time data collect described in step 1), and is connected as the centralized monitoring system (CSM) of station, electricity business section with O&M department.Historical data can be obtained from historical data base, and Real-time Monitoring Data needs to carry out alternately with corresponding data acquisition equipment.
Further, step 2) in data analytics server in storage during Monitoring Data, the Monitoring Data of format is stored among local file system with the form of text, as store in Excel or text, so more contribute to the processing of data, and logarithm Data preprocess step provides data supporting.
Further, step 2) in data analytics server pre-service that Monitoring Data is carried out, comprise and check and process the abnormity point in the Monitoring Data obtaining, check the integrality of data, Monitoring Data to different stations, electricity business section merges, to Monitoring Data convert, the operation such as normalization, with form and the span of uniform data.Historical Monitoring data are carried out to pre-service, comprise the steps such as the cleaning, data-switching, data normalization of data, remove noise data, incomplete data etc., Monitoring Data is standardized.
Further, when data analytics server is carried out feature selecting in step 3), according to the feature of the understanding of problem and data, utilize experience or feature selecting algorithm to select for the relevant data of problem.These data are extracted from raw data.Feature selecting support is artificial, the feature selecting of machine, for problem and SVM, selects suitable characteristic to process.Feature selection module need to be analyzed problem, finds out relevant feature, then the Monitoring Data after pre-service is selected.Input data using relevant feature as model training are processed, rather than process all data.The feature here refers to the value of data monitoring, as the magnitude of voltage of some track equipments, can be used as a feature and processes.
Further, when in step 4), data analytics server is carried out vectorization to data, by the analysis to input data layout, programming realizes the conversion of data layout, and the data of input are converted to vector pattern, is applicable to the vector space model form that SVM processes.If a problem has been selected to n relevant feature, and according to pretreated result, m group data have been comprised.Be exactly n*m data so, form the two-dimensional matrix of the capable n row of m.Also be equivalent to obtain m vector, each vector has n data.The value of feature is exactly the value of corresponding two-dimensional array relevant position.
Further, when data analytics server is carried out model training in step 5), first select suitable kernel, as linear kernel, the kernel of graph, tree core, polynomial kernel, neural network core, RBF core etc., different kernel functions is applicable to dissimilar problem.The selection of kernel function is mainly according to the understanding to problem, and the experience of different IPs function is selected.After having selected kernel function, need to classify to data, be divided into training data and test data two parts.Training data, for to model training, obtains corresponding parameter, and then use test data are tested, the generalization ability of verification model.Utilize the mode of ten times of cross validations to increase accuracy rate and the recall rate of category of model.In one embodiment, can be by be divided into ten groups average the VSM Monitoring Data obtaining, be numbered 1-10, carry out model training ten times, select unduplicated numbering as test set at every turn, 9 remaining piece of data are trained as training set, then use different parameters to carry out ten times of cross validations, obtain accuracy rate and parameter corresponding to recall rate more accurately.
Further, when step 6) data analytics server is analyzed Real-time Monitoring Data, need to purpose data classifying assembly, in CSM system, gather real-time track traffic Monitoring Data and carry out the step identical with Historical Monitoring data, comprise the steps such as pre-service, feature selecting, data vector, then utilize the disaggregated model having obtained, these data are classified, break down judging whether, and produce the reason of fault.
Step 1) to 6 above) can be for the malfunction monitoring data analysis of device level and O&M level.Device level Analysis on monitoring data is that data analysis algorithm carries out Monitoring Data collection, processing, model generation and fault analysis for the equipment of some appointments; O&M level Analysis on monitoring data is collection, processing, model generation and the fault analysis of carrying out Monitoring Data for a certain class fault of whole service system.
A track traffic fault diagnosis system based on SVM, comprising:
Purpose data classifying assembly, is positioned at track traffic O&M department, for Historical Monitoring data and the Real-time Monitoring Data of acquisition trajectory traffic, and is transferred in data analytics server;
Data analytics server, comprising:
Data storage component, connects described purpose data classifying assembly, all kinds of Monitoring Data that send over for storing purpose data classifying assembly;
Data pre-processing assembly, connects described data storage component, for Monitoring Data being carried out to pre-service with by its standardization;
Feature selecting assembly, connects described data pre-processing assembly, and the reason producing for analyzing concrete fault and fault, carries out feature selecting to Monitoring Data, maps out the Monitoring Data relevant to failure problems;
Data vector assembly, connects described feature selecting assembly, for characteristic is carried out to vectorization, is converted to the manageable vector space model data of SVM;
Model training assembly, connects described data vector assembly, for Historical Monitoring data are carried out to model training, produces corresponding Question Classification model;
Real-time data analysis assembly, connects described data vector assembly and described model training assembly, for Real-time Monitoring Data is carried out to computational analysis and classification, judges whether fault and must be out of order the reason producing.
Further, described track traffic O&M department comprises each station, electricity business section, and described purpose data classifying assembly is connected with the centralized monitoring system (CSM) of these O&M departments.Historical data can be obtained from historical data base, and Real-time Monitoring Data needs to carry out alternately with corresponding data acquisition equipment.
The present invention utilizes SVM to classify to Monitoring Data, by the monitoring means of robotization, is replaced manually and in the Monitoring Data of magnanimity, is carried out judgement and the analysis of fault, and compared with prior art, advantage is as follows:
1) the present invention has accelerated the speed of Fault Identification, adopt svm classifier device to carry out Fault Identification to track Traffic monitoring data, can accelerate the speed of Fault Identification, by Real-time Monitoring Data is analyzed, can find fast fault, and identify out of order type.
2) the present invention, by the identification fault that uses a model, has saved a large amount of human costs, no longer needs artificial going to observe monitoring information and then carries out Fault Identification and analysis.
3) the present invention expands by cloud platform, and Monitoring Data is carried out to distributed storage and parallel computation, can solve the Storage and Processing problem of ever-increasing track traffic Monitoring Data.Thereby can be more calm should complicated equipment failure and driving accident reason.
4) on basis of the present invention, add the learning ability of algorithm, can constantly improve the ability of Fault Identification, by continuous cumulative learning, can find manually also not sum up the new fault of appearance, and the new reason of fault generation, can improve pre-alerting ability, pre-diagnosis capability.
Accompanying drawing explanation
Fig. 1 is the flow chart of steps of track traffic Analysis on monitoring data method of the present invention.
Fig. 2 is the assembly connection diagram of track traffic Analysis on monitoring data system of the present invention.
Fig. 3 is the process flow diagram of O&M level track fault analysis example of the present invention.
Fig. 4 is the process flow diagram of device level track fault analysis example of the present invention.
Embodiment
Below by specific embodiments and the drawings, the present invention is described in detail.
Flow process of Monitoring Data being carried out to analyzing and processing based on SVM of the present invention as shown in Figure 1, is mainly divided into model training stage and real-time data analysis stage.Input data comprise Historical Monitoring data and Real-time Monitoring Data, output be the analysis result for Real-time Monitoring Data.By the present invention, can to Monitoring Data, analyze fast the real-time analysis results such as failure cause that obtain.
Fig. 2 is the corresponding system composition diagram of realizing above-mentioned treatment scheme.Mainly comprise: be positioned at the purpose data classifying assembly of O&M department, the data storage component that is positioned at data analytics server, data pre-processing assembly, feature selecting assembly, data vector assembly, model training assembly and real-time data analysis assembly.
Analysis on monitoring data model is mainly by two steps: the one, according to given training set, find suitable SVM kernel function and parameter, and be commonly referred to the model training stage; The 2nd, use the function model that the first step has been trained to analyze Real-time Monitoring Data, to obtain the reason that whether system breaks down and fault produces.Below in conjunction with Fig. 1 and Fig. 2, illustrate the function of each assembly.
1, purpose data classifying assembly
Take the Chinese railway system as example, comprise the purpose data classifying assembly that is positioned at each workshop, electricity business section, Railway Bureau, railway main office; Purpose data classifying assembly is connected with the centralized monitoring system (CSM) of correspondence position, obtains data wherein.The Monitoring Data of wherein obtaining is divided into Historical Monitoring data and Real-time Monitoring Data; Historical Monitoring data the model training stage use, for to model training to obtain disaggregated model; The model that obtains of training is for Real-time Monitoring Data is classified, to obtain the current running status of system, as whether there being the reason etc. of fault and fault.
Purpose data classifying assembly is connected with data analytics server, and purpose data classifying assembly is transferred to the data storage component in data analytics server by the Monitoring Data getting.During concrete enforcement, purpose data classifying assembly can be software module, and CSM provides an interface, and purpose data classifying assembly calls this interface exactly, the regular data of obtaining.Because different CSM data differences is larger, collects assembly and need to identify various data layouts.
2, data storage component
Data storage component is arranged in data analytics server, supports the data storage of format, half format and unformatted.Historical Monitoring data, because data volume is larger, for the ease of parallel processing, generally adopt the mode of file to store.Further can adopt distributed file system to carry out the storage of Historical Monitoring data, and adopt parallel computation framework to calculate Monitoring Data, to improve the ability of data storage and the ability that data are calculated.Data storage component externally provides the interface of data access.Purpose data classifying assembly utilizes itself and being connected of data analytics server, and the data memory interface of calling data memory module, is stored in Historical Monitoring data and Real-time Monitoring Data in data analytics server.
3, data pre-processing assembly
Data pre-processing assembly is arranged in data analytics server, by and data storage component between be connected, the interface of calling data access, carries out pre-service to the Monitoring Data obtaining.First check correctness and the integrality of data, then process accordingly, as data strip deletion etc.Further, Monitoring Data is normalized the Monitoring Data collection that form format is correct, sample space is complete.
4, feature selecting assembly and data vector assembly
Feature selecting assembly is connected with data pre-processing assembly, after having carried out feature selecting, by with being connected of data pre-processing assembly, the Monitoring Data collection of handling well is carried out to sampling, map out only relevant with feature partial data, form new Monitoring Data collection.Feature selecting assembly is connected with data vector assembly, and the Monitoring Data collection mapping out is transferred to data vector assembly, and data vector assembly carries out space vector to data, forms the data of the VSM form of SVM support.
5, model training assembly
Model training assembly is connected with data vector assembly, obtains the Monitoring Data of VSM form by this connection, then uses different parameters to carry out ten times of cross validations to these data.To obtain classification and best model and the parameter of general Huaneng Group power.By with being connected of real-time data analysis assembly, the model training is transferred to real-time data analysis assembly.
6, real-time data analysis assembly
Real-time data analysis assembly is connected with data vector assembly, and is connected with model training assembly.Real-time Monitoring Data also needs experience and flow process like Historical Monitoring data class, finally using the Real-time Monitoring Data of VSM form as input, be input to real-time data analysis assembly, by calculating, just can obtain current system and whether have specific fault, and the reason of this fault generation.
The meaning that SVM carries out in data analysis in track Traffic monitoring data is the data analysis processing that Historical Monitoring is arrived, and then obtains forecast model.Then utilize this mathematical model to carry out analyzing and processing to the Real-time Monitoring Data collecting, can obtain real-time analysis result, as fault pre-alarming, abnormal alarm etc.Avoided the waste of human resource that adopts manual analysis mode to cause, and the analysis difficulty that causes of the analysis result delay that too dependency analysis experience and focus, manual analysis bring and magnanimity Monitoring Data.Specifically, when the present invention utilizes SVM to carry out the analysis of track traffic Monitoring Data, comprise the following steps:
1) collect data
During SVM data analysis, need to use two class data, track traffic Historical Monitoring data and track traffic Real-time Monitoring Data:
Track traffic Historical Monitoring data: Historical Monitoring data need to comprise the partial data while there is all kinds of failure condition, for obtaining the feature of data in all kinds of situations;
Track traffic Real-time Monitoring Data: Real-time Monitoring Data refers to the real time data collecting from track traffic, SVM, to this part data analysis processing, can monitor under current running environment, whether breaks down, and the analysis of causes of fault.
2) Monitoring Data pre-service
When Historical Monitoring data and Real-time Monitoring Data are carried out to Treatment Analysis, need to carry out pre-service to Monitoring Data, comprise data cleansing, data integration, data transformation and data stipulations.
The dirty datas such as the storage of track traffic Monitoring Data is imperfect, inconsistent, cause carrying out data analysis or Result poor.For improve data analysis mass formation Data Preprocessing Technology.Data pre-service has several different methods: data scrubbing, data integration, data transformation, data reduction etc.These data processing techniques were used before data analysis, had greatly improved the quality of data analysis pattern, reduced the needed time of actual excavation.
Data cleansing: data scrubbing is by filling in the value of disappearance, smooth noise data, identification or deleting outlier and solve inconsistency and carry out " cleaning " data.Mainly to reach following target: standardized format, abnormal data is removed, error correcting, the removing of repeating data.Railway monitoring data sometimes there will be the information such as the condition of instant error of the monitor values such as voltage, circuit, can't impact system, therefore need these data to clean.Remove some exceptional values, if there is the situation of some monitor value vacancy, according to empirical value, fill up, or corresponding Monitoring Data is deleted.
Data integration: data integration combines the data in multiple data sources and unifies storage, and in fact the process of setting up data warehouse is exactly data integration.May there is mutual relationship between distinct device, website very much in track traffic Monitoring Data, data are integrated with to be beneficial to the relation between equipment, website is excavated, analyzed, as carry out association analysis etc.
Data transformation: by level and smooth gathering, Data generalization, the modes such as standardization become to be applicable to the form of data analysis by data-switching.Monitoring Data is because the signal difference of monitoring, as electric current, voltage etc., unlike signal value type and span difference larger, some is analog quantity, some is Boolean quantity.Need to carry out stipulations to these data, so more be conducive to the degree of accuracy of foundation and the analysis of model.
Data stipulations: during data analysis, often data volume is very large, on low volume data, carry out mining analysis and need long time, the reduction that data reduction techniques can be used for obtaining data set represents, it is much smaller, but still close to keeping the integrality of former data, and before result and reduction, come to the same thing or almost identical.
3) feature selecting
Feature selecting is also feature subset selection (FSS, Feature Subset Selection).Refer to and from an existing M feature (Feature), select N feature to make the specific indexes optimization of system.In this article, feature extraction is mainly to analyze according to concrete problem, finds out relative feature, then utilizes SVM to analyze these features, rather than whole features is analyzed.In this article, feature refers to the data that Monitoring Data obtains, and the kind of Monitoring Data is many, sample frequency is high, and therefore data volume is larger.For particular problem as track circuit failure judgement analyze time, it is only relevant to little several features, rather than whole Monitoring Data.By feature selecting, can greatly reduce the calculated amount of data, when improving computing velocity, can also avoid introducing too much uncorrelated feature, thereby improve the degree of accuracy of analyzing.
4) data vector
When SVM carries out Treatment Analysis to data, need to be converted to specific form-vector space model (VSM), vector space model is used to calculate the similarity of text the earliest.VSM concept is simple, and the processing of content of text is reduced to the vector operation in vector space, and it expresses semantic similarity with the similarity on space, visual and understandable.When document is represented as the vector of document space, just can measure the similarity between document by the similarity between compute vector.Therefore need programming to realize pretreated Monitoring Data is carried out to format conversion, become the data of vector space model form.
5) model training
The essence of SVM is that data are classified, for the study machine of finite sample situation, realization be structural risk minimization: between the precision of the data approximation to given and the complicacy of approximating function, seek to trade off, to obtaining best general Huaneng Group power.What its finally solved is a convex quadratic programming problem, and in theory, what obtain will be globally optimal solution, solve unavoidable local extremum problem in neural net method.It is transformed into practical problems the feature space of higher-dimension by nonlinear transformation, in higher dimensional space, construct linear decision function and realize the non-linear decision function in former space, solved dexterously problem of dimension, and guaranteed good Generalization Ability, and algorithm complex and sample dimension are irrelevant.Final purpose is by finding an optimum lineoid to divide data.
SVM supports polytype kernel function, as linear kernel, the kernel of graph, tree core, polynomial kernel, neural network core etc.Different IPs function is for different problems, in example herein, chooses RBF(radial basis function) core is as the kernel function of SVM.The selection of kernel function is generally based on experience.
RBF core is a kind of conventional kernel function.It is the most conventional kernel function in support vector machine classification.About the RBF of two sample x and x', endorse the proper vector that is expressed as certain " input space " (input space), shown in it is defined as follows:
Because the value of RBF kernel function with distance reduce, and between the 0(limit) and 1(in the time of x=x') between, so it is a kind of ready-made similarity measurement representation.The feature space of core has infinite many dimensions; For σ=1, its expansion is:
To the training of this paper SVM model, be in fact to find and can make two best parameters C of classifying quality and R by training data, make the svm classifier model based on RBF kernel there is best classification capacity and general Huaneng Group power.
In the process of training, adopt the mode of ten times of cross validations to carry out the accuracy rate of estimation model.Data set is divided into very, in turn will be wherein 9 parts as training data, 1 part as test data, tests.Each test all can draw corresponding accuracy (or error rate).The mean value of the accuracy (or error rate) of the result of 10 times, as the estimation to arithmetic accuracy, generally also needs to carry out repeatedly 10 folding cross validations (for example 10 10 folding cross validations), then asks its average, as the estimation to algorithm accuracy.
6) real-time data analysis
The Real-time Monitoring Data collecting is also needed to carry out treatment scheme above, comprise the steps such as data characteristics selection, data pre-service, space vector, then use the model that trained to analyze, can draw the result of classification, here the reason of fault etc. namely.
The analysis that the present invention utilizes SVM to carry out track traffic Monitoring Data can be analyzed for the fault of O&M level and two kinds of ranks of device level.The fault analysis of O&M level is analyzed take fault as unit, object is that certain fault of whole system is analyzed to identification, when fault is analyzed, all Monitoring Data relevant to fault be need to first obtain, then for these Monitoring Data, data analysis and fault diagnosis carried out.The fault analysis of device level is analyzed take equipment as unit, and object is to identify all faults of some equipment, only need to obtain all Monitoring Data of this equipment when model training and fault analysis.Below by example, O&M level and device level fault analysis are described respectively.
A) O&M level fault diagnosis is implemented
Fig. 3 is the schematic diagram of O&M level Analysis on monitoring data.O&M level fault diagnosis is deployed in this programme in equipment O&M platform, adopts private server to carry out the storage of Monitoring Data, and data acquisition equipment stores the data of collection into data analytics server by Ethernet; Data analytics server is carried out data analysis to the Monitoring Data obtaining, and the fault diagnosis model obtaining is stored in the knowledge base of equipment O&M platform.The Real-time Monitoring Data gathering for data acquisition assembly, data analytics server completes the fault diagnosis to equipment by calling fault diagnosis model in knowledge base, and according to the result of fault diagnosis, fault diagnosis model in knowledge base is assessed and revised.
Below by the analysis example of concrete fault, further illustrate the operational process of such scheme.
Track circuit failure is the most common failure of track traffic, and fault is divided into two large classes: indoor fault and outdoor fault.In existing system, when track circuit breaks down, be, need related personnel to recall relevant Monitoring Data, then the value of comprehensive various Monitoring Data is carried out discriminatory analysis, to obtain the type of fault.Technical capability and the experience of this analytical approach to staff has quite high requirement, and when fault occurs, needs the manual relevant Monitoring Data that finds to analyze.Will inevitably need like this regular hour to operate and analyze judgement.
Known by above-mentioned analysis, although existing system has been realized monitoring and statistics to track traffic circulation data, but the analysis of Monitoring Data is main still by manually carrying out, caused the waste of human resources and time resource, for track traffic, when system breaks down, must to fault, rush to repair timely and get rid of.
Fig. 3 is the process flow diagram that 25Hz phase-sensitive track circuits are distinguished indoor and outdoor fault.The track circuit failure analysis of causes shown in this figure is a classification problem, is applicable to very much using SVM to analyze judgement, below in conjunction with data, uses SVM to carry out automatic discriminatory analysis to fault type.As shown in Figure 3, carrying out fault while judging, relevant Monitoring Data comprises: junction box is subject to terminal voltage, throws cable terminal outside voltage, sending end voltage away, and the type of surveying data is analog quantity.
First the Analysis on monitoring data of O&M level needs to determine the fault that will analyze, and the Monitoring Data relevant to this fault.Two parts are not the targets (25Hz phase-sensitive track circuits are distinguished indoor and outdoor fault) of setting fault analysis above, and the Monitoring Data relevant to this fault (junction box is subject to terminal voltage, throws cable terminal outside voltage, sending end voltage etc. away).After having determined fault and relevant data, components of data analysis will be analyzed gathering the pretreated Monitoring Data of cake, to obtain fault diagnosis model.
After data pre-service and feature selecting, the feature extraction result completing is as shown in table 1.
Table 1. feature extraction result
| Junction box is subject to terminal voltage | Throw cable terminal outside voltage away | Sending end voltage |
| 25.00 | 25.00 | 25.00 |
| 24.00 | 24.00 | 25.00 |
| 27.00 | 27.00 | 27.00 |
| 0.00 | 0.00 | 0.00 |
| … | … | … |
For the purpose of simplifying the description, in upper table, the normal voltage value of three test points is all set to 25v.The classification of type of fault is as follows:
(1) non-fault;
(2) fault is indoor;
(3) fault is outdoor;
(4) indoor short circuit;
(5) indoor open circuit.
Above-mentioned data are carried out to vectorization, so that SVM calculates:
Instance data position:
0?1:25.02:25.03:25.0
0?1:25.02:25.03:25.0
0?1:25.02:25.03:25.0
4?1:30.02:25.03:25.0
4?1:30.02:35.03:20.0
1?1:0.02:0.03:0.0
2?1:0.02:25.03:25.0
3?1:0.02:50.03:25.0
3?1:15.02:50.03:25.0
1?1:0.02:0.03:0.0
1?1:0.02:0.03:0.0
The type of a top column of figure representing fault:
● 0 represents not have fault
● 1 represents that fault is indoor
● 2 represent that fault is outdoor
● 3 represent indoor short circuit
● 4 represent indoor open circuit
Because data volume is more just listed instance data, the input using these data as SVM is trained, and can obtain forecast model, then by inputting different test datas, can obtain the result that track circuit failure is analyzed.
B) device level fault diagnosis embodiment
Device level fault diagnosis can be deployed in this programme in special data analytics server and also can equally with equipment acquisition component be deployed in monitoring of equipment workstation.When data portion is deployed in data analytics server, similar with the treatment scheme of the equipment Inspection data of O&M level.When being deployed in local monitoring station, collection, storage and the analysis of data all can complete in workstation.Components of data analysis is carried out data analysis to device history Monitoring Data, and the fault diagnosis model obtaining is stored in local knowledge base.For equipment Real-time Monitoring Data, by calling fault diagnosis model in knowledge base, complete the fault diagnosis to equipment, and according to the result of fault diagnosis, fault diagnosis model in knowledge base is assessed and revised.
Fig. 4 is the fault diagnosis schematic diagram of power supply panel equipment.The equipment failure cognitron analysis of causes shown in this figure is also a classification problem, is applicable to using SVM to carry out fault analysis.Below in conjunction with data, use SVM to carry out fault analysis and diagnosis to power supply panel.As shown in Figure 4, carrying out fault while judging, relevant Monitoring Data comprises: junction box is subject to terminal voltage, throws cable terminal outside voltage, sending end voltage away, and the type of surveying data is Boolean quantity.
Concrete treatment scheme and O&M level similar, comprises data acquisition, pre-service, feature selecting, model training and Real-time Monitoring Data analysis.Difference is that the fault analysis of device level can carry out and also can in data analytics server, carry out in local monitoring equipment.
Above embodiment is only in order to technical scheme of the present invention to be described but not be limited, and those of ordinary skill in the art can modify or be equal to replacement technical scheme of the present invention, and protection scope of the present invention should be as the criterion with described in claim.
Claims (10)
1. the track traffic method for diagnosing faults based on SVM, its step comprises:
1) by Historical Monitoring data and the Real-time Monitoring Data of the traffic of purpose data classifying assembly acquisition trajectory, and be transferred in data analytics server;
2) data analytics server is stored all kinds of Monitoring Data, and it is carried out to pre-service with by its standardization;
3) reason that the concrete fault of data analytics server analysis and fault produce, carries out feature selecting to Monitoring Data, maps out the Monitoring Data relevant to failure problems;
4) data analytics server is carried out vectorization to characteristic, is converted into the vector space model data that can be processed by SVM;
5) data analytics server is carried out model training according to vector space model to Historical Monitoring data, produces corresponding Question Classification model;
6) data analytics server, according to the disaggregated model being obtained by Historical Monitoring data, is carried out computational analysis and classification to Real-time Monitoring Data, judges whether fault and must be out of order the reason producing.
2. the method for claim 1, it is characterized in that: step 2) in data analytics server in storage during Monitoring Data, the Monitoring Data of format is stored among local file system with the form of text, and logarithm Data preprocess step provides data supporting.
3. the method for claim 1, is characterized in that step 2) in data analytics server pre-service that Monitoring Data is carried out comprise data cleansing, data integration, data transformation and data stipulations.
4. the method for claim 1, it is characterized in that: when in step 3), data analytics server is carried out feature selecting, according to the feature of the understanding of problem and data, utilize experience or feature selecting algorithm to select the data relevant to problem, it is extracted from raw data.
5. the method for claim 1, it is characterized in that: in step 4), data analytics server is by the analysis to input data layout, programming realizes the conversion of data layout, and the data of input are converted to vector pattern and are applicable to the vector space model form that SVM processes.
6. the method for claim 1, it is characterized in that: when in step 5), data analytics server is carried out model training, first select suitable kernel, then data are divided into training data and test data two parts, training data is used for model training, obtain corresponding parameter, use test data are afterwards tested, the general Huaneng Group power of verification model.
7. method as claimed in claim 6, is characterized in that: step 5) utilizes the mode of ten times of cross validations to increase accuracy rate and the recall rate of category of model.
8. method as claimed in claim 7, it is characterized in that: step 5) is divided into ten groups by the VSM Monitoring Data obtaining, be numbered 1-10, carry out model training ten times, select unduplicated numbering as test set at every turn, 9 remaining piece of data are trained as training set, then use different parameters to carry out ten times of cross validations, obtain accuracy rate and parameter corresponding to recall rate more accurately.
9. the track traffic fault diagnosis system based on SVM, is characterized in that, comprising:
Purpose data classifying assembly, is positioned at track traffic O&M department, for Historical Monitoring data and the Real-time Monitoring Data of acquisition trajectory traffic, and is transferred in data analytics server;
Data analytics server, comprising:
Data storage component, connects described purpose data classifying assembly, all kinds of Monitoring Data that send over for storing purpose data classifying assembly;
Data pre-processing assembly, connects described data storage component, for Monitoring Data being carried out to pre-service with by its standardization;
Feature selecting assembly, connects described data pre-processing assembly, and the reason producing for analyzing concrete fault and fault, carries out feature selecting to Monitoring Data, maps out the Monitoring Data relevant to failure problems;
Data vector assembly, connects described feature selecting assembly, for characteristic is carried out to vectorization, is converted to the manageable vector space model data of SVM;
Model training assembly, connects described data vector assembly, for Historical Monitoring data are carried out to model training, produces corresponding Question Classification model;
Real-time data analysis assembly, connects described data vector assembly and described model training assembly, for Real-time Monitoring Data is carried out to computational analysis and classification, judges whether fault and must be out of order the reason producing.
10. system as claimed in claim 9, is characterized in that: described purpose data classifying assembly and described data analytics server are carried out data transmission by Ethernet; Or described purpose data classifying assembly and described data analytics server are integrated in a workstation, by data bus, carry out data transmission.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410009600.4A CN103745229A (en) | 2013-12-31 | 2014-01-09 | Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) |
| CN202110230869.5A CN112949715A (en) | 2013-12-31 | 2014-01-09 | SVM (support vector machine) -based rail transit fault diagnosis method |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310751395 | 2013-12-31 | ||
| CN201310751395.4 | 2013-12-31 | ||
| CN201410009600.4A CN103745229A (en) | 2013-12-31 | 2014-01-09 | Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110230869.5A Division CN112949715A (en) | 2013-12-31 | 2014-01-09 | SVM (support vector machine) -based rail transit fault diagnosis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103745229A true CN103745229A (en) | 2014-04-23 |
Family
ID=50502246
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110230869.5A Pending CN112949715A (en) | 2013-12-31 | 2014-01-09 | SVM (support vector machine) -based rail transit fault diagnosis method |
| CN201410009600.4A Pending CN103745229A (en) | 2013-12-31 | 2014-01-09 | Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110230869.5A Pending CN112949715A (en) | 2013-12-31 | 2014-01-09 | SVM (support vector machine) -based rail transit fault diagnosis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN112949715A (en) |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104596767A (en) * | 2015-01-13 | 2015-05-06 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
| CN105467975A (en) * | 2015-12-29 | 2016-04-06 | 山东鲁能软件技术有限公司 | Equipment fault diagnosis method |
| CN105548764A (en) * | 2015-12-29 | 2016-05-04 | 山东鲁能软件技术有限公司 | Electric power equipment fault diagnosis method |
| CN106199332A (en) * | 2016-06-29 | 2016-12-07 | 中国铁路总公司 | A kind of track circuit red band method for diagnosing faults of improvement |
| CN106210615A (en) * | 2015-04-30 | 2016-12-07 | 北京文安智能技术股份有限公司 | A kind of city management automatic monitoring method, Apparatus and system |
| CN106274991A (en) * | 2016-08-23 | 2017-01-04 | 四川网达科技有限公司 | Railway fail analysis device and method |
| CN106404441A (en) * | 2016-09-22 | 2017-02-15 | 宁波大学 | Nonlinear similarity index based fault classification and diagnosing method |
| CN106407732A (en) * | 2016-12-09 | 2017-02-15 | 珠海安润普科技有限公司 | Method, device and system for identifying states of target object and intelligent clothes |
| CN104091070B (en) * | 2014-07-07 | 2017-05-17 | 北京泰乐德信息技术有限公司 | Rail transit fault diagnosis method and system based on time series analysis |
| CN106897792A (en) * | 2017-01-10 | 2017-06-27 | 广东广业开元科技有限公司 | A kind of structural fire protection risk class Forecasting Methodology and system |
| CN106933956A (en) * | 2017-01-22 | 2017-07-07 | 深圳市华成峰科技有限公司 | Data digging method and device |
| CN107065839A (en) * | 2017-06-06 | 2017-08-18 | 苏州大学 | A kind of method for diagnosing faults and device based on diversity recursion elimination feature |
| CN107144721A (en) * | 2017-05-26 | 2017-09-08 | 北京戴纳实验科技有限公司 | Framework is analyzed for the electric current of laboratory equipment and the big data of voltage |
| CN107577738A (en) * | 2017-08-28 | 2018-01-12 | 电子科技大学 | A kind of FMECA method by SVM text mining processing datas |
| CN108039971A (en) * | 2017-12-18 | 2018-05-15 | 北京搜狐新媒体信息技术有限公司 | A kind of alarm method and device |
| CN108173708A (en) * | 2017-12-18 | 2018-06-15 | 北京天融信网络安全技术有限公司 | Anomalous traffic detection method, device and storage medium based on incremental learning |
| CN108171234A (en) * | 2017-12-26 | 2018-06-15 | 爱保理网（北京）科技有限公司 | A kind of operation association operation state monitoring system and method based on cross validation |
| CN108304941A (en) * | 2017-12-18 | 2018-07-20 | 中国软件与技术服务股份有限公司 | A kind of failure prediction method based on machine learning |
| CN108536945A (en) * | 2018-04-02 | 2018-09-14 | 国网湖南省电力有限公司 | A kind of fault diagnosis method and system for large-scale phase modifier |
| CN108674453A (en) * | 2018-05-23 | 2018-10-19 | 深圳市阡丘越科技有限公司 | A kind of rail traffic monitoring system and control method based on AI technologies |
| CN108700872A (en) * | 2016-02-29 | 2018-10-23 | 三菱电机株式会社 | Machine sort device |
| CN109522715A (en) * | 2018-11-07 | 2019-03-26 | 浙江大学华南工业技术研究院 | A kind of data fusion classification method and system towards safe and intelligent power grid |
| CN109684447A (en) * | 2018-12-13 | 2019-04-26 | 贵州电网有限责任公司 | A kind of dispatching of power netwoks running log fault information analysis method based on text mining |
| CN109917228A (en) * | 2019-03-28 | 2019-06-21 | 国网上海市电力公司 | A kind of traveling wave method distribution net work earthing fault localization method based on RBF neural |
| CN109934759A (en) * | 2019-03-20 | 2019-06-25 | 中国铁道科学研究院集团有限公司 | A kind of locomotive Analysis on monitoring data method and system |
| CN110262417A (en) * | 2019-06-10 | 2019-09-20 | 中云开源数据技术（上海）有限公司 | A kind of production equipment failure predication diagnostic method |
| CN110428109A (en) * | 2019-08-07 | 2019-11-08 | 佳都新太科技股份有限公司 | Subway shield door time between failures prediction model is established and prediction technique |
| CN111077871A (en) * | 2018-10-19 | 2020-04-28 | 北京全路通信信号研究设计院集团有限公司 | Railway signal system fault intelligent analysis platform |
| CN111260822A (en) * | 2019-12-31 | 2020-06-09 | 杭州安脉盛智能技术有限公司 | Rail transit vehicle health state analysis method and terminal based on big data |
| CN111650898A (en) * | 2020-05-13 | 2020-09-11 | 大唐七台河发电有限责任公司 | Distributed control system and method with high fault tolerance performance |
| CN112181955A (en) * | 2020-09-01 | 2021-01-05 | 西南交通大学 | Data standard management method for information sharing of heavy haul railway comprehensive big data platform |
| CN112469613A (en) * | 2018-05-16 | 2021-03-09 | 西门子交通奥地利有限责任公司 | Method and device for diagnosing and monitoring vehicles, vehicle components and traffic lanes |
| CN112540580A (en) * | 2020-10-30 | 2021-03-23 | 红云红河烟草(集团)有限责任公司 | Method for monitoring equipment state of wire making and feeding system based on historical production data |
| CN112578213A (en) * | 2020-12-23 | 2021-03-30 | 交控科技股份有限公司 | Fault prediction method and device for rail power supply screen |
| CN112858804A (en) * | 2020-12-25 | 2021-05-28 | 广西交控智维科技发展有限公司 | Power supply screen fault diagnosis method and device for railway vehicle |
| CN113101057A (en) * | 2020-01-10 | 2021-07-13 | 尤妮佳股份有限公司 | Collection device, collection method, and computer-readable recording medium |
| CN113485262A (en) * | 2021-06-29 | 2021-10-08 | 华能(浙江)能源开发有限公司玉环分公司 | SVM-based fault analysis method for fuel system of thermal power plant |
| CN114442543A (en) * | 2021-10-29 | 2022-05-06 | 南京河海南自水电自动化有限公司 | Computer monitoring method suitable for early warning of hydropower station fault |
| CN117076861A (en) * | 2023-08-18 | 2023-11-17 | 深圳市深国际湾区投资发展有限公司 | Data fusion-based related data processing system, method and medium |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113109666B (en) * | 2021-04-09 | 2024-03-15 | 河南省博海大数据科技有限公司 | Rail circuit fault diagnosis method based on deep convolutional neural network |
| CN114666374B (en) * | 2022-03-31 | 2023-05-02 | 金陵科技学院 | Automatic construction method for fault path of communication system special for rail transit |
| CN114996258A (en) * | 2022-06-23 | 2022-09-02 | 中铁第四勘察设计院集团有限公司 | Contact network fault diagnosis method based on data warehouse |
| CN116523722A (en) * | 2023-06-30 | 2023-08-01 | 江西云绿科技有限公司 | Environment monitoring analysis system with machine learning capability |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050105791A1 (en) * | 2003-10-29 | 2005-05-19 | Lee Ken K. | Surface inspection method |
| CN102314598A (en) * | 2011-09-22 | 2012-01-11 | 西安电子科技大学 | Retinex theory-based method for detecting human eyes under complex illumination |
| CN102682601A (en) * | 2012-05-04 | 2012-09-19 | 南京大学 | Expressway traffic incident detection method based on optimized support vector machine (SVM) |
| CN102749585A (en) * | 2011-04-21 | 2012-10-24 | 李昌 | Multi-layer SVM (support vector machine) based storage battery on-line monitoring method |
| CN103234767A (en) * | 2013-04-21 | 2013-08-07 | 蒋全胜 | Nonlinear fault detection method based on semi-supervised manifold learning |
-
2014
- 2014-01-09 CN CN202110230869.5A patent/CN112949715A/en active Pending
- 2014-01-09 CN CN201410009600.4A patent/CN103745229A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050105791A1 (en) * | 2003-10-29 | 2005-05-19 | Lee Ken K. | Surface inspection method |
| CN102749585A (en) * | 2011-04-21 | 2012-10-24 | 李昌 | Multi-layer SVM (support vector machine) based storage battery on-line monitoring method |
| CN102314598A (en) * | 2011-09-22 | 2012-01-11 | 西安电子科技大学 | Retinex theory-based method for detecting human eyes under complex illumination |
| CN102682601A (en) * | 2012-05-04 | 2012-09-19 | 南京大学 | Expressway traffic incident detection method based on optimized support vector machine (SVM) |
| CN103234767A (en) * | 2013-04-21 | 2013-08-07 | 蒋全胜 | Nonlinear fault detection method based on semi-supervised manifold learning |
Cited By (55)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104091070B (en) * | 2014-07-07 | 2017-05-17 | 北京泰乐德信息技术有限公司 | Rail transit fault diagnosis method and system based on time series analysis |
| CN104596767A (en) * | 2015-01-13 | 2015-05-06 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
| CN104596767B (en) * | 2015-01-13 | 2017-04-26 | 北京工业大学 | Method for diagnosing and predicating rolling bearing based on grey support vector machine |
| CN106210615A (en) * | 2015-04-30 | 2016-12-07 | 北京文安智能技术股份有限公司 | A kind of city management automatic monitoring method, Apparatus and system |
| CN105467975A (en) * | 2015-12-29 | 2016-04-06 | 山东鲁能软件技术有限公司 | Equipment fault diagnosis method |
| CN105548764A (en) * | 2015-12-29 | 2016-05-04 | 山东鲁能软件技术有限公司 | Electric power equipment fault diagnosis method |
| CN105467975B (en) * | 2015-12-29 | 2018-08-03 | 山东鲁能软件技术有限公司 | A kind of equipment fault diagnosis method |
| CN105548764B (en) * | 2015-12-29 | 2018-11-06 | 山东鲁能软件技术有限公司 | A kind of Fault Diagnosis for Electrical Equipment method |
| CN108700872A (en) * | 2016-02-29 | 2018-10-23 | 三菱电机株式会社 | Machine sort device |
| CN108700872B (en) * | 2016-02-29 | 2021-08-06 | 三菱电机株式会社 | Machine sorting device |
| CN106199332A (en) * | 2016-06-29 | 2016-12-07 | 中国铁路总公司 | A kind of track circuit red band method for diagnosing faults of improvement |
| CN106199332B (en) * | 2016-06-29 | 2019-05-07 | 中国铁路总公司 | A kind of improved track circuit red band method for diagnosing faults |
| CN106274991A (en) * | 2016-08-23 | 2017-01-04 | 四川网达科技有限公司 | Railway fail analysis device and method |
| CN106404441A (en) * | 2016-09-22 | 2017-02-15 | 宁波大学 | Nonlinear similarity index based fault classification and diagnosing method |
| CN106407732A (en) * | 2016-12-09 | 2017-02-15 | 珠海安润普科技有限公司 | Method, device and system for identifying states of target object and intelligent clothes |
| CN106897792A (en) * | 2017-01-10 | 2017-06-27 | 广东广业开元科技有限公司 | A kind of structural fire protection risk class Forecasting Methodology and system |
| CN106933956A (en) * | 2017-01-22 | 2017-07-07 | 深圳市华成峰科技有限公司 | Data digging method and device |
| CN107144721A (en) * | 2017-05-26 | 2017-09-08 | 北京戴纳实验科技有限公司 | Framework is analyzed for the electric current of laboratory equipment and the big data of voltage |
| CN107065839B (en) * | 2017-06-06 | 2019-09-27 | 苏州大学 | A kind of method for diagnosing faults and device based on diversity recursion elimination feature |
| CN107065839A (en) * | 2017-06-06 | 2017-08-18 | 苏州大学 | A kind of method for diagnosing faults and device based on diversity recursion elimination feature |
| CN107577738A (en) * | 2017-08-28 | 2018-01-12 | 电子科技大学 | A kind of FMECA method by SVM text mining processing datas |
| CN108173708A (en) * | 2017-12-18 | 2018-06-15 | 北京天融信网络安全技术有限公司 | Anomalous traffic detection method, device and storage medium based on incremental learning |
| CN108304941A (en) * | 2017-12-18 | 2018-07-20 | 中国软件与技术服务股份有限公司 | A kind of failure prediction method based on machine learning |
| CN108039971A (en) * | 2017-12-18 | 2018-05-15 | 北京搜狐新媒体信息技术有限公司 | A kind of alarm method and device |
| CN108171234A (en) * | 2017-12-26 | 2018-06-15 | 爱保理网（北京）科技有限公司 | A kind of operation association operation state monitoring system and method based on cross validation |
| CN108171234B (en) * | 2017-12-26 | 2021-11-02 | 爱保理网（北京）科技有限公司 | Operation complex operation state monitoring system and method based on cross validation |
| CN108536945A (en) * | 2018-04-02 | 2018-09-14 | 国网湖南省电力有限公司 | A kind of fault diagnosis method and system for large-scale phase modifier |
| CN112469613A (en) * | 2018-05-16 | 2021-03-09 | 西门子交通奥地利有限责任公司 | Method and device for diagnosing and monitoring vehicles, vehicle components and traffic lanes |
| CN108674453A (en) * | 2018-05-23 | 2018-10-19 | 深圳市阡丘越科技有限公司 | A kind of rail traffic monitoring system and control method based on AI technologies |
| CN111077871A (en) * | 2018-10-19 | 2020-04-28 | 北京全路通信信号研究设计院集团有限公司 | Railway signal system fault intelligent analysis platform |
| CN111077871B (en) * | 2018-10-19 | 2021-06-22 | 北京全路通信信号研究设计院集团有限公司 | Railway signal system fault intelligent analysis platform |
| CN109522715A (en) * | 2018-11-07 | 2019-03-26 | 浙江大学华南工业技术研究院 | A kind of data fusion classification method and system towards safe and intelligent power grid |
| CN109684447A (en) * | 2018-12-13 | 2019-04-26 | 贵州电网有限责任公司 | A kind of dispatching of power netwoks running log fault information analysis method based on text mining |
| CN109934759B (en) * | 2019-03-20 | 2021-11-09 | 中国铁道科学研究院集团有限公司 | Locomotive monitoring data analysis method and system |
| CN109934759A (en) * | 2019-03-20 | 2019-06-25 | 中国铁道科学研究院集团有限公司 | A kind of locomotive Analysis on monitoring data method and system |
| CN109917228A (en) * | 2019-03-28 | 2019-06-21 | 国网上海市电力公司 | A kind of traveling wave method distribution net work earthing fault localization method based on RBF neural |
| CN110262417A (en) * | 2019-06-10 | 2019-09-20 | 中云开源数据技术（上海）有限公司 | A kind of production equipment failure predication diagnostic method |
| CN110262417B (en) * | 2019-06-10 | 2020-09-01 | 中云开源数据技术（上海）有限公司 | Production equipment fault prediction and diagnosis method |
| CN110428109B (en) * | 2019-08-07 | 2020-11-03 | 佳都新太科技股份有限公司 | Subway shield door fault interval time prediction model establishing and predicting method |
| CN110428109A (en) * | 2019-08-07 | 2019-11-08 | 佳都新太科技股份有限公司 | Subway shield door time between failures prediction model is established and prediction technique |
| CN111260822B (en) * | 2019-12-31 | 2022-07-26 | 杭州安脉盛智能技术有限公司 | Rail transit vehicle health state analysis method and terminal based on big data |
| CN111260822A (en) * | 2019-12-31 | 2020-06-09 | 杭州安脉盛智能技术有限公司 | Rail transit vehicle health state analysis method and terminal based on big data |
| CN113101057A (en) * | 2020-01-10 | 2021-07-13 | 尤妮佳股份有限公司 | Collection device, collection method, and computer-readable recording medium |
| CN113101057B (en) * | 2020-01-10 | 2024-03-26 | 尤妮佳股份有限公司 | Collecting device, collecting method, and computer-readable recording medium |
| CN111650898A (en) * | 2020-05-13 | 2020-09-11 | 大唐七台河发电有限责任公司 | Distributed control system and method with high fault tolerance performance |
| CN111650898B (en) * | 2020-05-13 | 2023-10-20 | 大唐七台河发电有限责任公司 | Distributed control system and method with high fault tolerance performance |
| CN112181955A (en) * | 2020-09-01 | 2021-01-05 | 西南交通大学 | Data standard management method for information sharing of heavy haul railway comprehensive big data platform |
| CN112181955B (en) * | 2020-09-01 | 2022-12-09 | 西南交通大学 | Data standard management method for information sharing of heavy haul railway comprehensive big data platform |
| CN112540580A (en) * | 2020-10-30 | 2021-03-23 | 红云红河烟草(集团)有限责任公司 | Method for monitoring equipment state of wire making and feeding system based on historical production data |
| CN112578213A (en) * | 2020-12-23 | 2021-03-30 | 交控科技股份有限公司 | Fault prediction method and device for rail power supply screen |
| CN112858804A (en) * | 2020-12-25 | 2021-05-28 | 广西交控智维科技发展有限公司 | Power supply screen fault diagnosis method and device for railway vehicle |
| CN113485262A (en) * | 2021-06-29 | 2021-10-08 | 华能(浙江)能源开发有限公司玉环分公司 | SVM-based fault analysis method for fuel system of thermal power plant |
| CN113485262B (en) * | 2021-06-29 | 2022-04-29 | 华能(浙江)能源开发有限公司玉环分公司 | SVM-based fault analysis method for fuel system of thermal power plant |
| CN114442543A (en) * | 2021-10-29 | 2022-05-06 | 南京河海南自水电自动化有限公司 | Computer monitoring method suitable for early warning of hydropower station fault |
| CN117076861A (en) * | 2023-08-18 | 2023-11-17 | 深圳市深国际湾区投资发展有限公司 | Data fusion-based related data processing system, method and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112949715A (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103745229A (en) | Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine) | |
| CN109947086B (en) | Mechanical fault migration diagnosis method and system based on counterstudy | |
| CN103699698A (en) | Method and system for track traffic failure recognition based on improved Bayesian algorithm | |
| CN108830745B (en) | Power grid cascading failure diagnosis, early warning and evaluation system based on monitoring information | |
| CN108320043A (en) | A kind of distribution network equipment state diagnosis prediction method based on electric power big data | |
| CN105574593B (en) | Track state static detection and control system and method based on cloud computing and big data | |
| CN103630244A (en) | Equipment fault diagnosis method and system of electric power system | |
| CN112926666A (en) | Rail transit fault diagnosis method | |
| CN111199257A (en) | Fault diagnosis method and device for high-speed rail driving equipment | |
| CN104077552A (en) | Rail traffic signal comprehensive operation and maintenance method and system based on cloud computing | |
| CN103760901A (en) | Rail transit fault identification method based on association rule classifier | |
| CN103714383A (en) | Rail transit fault diagnosis method and system based on rough set | |
| CN105677791A (en) | Method and system used for analyzing operating data of wind generating set | |
| CN110738523B (en) | Maintenance order quantity prediction method and device | |
| CN114267178B (en) | Intelligent operation maintenance method and device for station | |
| CN108304567B (en) | Method and system for identifying working condition mode and classifying data of high-voltage transformer | |
| CN107340766A (en) | Power scheduling alarm signal text based on similarity sorts out and method for diagnosing faults | |
| CN107124325A (en) | A kind of power communication network operation safety evaluation method and system | |
| CN111652395A (en) | Health assessment method for high-speed railway contact network equipment | |
| CN105608541A (en) | Electric power material supply whole-course early-warning supervise system and method | |
| Brahimi et al. | Development of a prognostics and health management system for the railway infrastructure—Review and methodology | |
| CN106682835A (en) | Data-driven complex electromechanical system service quality state evaluation method | |
| CN111915026A (en) | Fault processing method and device, electronic equipment and storage medium | |
| Shubinsky et al. | Application of machine learning methods for predicting hazardous failures of railway track assets | |
| CN103942251A (en) | Method and system for inputting high altitude meteorological data into database based on multiple quality control methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination |