Fault early warning method and system for key equipment in traction substation based on prediction model

Through the fault warning method of key equipment of the traction substation based on the prediction model, the intelligent diagnosis and independent status evaluation of railway traction power supply equipment are realized, the problem of low fault detection efficiency is solved, and the reliability and economic benefits of railway transportation are improved.

CN116720324BActive Publication Date: 2024-11-15CHINA RAILWAY SIYUAN SURVEY & DESIGN GRP CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310555933.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-05-15
Publication Date
2024-11-15
Estimated Expiration
2043-05-15

AI Technical Summary

Technical Problem

The fault detection efficiency of railway traction power supply equipment is low, resulting in potential problems in railway transportation, which makes it difficult to detect and solve faults in a timely manner, affecting transportation reliability and economic benefits.

Method used

The fault warning method of key equipment of the traction substation based on the prediction model is adopted, and through information collection, cluster analysis, data-driven prediction and reliability evaluation, potential faults are identified and maintenance decisions are made to realize intelligent diagnosis and autonomous status evaluation of the equipment.

Benefits of technology

Transform the passive operation and maintenance of railway power supply into active operation and maintenance, improve the efficiency of fault detection, realize intelligent diagnosis and independent status evaluation of traction power supply and substation equipment, optimize maintenance decisions, and improve the reliability and economic benefits of railway transportation. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116720324B_ABST
    Figure CN116720324B_ABST
Patent Text Reader

Abstract

The present application proposes a method and system for early warning of key equipment failures in traction substations based on a prediction model, which relates to the field of power supply and transformation. The method includes: obtaining information data on key equipment information, fault categories, fault modes, fault causes, and fault influencing factors of power supply and transformation through information collection technology; classifying the information data obtained through information collection technology through a clustering method, and performing correlation analysis to mine potential correlation relationships; identifying unconventional data and / or finding data of unconventional discrete points for statistical analysis through the data influence characteristics of the information data in the association process. In order to effectively improve operation and maintenance efficiency, save operation and maintenance costs, and reduce the total workload of maintenance personnel, research and development of early warning and prediction methods for key equipment failures in power supply and transformation are carried out, so as to realize intelligent diagnosis and autonomous comprehensive status evaluation of key equipment in traction power supply and transformation, and make maintenance decisions based on equipment status.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of power supply and transformation, and specifically, to a method and system for early warning of failure of key equipment in a traction substation based on a prediction model. Background Art

[0002] With the development of motor and electrical manufacturing industry, electronic industry and electric power industry, electrified railway transportation has been widely valued by countries around the world for its huge economic benefits and has developed rapidly. Traction power supply refers to the power supply method for the electric energy required for traction vehicle transportation; the traction power supply system is the infrastructure and an important component of electrified railways, providing power energy for trains, and its operating reliability is directly related to the reliability of electrified railways. Due to the complexity of the operating environment and operating conditions, failures of traction power supply and substation equipment, contact network, etc. are inevitable. The traction power supply system refers to the entire power supply system that introduces 220 (110) KV power from the local area, reduces the voltage to 27.5 KV through the traction substation and sends it to the electric locomotive. However, the current railway traction power supply has many equipment and a large span. Once a problem occurs, it is not easy to detect, resulting in problems in railway transportation. Improve the detection efficiency.

[0003] Therefore, railway traction power supply equipment fault warning is a key issue in the railway transportation field. A large amount of on-site operation and maintenance and testing and other complicated and redundant workloads have led to backward production efficiency and low economic benefits in distribution network fault monitoring. Summary of the invention

[0004] The purpose of this application is to provide a method for early warning of key equipment failures in traction substations based on a predictive model, which can realize intelligent diagnosis and autonomous comprehensive status evaluation of key traction power supply and substation equipment, and make maintenance decisions based on the equipment status.

[0005] Another object of the present application is to provide a traction substation key equipment failure warning system based on a prediction model, which can run a traction substation key equipment failure warning method based on a prediction model.

[0006] The embodiment of the present application is implemented as follows:

[0007] In the first aspect, an embodiment of the present application provides a method for early warning of key equipment failures in a traction substation based on a predictive model, which includes obtaining information data on key power supply and substation equipment information, fault categories, fault modes, fault causes, and fault influencing factors through information collection technology; classifying the information data obtained through information collection technology through a clustering method, and performing correlation analysis to mine potential association relationships; identifying unconventional data and / or finding data of unconventional discrete points through data influence characteristics in the association process of the information data for statistical analysis.

[0008] In some embodiments of the present application, the above also includes: predicting the cause and location of possible failure before the failure occurs by testing the status parameters of the equipment, collecting and analyzing fault symptom data, wherein the fault prediction is performed through data-driven, and / or, acquiring data through simulation and / or experiment, and establishing a life model for existing products for prediction.

[0009] In some embodiments of the present application, the above-mentioned data-driven fault prediction also includes: data-driven fault prediction is a process of fault identification through data collection, feature extraction, and trend prediction, that is, the preparation of prediction sample data is completed through data collection, the training of the prediction model is completed through feature extraction of the prediction sample data, and finally the fault trend prediction is performed.

[0010] In some embodiments of the present application, the above also includes: analyzing and evaluating the reliability level of the system as a whole and each device, predicting the remaining life of the equipment, and taking targeted and differentiated protective measures in combination with the various risk factors that may be brought about by the external operating environment, wherein the methods for analyzing and evaluating the reliability level of the system as a whole and each device include analytical methods, Monte Carlo methods and hybrid methods.

[0011] In some embodiments of the present application, the above analytical method includes: building a reliability model according to the functional relationship between system components, integrating all fault states, and analyzing the fault states to finally obtain a reliability index.

[0012] In some embodiments of the present application, the Monte Carlo method includes: assigning random operating states to various components of the system through random sampling, and then calculating the required reliability indicators using probability statistics methods.

[0013] In some embodiments of the present application, the above-mentioned hybrid method includes: using the Monte Carlo method to simulate the random operating state of each component of the system, and then using the analytical method to calculate the average duration of each state, instead of the sampling of state time originally required, thereby reducing the amount of calculation of the analytical method while reducing the iterative convergence time of the Monte Carlo method.

[0014] In some embodiments of the present application, the above also includes: using the triangular whitening weight function in the grey theory to evaluate the technical risk of the project, whitening the grey system with insufficient clear or overall information, so as to obtain correct risk assessment results based on less information.

[0015] In a second aspect, an embodiment of the present application provides a traction substation key equipment fault warning system based on a prediction model, which includes a main statistical quantity statistical submodule for obtaining information data of key power supply and substation equipment information, fault category, fault mode, fault cause, and fault influencing factors through information collection technology;

[0016] The correlation analysis module is used to classify the information data obtained through information collection technology through clustering methods, and to perform correlation analysis to explore potential correlation relationships;

[0017] A time regression analysis module is used to identify unconventional data and / or find data of unconventional discrete points for statistical analysis through the data influence characteristics of information data in the association process;

[0018] A fault prediction module is used to predict the cause and location of a possible fault before it occurs by testing the state parameters of the equipment, collecting and analyzing fault symptom data, wherein the fault prediction is performed by data-driven, and / or, data is obtained by simulation and / or experiment, and a life model is established for an existing product for prediction;

[0019] The reliability analysis module is used to analyze and evaluate the reliability level of the entire system and each device, predict the remaining life of the equipment, and take targeted and differentiated protective measures based on the various risk factors that may be brought about by the external operating environment. The methods for analyzing and evaluating the reliability level of the entire system and each device include analytical method, Monte Carlo method and hybrid method;

[0020] The risk assessment module is used to evaluate the technical risks of the project using the triangular whitening weight function in the grey theory, and to whiten the grey system with insufficient clear or overall information so as to obtain the correct risk assessment results based on less information.

[0021] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method such as any one of the methods for early warning of failure of key equipment in a traction substation based on a prediction model.

[0022] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0023] The current passive operation and maintenance of railway power supply can be transformed into active operation and maintenance. By studying the fault warning and prediction methods of key equipment in traction substations, intelligent diagnosis and autonomous comprehensive status evaluation of key traction power supply and substation equipment can be realized, and maintenance decisions can be made based on the equipment status. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 A schematic diagram of the steps of a method for early warning of key equipment failure in a traction substation based on a prediction model provided in an embodiment of the present application;

[0026] Figure 2 A schematic diagram of a module of a traction substation key equipment failure warning system based on a prediction model provided in an embodiment of the present application;

[0027] Figure 3 An electronic device is provided in an embodiment of the present application.

[0028] Icon: 101 - memory; 102 - processor; 103 - communication interface. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0030] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0031] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0032] It should be noted that the term "comprises" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article, or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or device that includes the element.

[0033] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0034] Example 1

[0035] The important theoretical basis of machine learning algorithms is classical statistics, and its main research content is the asymptotic theory when the number of samples tends to infinity. However, in practical problems, the number of samples is often limited, and the intrinsic correlation of these samples is unknown. In general, the cost of obtaining additional samples is very high. Therefore, the machine learning algorithms designed based on the asymptotic theory when the number of samples tends to infinity are incapable of solving practical problems with a limited number of samples. In particular, when the number of samples is limited and the dimensionality of the sample space is extremely high (such as several thousand dimensions or hundreds of thousands of dimensions), machine learning methods based on classical statistics are even more helpless.

[0036] In the mid-1990s, Vapnik and the AT&T Bell Laboratory team he led proposed a support vector machine (SVM) learning algorithm based on statistical learning theory and kernel function mapping theory. This is a typical small sample machine learning algorithm. Once proposed, it has attracted widespread attention from researchers in the field of machine learning for its excellent theoretical performance and good generalization performance.

[0037] See also Figure 1 , Figure 1 A schematic diagram of a method for early warning of key equipment failure in a traction substation based on a prediction model is provided in an embodiment of the present application, as shown below:

[0038] Step S100, obtaining information data of key power supply and transformation equipment information, fault type, fault mode, fault cause, and fault influencing factors through information collection technology;

[0039] In some embodiments, the main statistics include the number of faults, the number of parameter anomalies, the number of system function reduction / non-working states, the failure rate, the first failure time, the failure interval time, the power interruption time, the repair rate, the failure repair time, etc.

[0040] Taking EMU as an example, with the help of information collection technology, the operation status, fault information and maintenance status of EMU throughout its life cycle can be recorded in a timely and complete manner. This information provides a basis for the diagnosis of EMU faults. Through statistical analysis of fault information, it helps to optimize the EMU maintenance process, the type and quantity of after-sales spare parts, and the normal operation of EMU equipment. Through macro-statistical analysis of fault information, staff can not only obtain the time, lines and equipment where faults occur frequently, but also obtain the main fault modes, causes and influencing factors of faults. In-depth statistical analysis of fault information can optimize the maintenance interval of equipment, adjust the maintenance content of each level of maintenance, and reasonably allocate maintenance resources.

[0041] In some implementations, key power supply and transformation equipment (traction transformers, high-voltage circuit breakers, gas-insulated metal-enclosed switchgear (GIS), lightning arresters, high-voltage cables, transformers, etc.) and fault categories can be clearly identified, and necessary collection data can be obtained through the operation and maintenance management unit or relevant departments.

[0042] Step S110, classifying the information data acquired through the information collection technology by means of a clustering method, and performing a correlation analysis to mine potential correlation relationships;

[0043] In some embodiments, the correlation analysis includes correlation analysis functions of various monitoring and detection data, fault data, maintenance data, and monitoring data in different time periods and sections, etc. For example, clustering methods are used to classify data and explore potential correlation relationships.

[0044] Step S120, identifying unconventional data and / or finding data of unconventional discrete points for statistical analysis through data influence features of information data in the association process;

[0045] In some embodiments, time regression analysis means: within a specified railway line, under different working conditions and environmental parameters, a time regression function is provided to determine the changing trend of data, allowing users to clearly understand the evolution of corresponding characterization parameters or equipment failure frequency based on the trend, thereby providing a basis for fault warning.

[0046] In some embodiments, historical data is obtained, and at the same time, parameters that affect the data are found, and professional data software such as SPSS is used to adopt association analysis methods, using the data influence characteristics of the association process of data parameters to identify unconventional data, or using formula calculations to find data of unconventional discrete points. When the amount of data obtained is insufficient, the SVM algorithm can be used to make up for the impact of insufficient data that cannot be processed by conventional machine learning; an improved gray correlation analysis model can also be constructed.

[0047] Fault warning is to automatically give warning information to key components based on the fault prediction model and warning threshold, and remind the on-duty personnel by pop-up text windows, sound and light alarms, etc. In the early stage of equipment failure or when the failure is in the latent period, the hidden faults are discovered in time, the future development trend of the failure is accurately predicted, and the failure is warned and eliminated in time before the consequences of the failure appear. It is necessary to investigate and understand the actual working method of fault warning, the current fault prediction model, and how to determine the warning threshold.

[0048] Step S130, by testing the state parameters of the equipment, collecting and analyzing the fault symptom data, and predicting the cause and location of the possible fault before the fault occurs, wherein the fault prediction is performed by data-driven, and / or, data is obtained by simulation and / or experiment, and a life model is established for the existing product for prediction;

[0049] In some implementations, different fault symptoms may be caused by different fault locations or causes, and there is a linear or nonlinear mapping relationship between the fault symptoms and the fault causes or locations. This mapping relationship is described using reasoning methods or applicable models, and by testing the state parameters of the equipment, collecting and analyzing the fault symptom data, the possible causes and locations of the fault are predicted before the fault occurs, thus forming a fault prediction and health management technology.

[0050] In some implementations, fault prediction can be achieved in two ways: one is to predict based on an existing product life model, and the other is to predict based on a model established based on historical data. The former can obtain data through simulation or experiment, while the latter can be achieved through data-driven, which has a wide range of applications, namely, the process of data collection, feature extraction, trend prediction, and fault identification:

[0051] In some implementations, it is difficult to obtain full life cycle characteristic data in actual use of the product. Therefore, if a relatively complete cycle model is required, there are two methods to obtain data: simulation and experiment. The cost of simulation is low, and multiple fault injection verifications can be achieved through simulation, thereby obtaining different response characteristic data of the product.

[0052] The limitation of simulation is that its research results largely rely on the accuracy of the model and the depth of understanding of the product failure mechanism. Currently, the most widely used application is to use reliability experimental technology to carry out accelerated life experiments on products with a certain sample size, monitor the product performance degradation characteristics of the entire experimental process in real time, and use similarity modeling-based methods to predict faults. Commonly used accelerated life experiment models include temperature acceleration model and temperature and humidity acceleration model.

[0053] In some implementations, data-driven prediction is to understand and master the operating status of the system and the law of state change, and to infer the change trend of its state according to certain prediction methods, estimate the propagation and development of faults and the performance degradation trend of the system, which is an important means of accident prevention, condition-based maintenance and health management. Among them, the data-driven fault prediction method mainly uses the historical working data, fault injection data and simulation experiment data of the equipment, and performs trend prediction through various data analysis and processing algorithms, which is currently a widely used prediction method.

[0054] Data-driven fault prediction is a process of data collection, feature extraction, trend prediction, and fault identification, which is mainly divided into the following three steps: prediction sample data preparation and prediction model training, trend prediction, and trend prediction.

[0055] Step S140, analyzing and evaluating the reliability level of the whole system and each device, predicting the remaining life of the device, and taking targeted and differentiated protective measures in combination with various risk factors that may be brought about by the external operating environment. The methods for analyzing and evaluating the reliability level of the whole system and each device include analytical method, Monte Carlo method and hybrid method;

[0056] In some implementations, from the perspective of long-term operation, the reliability level of the system as a whole and each device is analyzed and evaluated, the remaining life of the equipment is predicted, and differentiated protective measures are taken in a targeted manner in combination with various risk factors that may be brought about by the external operating environment. Probability theory and mathematical statistics are frequently used in reliability analysis, so the evaluation methods derived from them mainly include analytical methods, simulation methods, hybrid methods, etc.:

[0057] In some implementations, the analytical method mainly builds a reliability model based on the functional relationship between system components, integrates all fault states, and analyzes them to finally obtain relevant reliability indicators. Its advantages are clear concepts and high model accuracy. Its disadvantages are that when applied to larger systems, it is more complicated and difficult to implement because all fault states need to be listed and analyzed one by one. In practical applications, the original analytical method can be improved accordingly according to different actual conditions to achieve the purpose of reducing the workload. The new methods formed mainly include the minimum cut set method, the minimum path set method, the state space truncation method, the network equivalence method, etc.

[0058] In some embodiments, the simulation method can also be called the Monte Carlo method, the principle of which is to assign random operating states to each component of the system through random sampling, and then calculate the required reliability index using probability statistics and other methods. The advantage of this method is that it is easier to implement than the analytical method and can be applied to more complex systems. Its disadvantage is that the calculation convergence speed is slow and the error is large.

[0059] In some embodiments, the central idea of ​​the hybrid method is to first use the Monte Carlo method to simulate the random operating state of each component of the system, and then use the analytical method to calculate the average duration of each state, replacing the sampling of state time required originally. This approach can reduce the amount of calculation of the analytical method while reducing the iterative convergence time of the Monte Carlo method. ④ Artificial intelligence method.

[0060] In some embodiments, a study of failure data of a large number of different types of components shows that the component failure rate curve is bathtub-shaped, so an appropriate probability distribution (such as exponential distribution, normal distribution, and Weibull distribution) can be selected as a failure rate curve; based on the components or devices, one of the analytical methods, simulation methods, or hybrid methods is used to establish system-level reliability.

[0061] Step S150, using the triangular whitening weight function in the grey theory to evaluate the technical risk of the project, whitening the grey system with insufficient clear or overall information, so as to obtain the correct risk assessment result based on less information.

[0062] In some implementations, many new technologies and methods are used to improve the intelligence of fault diagnosis, remaining life prediction and maintenance decision-making. Since new technologies lack a large amount of historical data as the basis for evaluation, they are mostly based on human experience and understanding. Therefore, the evaluation of technical risks is somewhat "gray".

[0063] In some implementations, the evaluation content is clarified; the triangular whitening weight function in the grey theory is used to evaluate the technical risk of the project, in order to whiten the grey system which is not very clear and has insufficient overall information as much as possible, so as to obtain the correct evaluation result based on less information.

[0064] It also includes maintenance decision-making assistance: combining the aforementioned functions, comprehensively formulating reasonable maintenance strategies, determining the best maintenance cycle, the best time and method for fault repair, and the selection of various status thresholds.

[0065] In some implementations, based on the current health status of the system, starting from the reliability and maintenance cost of the system, an optimized maintenance decision reference is provided. With the lowest overall maintenance cost and the highest overall system reliability as the optimization goal, the optimal maintenance plan is generated by optimizing parameters such as the comprehensive maintenance cycle, maintenance method, staffing, and planned maintenance times.

[0066] Specifically, it is necessary to consider the various dynamics and randomness in maintenance activities on the basis of the previous research content, examine the multi-scale effects of maintenance activities in time (such as the urgency of accident repair, the real-time nature of condition maintenance, and the periodicity of regular maintenance, etc.) and space (such as component-level, equipment-level, subsystem-level, system-level maintenance, etc.), establish a multi-factor, multi-scale spatiotemporal joint decision-making model, and focus on the decision-making issues in maintenance time, maintenance cycle, maintenance degree and maintenance method; apply sensitivity analysis methods, nonlinear mathematical programming methods and multi-objective intelligent optimization algorithms to study the optimization and optimal combination of maintenance decision variables; analyze the various attributes of the traction power supply system in maintenance activities, reasonably allocate their weights, form a set of maintenance influencing factors, and establish a maintenance library in combination with spare parts, tools and types of work used in maintenance activities, and finally provide a complete set of customized maintenance solutions for each traction power supply system with different structures and performances. Through the auxiliary optimization decision-making part, the project team will provide general algorithm programs and typical case verification results.

[0067] Example 2

[0068] See also Figure 2 , Figure 2 A module schematic diagram of a traction substation key equipment failure warning system based on a prediction model provided in an embodiment of the present application is as follows:

[0069] The main statistical quantity statistical submodule is used to obtain information data on key power supply and substation equipment, fault categories, fault modes, fault causes, and fault influencing factors through information collection technology;

[0070] The correlation analysis module is used to classify the information data obtained through information collection technology through clustering methods, and to perform correlation analysis to explore potential correlation relationships;

[0071] A time regression analysis module is used to identify unconventional data and / or find data of unconventional discrete points for statistical analysis through the data influence characteristics of information data in the association process;

[0072] A fault prediction module is used to predict the cause and location of a possible fault before it occurs by testing the state parameters of the equipment, collecting and analyzing fault symptom data, wherein the fault prediction is performed by data-driven, and / or, data is obtained by simulation and / or experiment, and a life model is established for an existing product for prediction;

[0073] The reliability analysis module is used to analyze and evaluate the reliability level of the entire system and each device, predict the remaining life of the equipment, and take targeted and differentiated protective measures based on the various risk factors that may be brought about by the external operating environment. The methods for analyzing and evaluating the reliability level of the entire system and each device include analytical method, Monte Carlo method and hybrid method;

[0074] The risk assessment module is used to evaluate the technical risks of the project using the triangular whitening weight function in the grey theory, and to whiten the grey system with insufficient clear or overall information so as to obtain the correct risk assessment results based on less information.

[0075] like Figure 3 As shown, an embodiment of the present application provides an electronic device, which includes a memory 101 for storing one or more programs and a processor 102. When the one or more programs are executed by the processor 102, any method in the first aspect described above is implemented.

[0076] The memory 101, the processor 102 and the communication interface 103 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines. The memory 101 can be used to store software programs and modules, and the processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be used to communicate signaling or data with other node devices.

[0077] Among them, the memory 101 can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0078] The processor 102 may be an integrated circuit chip with signal processing capability. The processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0079] In the embodiments provided in the present application, it should be understood that the disclosed method and system can also be implemented in other ways. The method and system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the method and system, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0080] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0081] On the other hand, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by the processor 102, the method of any one of the first aspects described above is implemented. If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk.

[0082] To sum up, the embodiment of the present application provides a method and system for early warning of key equipment faults in traction substations based on a prediction model, which can transform the current "passive" operation and maintenance of railway power supply into "active" operation and maintenance. By studying the early warning and prediction method for key equipment faults in traction substations, intelligent diagnosis and autonomous comprehensive status evaluation of key equipment in traction power supply and substation can be achieved, and maintenance decisions can be made based on the equipment status.

[0083] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0084] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for early warning of key equipment failure in traction substation based on a prediction model, characterized in that: include: Through information collection technology, we can obtain information data on key power supply and substation equipment, fault types, fault modes, fault causes, and fault influencing factors; Use clustering methods to classify the information data obtained through information collection technology, conduct correlation analysis, and explore potential correlation relationships; Identify irregular data and / or find irregular discrete points for statistical analysis through the data impact characteristics of information data in the association process; Also includes: By testing the equipment's status parameters, collecting and analyzing fault symptom data, the cause and location of possible faults are predicted before the fault occurs, wherein the fault prediction is performed through data-driven, and / or, data is obtained through simulation and / or experiments, and a life model is established for existing products for prediction; The data-driven fault prediction further includes: Data-driven fault prediction is a process of fault identification through data collection, feature extraction, and trend prediction. That is, the prediction sample data is prepared through data collection, the prediction model is trained through feature extraction of the prediction sample data, and finally the fault trend is predicted. Also includes: Analyze and evaluate the reliability level of the entire system and each device, predict the remaining life of the equipment, and take targeted and differentiated protective measures based on the various risk factors that may be brought about by the external operating environment. The methods for analyzing and evaluating the reliability level of the entire system and each device include analytical method, Monte Carlo method and hybrid method; The analytical method includes: Build a reliability model based on the functional relationship between system components, integrate all fault conditions, analyze the fault conditions, and finally obtain reliability indicators; The Monte Carlo method includes: By random sampling, each component of the system is given a random operating state, and the required reliability index is calculated using probability statistics methods; The hybrid method includes: The Monte Carlo method is used to simulate the random operating states of each component of the system, and then the average duration of each state is calculated using the analytical method, replacing the sampling of state time required originally. This reduces the computational workload of the analytical method and the iterative convergence time of the Monte Carlo method. Also includes: The triangular whitening weight function in the grey theory is used to evaluate the technical risk of the project, and the grey system with insufficient clear or overall information is whitened so that the correct risk assessment result can be obtained based on less information.

2. A traction substation key equipment failure early warning system based on a prediction model, characterized in that: include: The main statistical quantity statistical submodule is used to obtain information data on key power supply and transformation equipment, fault categories, fault modes, fault causes, and fault influencing factors through information collection technology; A correlation analysis module is used to classify the information data obtained through the information collection technology through clustering methods, and to perform correlation analysis to explore potential correlation relationships; a time regression analysis module is used to identify unconventional data and / or find data of unconventional discrete points for statistical analysis through the data influence characteristics of the information data in the correlation process; a fault prediction module is used to predict the cause and location of possible faults before the fault occurs by testing the state parameters of the equipment, collecting and analyzing fault symptom data, wherein fault prediction is performed through data-driven, and / or, data is obtained through simulation and / or experiments, and a life model is established for existing products for prediction; The reliability analysis module is used to analyze and evaluate the reliability level of the entire system and each device, predict the remaining life of the equipment, and take targeted and differentiated protection measures in combination with the various risk factors that may be brought about by the external operating environment. Among them, the methods for analyzing and evaluating the reliability level of the entire system and each device include analytical method, Monte Carlo method and hybrid method; The risk assessment module is used to evaluate the technical risks of the project using the triangular whitening weight function in the grey theory, and whiten the grey system with unclear or insufficient overall information, so as to obtain correct risk assessment results based on less information.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method as claimed in claim 1 is implemented.