Power distribution equipment state visualization platform based on big data

The technical field of the visualization platform for the status of power distribution equipment based on big data

[0001] The present invention relates to the technical field of equipment management, and in particular to a big data-based visualization platform for the status of power distribution equipment.

Background technique

[0002] At present, major power grid companies mainly rely on the power production management system (PMS) for information management of power grid equipment. PMS can provide most of the information required for equipment management, including equipment account, defect, status evaluation, maintenance test, online monitoring and other data. However, from the perspective of strengthening the all-round management and control of the equipment status, the effective information provided by the PMS is still very limited, and the sensitivity to status changes and exposure to risks is far from enough, and the management efficiency and effectiveness are not high.

[0003] In recent years, equipment failures caused by quality problems have occurred from time to time, and extreme weather such as thunderstorms and line winds are frequently affected by global climate change, which has an increasingly prominent impact on the safe operation of power grid equipment. Therefore, there is an urgent need for a management platform that can combine information provided by real-time systems such as scheduling, weather, and video surveillance, and add advanced intelligent diagnosis and auxiliary analysis functions to better meet the work needs of the transportation inspection department.

Summary of the invention

[0004] The present invention aims to solve at least one of the above technical problems.

[0005] To this end, the purpose of the present invention is to propose a big data-based visualization platform for the status of power distribution equipment, which can meet the business needs of intensive development and lean management, and improve the ability of equipment multi-source information interaction and integration to achieve The panoramic real-time perception, multi-dimensional intelligent monitoring and control of power distribution equipment can prevent equipment operation risks in advance, improve the response speed to emergencies, and make the life cycle management of equipment transparent and efficient.

[0006] In order to achieve the above objective, the embodiment of the present invention proposes a big data-based visualization platform for the status of power distribution equipment, including: a data processing module for acquiring multi-platform data and processing the multi-platform data , And display the processed data; data analysis module, used for big data integration, storage, retrieval, and data mining analysis; evaluation module, used to generate big data-based power distribution equipment evaluation model, and according to the power distribution equipment The evaluation model evaluates power distribution equipment and generates corresponding processing strategies based on the evaluation results.

[0007] In addition, the big data-based power distribution equipment status visualization platform according to the above-mentioned embodiments of the present invention may also have the following additional technical features:

[0008] In some examples, the multi-platform data includes at least: production management system data, online monitoring system data, spatial geographic information system data, meteorological system data, and video monitoring platform data.

[0009] In some examples, the data processing module is used to preprocess and clean the acquired multi-platform data, including: according to the business system, type, structure, and size of the multi-platform data, marking a unified specification The mark is used to identify the source and type of the data. At the same time, combined with the preset data rule library, the corresponding rules and data are encapsulated according to the data mark. The encapsulated data is identifiable, controllable and has corresponding The cleaning rules can be sent to the data cleaning stage for cleaning work.

[0010] In some examples, the data analysis module includes a perception layer, a network layer, and an application layer, wherein the perception

The layer is used for data collection; the network layer is used for data transmission; the application layer further includes a service layer, a business layer, a presentation layer, and a tool set. The service layer is used to provide data mining and analysis capabilities, The business layer is used to realize the business requirements of specific products, the presentation layer is used to provide an interactive interface, and the tool set is used to provide installation and deployment tools, data mining tools, business modeling tools, and code generation tools.

[0011] In some examples, the perception layer, the network layer, and the application layer interact, and the interaction includes a message flow and a data flow, and the processing of the data flow is controlled through the message flow.

[0012] In some examples, the power distribution equipment evaluation model includes at least: transformer equipment fault prediction models, switch and combined electrical equipment status development trends and failure probability dynamic prediction models, and transmission line faults based on complex correlations Forecast model.

[0013] In some examples, the evaluation module is used to use a multi-factor state evaluation analysis algorithm, including: 1) Analyze decision-making problems to construct a system's proposition set, that is, the system's identification framework Q = {A1,A2 ,......,Ak}; 2) For the target information system, construct the evidence body Ei (i = 1,2,......,m) based on the recognition framework; 3) According to the collected data of each evidence body, a total amount of evidence Data, combined with the characteristics of each proposition set in the recognition framework, determine the basic credibility distribution of each evidence body mi(Aj) = ......, Κ, indicating the ability of different state information to respond to the device state; 4) According to the basic credibility Degree allocation mi (Aj), respectively calculate the reliability interval of each proposition in the recognition framework under the action of a single body of evidence [Beli, Pli]; 5) Use the DS synthesis rule to calculate the basic credibility allocation m( Aj) and the confidence interval [Bel, Pl]; 6) Construct the corresponding decision rule according to the specific problem; 7) Draw the decision conclusion according to the decision rule.

[0014] In some examples, the evaluation module evaluates the power distribution equipment, including: A) Scan the data one by one according to the relevant requirements in the power distribution equipment state evaluation guideline, corresponding to each state quantity threshold in the guideline, When any data exceeds the threshold defined in the guidelines, the data is marked as an abnormal value and separated from the original data; B) The data is transformed into a multivariate time series, and the mutual covariance function and mutual covariance function of each one-dimensional time series are calculated. Correlation function, so as to obtain the order and delay parameters of the transfer function numerator and denominator polynomial, then fit the transfer function model, and finally determine the interference time and the abnormal data generated according to the ACF test of the model residual sequence; C) Based on incremental recurrence Inferred least squares regression parameter estimation and generalized likelihood ratio change point detection, the incremental mechanism is used to determine the data sequence regression model parameters and segmentation points, the data trend characteristics are extracted in real time, and the trend change data is marked as abnormal data.

[0015] In some examples, the failure rate of the distribution line under different weather conditions is the number of failures when the unit of time is converted into a year, and the average of the failure rate when the unit is 1 calendar year can be expressed by for:

[0017] Wherein, N is the expected duration of normal weather, and S is the expected duration of severe weather. Indicates the expected value of component failure rate in normal weather,'is the expected value of component failure rate in severe weather;

[0018] The failure rate and cumulative probability distribution function of the transformer can be expressed as:

[0019] child (One ^ birth

[0020] ^_ hum) = l-e2

[0021] Among them, the model assumes that the parameters 8 and C remain unchanged at different temperatures, and a two-state weather model is used to describe the accidental failure mode failure rate of the transformer, and its expression is:

[0022]

4,.( k) =« .N+S Λ ----Ν (1-wide) w = Ο »r = 1

[0023] Among them, 3 is the statistical average of accidental failures of the transformer, N is the duration of normal weather, S is the duration of severe weather, F is the proportion of faults that occur in severe weather, and w is the current weather of the transformer. Conditions, normal weather w = 0, severe weather w = l.

[0024] In some examples, the evaluation module is also used to evaluate the importance of equipment based on equipment status and system risks, including: a) Using the PHM model to calculate the system based on the evaluation results of the big data status, operating information, and micro-meteorological data The component considers the real-time failure probability of big data; b) Use enumeration to select the system state, enumerate to the third-order failure, form the expected failure event, and calculate the probability of the failure event; c) Perform static safety analysis on the selected system state ,Using the optimal power flow to calculate whether the system state satisfies the sufficiency, if the load shedding is required, the system state is an emergency state, go to step d), if the load shedding is not required, the system state is the alert state or the healthy state, and the system is NT Check, if it meets the safety criteria, it is in a healthy state, return to step b), otherwise, it is in an alert state, go to step d); d) Calculate the emergency index or the number of warning rings in the system state, and use the risk tracking model to calculate the state in this state The contribution value of each faulty component; e) Return to step b) Until all the fault events in the expected fault set are traversed; f) Calculate the total emergency index and total alarm ring number of the system, and calculate the component emergency importance index and alert importance index, according to the important The degree index is sorted to determine the weak equipment of the system.

[0025] According to an embodiment of the present invention, a big data-based power distribution equipment status visualization platform uses multi-dimensional visualization display functions to maximize the reuse of established various power distribution equipment basic data and maintenance management resources, and use data integration And data mining technology to achieve comprehensive equipment analysis, carry out state evaluation, fault diagnosis, risk assessment, life prediction and maintenance decision-making work for transmission and transformation equipment, assist with technical supervision and intelligent report management, and provide a panoramic, real-time , Multi-dimensional, intelligent equipment management and control information platform, so that a large number of equipment with different conditions, new and old, are always under strict monitoring, prevent equipment operation risks in advance, improve the speed of response to emergencies, and meet intensive development , The business needs of lean management, improve the equipment's multi-source information interaction and integration capabilities, realize panoramic real-time perception, multi-dimensional intelligent monitoring and control of power distribution equipment, and make the equipment's life cycle management transparent and efficient.

[0026] Additional aspects and advantages of the present invention will be partly given in the following description, and part of them will become obvious from the following description, or be understood through the practice of the present invention.

Description of the drawings

[0027] The above and/or additional aspects and advantages of the present invention will become obvious and easy to understand from the description of the embodiments in conjunction with the following drawings, in which:

[0028] FIG. 1 is a structural block diagram of a big data-based power distribution equipment state visualization platform according to an embodiment of the present invention;

[0029] FIG. 2 is a schematic diagram of a data cleaning process according to a specific embodiment of the present invention;

[0030] FIG. 3 is a schematic diagram of a fault prediction process of transformer equipment according to a specific embodiment of the present invention;

[0031] FIG. 4 is a schematic diagram of a transmission line fault prediction process according to another specific embodiment of the present invention;

[0032] FIG. 5 is a schematic diagram of a state evaluation analysis algorithm integrating multiple factors according to a specific embodiment of the present invention;

[0033] FIG. 6 is a schematic diagram of a differential state evaluation process of transformer equipment according to a specific embodiment of the present invention;

[0034] FIG. 7 is a schematic diagram of a rapid assessment process of a device status according to a specific embodiment of the present invention.

Detailed ways

[0035] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals indicate the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary, and are only used to explain the present invention, but should not be understood as limiting the present invention.

[0036] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right" "," "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, only for ease of description The present invention and simplified description do not indicate or imply that the pointed device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance.

[0037] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installation", "connected", and "connected" should be understood in a broad sense, for example, it may be a fixed connection or It is a detachable connection or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication between two components. For those of ordinary skill in the art, the specific meanings of the above-mentioned terms in the present invention can be understood in specific situations.

[0038] The following describes a big data-based power distribution equipment status visualization platform according to an embodiment of the present invention with reference to the accompanying drawings.

[0039] FIG. 1 is a structural block diagram of a big data-based power distribution equipment state visualization platform according to an embodiment of the present invention. As shown in FIG. 1, the big data-based power distribution equipment status visualization platform 100 includes: a data processing module 110, a data analysis module 120 and an evaluation module 130.

[0040] The data processing module 110 is used to obtain multi-platform data, process the multi-platform data, and display the processed data. Among them, the multi-platform data includes at least production management system (PMS) data, online monitoring system data, spatial geographic information system (GIS, Geographic Information System) data, meteorological system data, and video monitoring platform data, for example.

[0041] In other words, the data processing module 110 can implement multi-platform data acquisition. For example, the state visualization platform of power distribution equipment needs to integrate multiple systems horizontally, and access its data through the data processing module 110. At this stage, for example, it includes production management system (PMS) data, online monitoring system data, and spatial geographic information system (GIS). ) Data, weather system data and video surveillance platform data, etc. The horizontal data integration of the power distribution equipment status visualization platform and these systems follows a unified interface specification, and the data interface method is given priority to using the Web Service method, and at the same time comprehensive processing is combined with the actual situation of the project.

[0042] In a specific example, such as shown in Table 1, shows some examples of data source systems and their access objects.

CN 108564254 Β

<td>Data source system</td><td>Access object</td>

<td>Production management system</td><td>Ledger data of power distribution equipment, including equipment nameplate, technical parameters, equipment change management information and other data: equipment inspection, maintenance and operation and maintenance information, mainly including defect records, hidden danger records, fault records, operation and maintenance records, operation and maintenance records, etc. Data, real-time voltage, real-time current, etc.</td>

<td>Online monitoring system</td><td>Transformer equipment status monitoring parameters, including oil temperature and oil pressure, oil chromatogram, partial discharge, ground current, bushing dielectric loss capacitance, OLTC status, vibration spectrum, etc.: GIS equipment status monitoring parameters, mainly including partial discharge, gas density and micro Water content, gas leakage, gas composition, circuit breaker operating mechanism current and stroke; overhead line monitoring parameters, mainly including icing, micro-weather, temperature (sag), wind deviation, galloping, vibration</td>

<td>[0044]</td><td></td><td>Movement, tower tilt, vibration, etc.; cable monitoring parameters, mainly temperature and partial discharge along the line</td>

[0045] Table 1

[0046] Further, the implementation of the interface includes, for example, the following:

[0047] Web Service service call interface: For the power distribution equipment status visualization platform, the status monitoring information that needs to be further processed, and the status detection has provided service interfaces (for the power distribution equipment status visualization platform, online monitoring of data that does not provide a service interface is required , Obtain the data in the status monitoring through service call, the principle is to fetch and use as needed, and the data of non-special needs are not stored in the database of the power distribution equipment status visualization platform.

[0048] Page embedded integration interface: For the power distribution equipment status visualization platform, no further processing of the status monitoring information is required, and the status detection has provided the corresponding module page, then the corresponding function page is called through the URL.

[0049] Structured data acquisition interface: For conventional relational database data, programming interfaces such as JDBC/ODBC are used to directly acquire database data. For data with higher security and more privacy, the business system provides an interface for data acquisition/conversion The device is called to obtain or is actively pushed by the business system, and the relevant data is sent to the enterprise message bus, and the data acquisition/conversion device monitors the message bus to obtain the data.

[0050] Unstructured data acquisition interface: For unstructured data such as documents, audio, surveillance video, and pictures obtained from inspections, the data acquisition/conversion device directly reads and calls related files through a common file transfer protocol, and performs follow-up Related cleaning, conversion and other processing work.

[0051] Grid spatial data acquisition interface: grid spatial data is relatively complex, including structured data such as coordinate axes, longitude and latitude, and unstructured data such as images and texts. The data acquisition/conversion device uses a structured data interface and an unstructured data interface to acquire data from the system according to different data types. For the data pulled from the business system by the data acquisition/conversion device calling the programming interface or the system interface, it is necessary to configure the relevant strategy in the device, define the relevant interface, cycle, calling frequency, calling object and other related parameters, and data acquisition/ The conversion device will automatically perform related tasks and pull data from the business system. Data acquisition is mainly divided into information intranet data acquisition and information extranet data acquisition. The data acquisition/conversion device is deployed on the information intranet, and the data acquisition of the business system on the information extranet needs to pass the security

Full isolation device, based on safe transmission channel to obtain. The overall idea of data acquisition is based on the cross-platform programming interface enterprise service bus, which uses data interfaces, data center sharing, and secure file transmission under network isolation to solve cross-platform database access, cross-platform big data file high-speed concurrent reading, and cross-platform Key technologies such as platform data secure transmission and synchronization. [0052] On the other hand, the power distribution equipment status visualization platform needs to interact with numerous information systems, and needs to be connected in a loosely coupled manner. For example, a service-oriented architecture (Service-Oriented Architecture, SOA) can be used. SOA is a component model that connects different functional units (called services) of an application through well-defined interfaces and contracts between these services. The interface is defined in a neutral way, and it should be independent of the hardware platform, operating system, and programming language that implement the service. This enables services built in various such systems to interact in a unified and universal way.

[0053] In an embodiment of the present invention, the data processing module 110 is used to preprocess and clean the acquired multi-platform data. This is due to the fact that the acquired data has many sources of information, different structures, and various attributes. Therefore, it is necessary to preprocess and clean the data during the data quality management process. Specifically: the preprocessing is mainly based on the business system, type, structure, size, etc. of the data, marking a unified standard mark to identify the source, type and other attributes of the data. At the same time, combined with the preset data rule library, the corresponding rules and data are encapsulated according to the data tags. The encapsulated data is identifiable, controllable and has corresponding cleaning rules, and can be sent to the data cleaning stage for cleaning work. For example, as shown in Figure 2. [0054] The data stored in the equipment state assessment data source system usually maintains the original data characteristics, and the data quality is improved through data conversion technology, thereby improving the accuracy and performance of data mining or data flow mining. Data conversion mainly uses operations such as data generalization, data standardization, and data attribute construction to further improve data quality, and improve the accuracy and performance of data mining or data stream mining. Data generalization uses concept layering, replacing low-level "raw" data with high-level concepts, and generalizing the original data in the database into conceptual-level, aggregated, and statistically significant metadata that users are interested in.

[0055] The data analysis module 120 is used to perform big data integration, storage, retrieval, and data mining analysis.

[0056] In order to facilitate understanding, first describe the status quo of big data technology: With the development of smart grids, the establishment of various information platforms, the construction of smart substations, and the gradual application of smart power distribution equipment, a large amount of information from different sources is networked Integration and sharing is an inevitable trend in the development of equipment status evaluation, and promotes the development of power distribution equipment status evaluation and prediction to the direction of information integration and comprehensive analysis based on panoramic status. However, there are many factors that affect the operating status of power distribution equipment. The explosive growth of status monitoring data (such as partial discharge, vibration, images, video, etc.) plus the power grid operation and meteorological environment closely related to the status of the equipment has a huge amount of data. With rapid growth, it is difficult to establish a complete and accurate equipment state assessment mechanism model and causality model to analyze these data, which poses new challenges for improving the operation and maintenance management of power grid equipment. The lag in data mining and analysis technology for structural data has become a bottleneck in controlling the status of power distribution equipment and ensuring the safety of the power grid. Under this background, it is necessary to make full, reasonable and effective use of the data in the various information systems that have been built, to organically integrate a large number of scattered power distribution equipment status, operation and environmental weather and other multi-source information, and use advanced big data The processing technology realizes a differentiated, diversified, and complex all-round analysis, discovers from a large amount of data valuable rules for equipment state evaluation, timely captures the precursor information of early equipment failures, predicts the probability of failure, and provides an overview of the equipment state. Refined evaluation and prediction provide brand-new solution ideas and technical means, and ultimately effectively improve the accuracy of power distribution equipment evaluation, which is conducive to timely detection, rapid diagnosis and elimination of hidden troubles, improving equipment utilization, and ensuring safe and reliable operation of equipment and power grids .

[0057] Based on this, in an embodiment of the present invention, the data analysis module 120 includes, for example, a perception layer, a network layer, and an application layer. The perception layer is used for data collection; the network layer is used for data transmission; the application layer further includes a service layer, a business layer, a presentation layer, and a tool set. The service layer is used to provide data mining and analysis capabilities, and the business layer is used to implement With

To meet the business requirements of products, the presentation layer is used to provide interactive interfaces, such as mobile APP applications and Web browser applications, and the tool set is used to provide installation and deployment tools, data mining tools, business modeling tools, and code generation tools.

[0058] Specifically, the perception layer, network layer, and application layer all have the functions of collection, storage, analysis, and transmission, but each layer has a different focus. The collection of itself is used to optimize the system. The first layer has the ability to access third-party systems, and the storage ensures that any node failure data will not be lost, and analyzes the point-to-surface analysis from different perspectives of the perception layer, network layer, and application layer.

[0059] In an embodiment of the present invention, the perception layer, the network layer, and the application layer interact, and the interaction includes a message flow and a data flow, and the processing of the data flow is controlled through the message flow.

[0060] Further, the application layer, for example, further includes a storage layer. The storage layer is used for data storage. For example, Redis is responsible for real-time data storage, and event-driven data can be persisted to historical databases HBASE, Oracle/MySql, SqlLite. HBASE is used for large-scale data, and Oracle/MySql is used for medium-scale data. , SqlLite is used for small-scale data, you can use a unified interface to access Redis, HBASE, Oracle, MySql, SqlLite.

[0061] In a specific example, the service layer YARN is the resource scheduling manager of Hadoop2, on which Spark and MapReduce are provided. MapReduce provides offline parallel computing capabilities for big data, and Spark's Streaming provides online big data. Streaming computing capabilities, Sparks Shark provides Sql interactive computing capabilities, Sparks GraphX provides graph computing capabilities, and Mahout and MLlib provide data mining and machine learning functions. The embodiment of the present invention uses data mining tools in the tool set (similar to In PRiSM) for modeling, and then inject the model into the Internet of Things platform, which can be driven by the model engine to conduct real-time data mining and analysis.

[0062] On the basis of the system platform, the presentation layer uses WebSocket, Node.js, JQuery, and HTML5 to uniformly implement mobile and Web interactive interfaces.

[0063] The interaction between the storage layer, the service layer, and the presentation layer is event-driven, which ensures that the collected data can be displayed to the user interface in real time.

[0064] In a specific example, the mining algorithm and coupling analysis method used by the data analysis module are described as follows: The heterogeneous big data storage of the system adopts HADOOP2.0 distributed file storage, distributed NoSQL column database, and scalable data. Warehouse and other technologies provide big data application support and can meet the relevant technical requirements for big data management platforms. Under the Hadoop framework, big data management platform software implements a distributed, malicious, multi-dimensional data storage system. It is mainly composed of the following parts: distributed collaborative work system, distributed file system, distributed database, distributed data warehouse, unstructured data preprocessing and multi-level comprehensive indexing. Distributed file system: The system uses a distributed file system based on HDFS and HBASE. For massive unstructured small files and complex and changeable structured data, use the key value storage of HBASE. For a large single file, it can be directly stored in the HDFS file system. Distributed database: Build a distributed NoSQL database with one write and multiple reads and multiple writes and multiple reads in the system. Through the master-slave replication technology based on acceleration components, the data consistency between multiple databases is ensured, the disaster recovery function is realized, and the pressure when reading data is shared.

[0065] The comprehensive analysis system needs to perform complex data query, but HADOOP's simple key-value based indexing technology is difficult to support multi-dimensional data query. Therefore, it is necessary to establish a multi-level comprehensive index to improve the retrieval performance of the evaluation data of similar features. The specific primary index adopts a multi-dimensional R-tree structure to achieve key data feature dimensions. The relative density of clustered objects in the same cluster is close to each other, and the relative density of objects in different clusters is lower, so as to achieve the purpose of clustering objects. This can overcome the locality brought by the existence of a simple global distance standard as the basis of detection in most clustering or proximity algorithms: it is related to the distribution of neighbors in a certain range. By overcoming certain limitations brought about by the global distance threshold, based on the adjacent density algorithm, an adjustable and scalable effective clustering method will be formed, which can better support the transmission line and other power distribution equipment with super-grid

Data characteristics.

[0066] The evaluation module 130 is used to generate an evaluation model of power distribution equipment based on big data, evaluate the power distribution equipment according to the evaluation model of the power distribution equipment, and generate a corresponding processing strategy according to the evaluation result.

[0067] In one embodiment of the present invention, the power distribution equipment evaluation model includes at least: transformer equipment fault prediction model, switch and combined electrical equipment status development trend and failure probability dynamic prediction model, and power transmission based on complex association relationships. Line fault prediction model. Considering the differences of different equipment structures and fault types, the fault prediction methods of transformers, GIS/circuit breakers and transmission lines are elaborated in depth.

[0068] Transformer equipment fault prediction model: using traditional online monitoring, operating conditions, test and maintenance records, historical working conditions, defect records in the big data information system, combined with laboratory-related aging, defect pattern recognition and other tests, as well as new family differences , Relevant analogy tests, etc. to build a characteristic parameter data platform for fault diagnosis and dynamic prediction of transformer equipment; Analyze the typical defects of transformer equipment statistically, and use the methods of fitting and analogy in big data information to study the lack of some characteristic data. Manual completion method of data. Use deep learning (classification), cluster analysis and other technologies to analyze a large number of sample data, use data association algorithms to mine the change law and weight combination of typical defects and failure mode associated status information of power distribution equipment, and analyze the types, locations, and combinations of defects. The relationship between the severity and the related state, combined with the influence of bad working conditions, power grid operating state, and family defects on equipment state changes, builds a multi-dimensional equipment fault diagnosis and prediction model based on big data samples, as shown in Figure 3.

[0069] The development trend and failure probability dynamic prediction model of switch and combined electrical equipment status: First, statistical analysis of major defects or failure history data, laboratory simulation defect data. Use association rule mining, multi-correspondence analysis, principal component analysis and other related relationship recognition technologies to distinguish fault types and find the effective data combination that plays a leading role. For GIS and circuit breakers, the known effective that directly affects the evaluation results The data includes: closing resistance, such as SF6 humidity, SF6 gas pressure, partial discharge, vibration, etc., combined with equipment conditions to carry out related tests, and collect new data, establish the effective data multiple logic model and correlation matrix of GIS typical failure modes. Mining association rules for big data information that integrates power grid information, equipment status information, and natural environment information, mining typical GIS defects and failure modes of effective data changes, and analyzing the relationship between the type, location, and severity of defects and effective data , Using time series model, gray model, support vector machine, regression model and other methods to calculate the relationship between data that is closely related to valid data (such as bad working conditions, power grid operating status, family defects and equipment status changes, etc.) and valid data index. It is used to dynamically adjust the weight of effective data (closing resistance, partial discharge, etc.), and build a multi-dimensional equipment dynamic fault diagnosis model based on big data samples. Some time series are a cluster of time variables that depend on time. Although the individual sequence values constituting the time series are uncertain, However, the changes in the entire sequence do have certain regularities, which can be approximated by corresponding mathematical models. Combining the multiple logic model of the equipment's effective data and the state evolution history data, based on the technology of multiple time series to study the related evolution law and state distribution change of the fault characteristic information of GIS switchgear, combined with the state confirmation and diagnosis analysis results, it is proposed based on the ARMA (Auto-Auto- Regressive and Moving Average Model (Autoregressive and Moving Average Model) model of regression algorithm GIS switch equipment status development trend and failure probability dynamic prediction method.

[0070] Transmission line fault prediction model based on complex association relationship: according to the time and location information of the transmission line fault, the segmented mapping of the status data is realized; the support degree of all attributes after division is further calculated, and the cause of the transmission line fault is analyzed Correlation with changes in other state parameters, such as the relationship between transmission lines and icing, wind deviation, thunderstorms, pollution flashovers, etc., analyze the objective law of fault development, obtain frequent itemsets, and extract faults from frequent itemsets Development of association rules; using historical state evolution data, combined with multi-dimensional association rule analysis of equipment failure modes, based on the multivariate time series method to study the association evolution law and state distribution changes of the transmission line fault characteristic information, and finally establish the transmission line fault

Probabilistic dynamic prediction model, such as shown in Figure 4.

[0071] As a specific example, the following takes transformer equipment, switches and GIS equipment, and power distribution line (cable) equipment as examples to illustrate their specific technical routes. According to the cascading relationship and connotation mechanism of equipment status information, analyze the relationship between the global full data of the equipment and the equipment status, determine the status evaluation index system of the equipment, and propose the relevant criteria and the equipment status comprehensive evaluation model for the relevant characteristic parameters to reflect the equipment status. The online self-learning method using big data samples analyzes the individualized laws of equipment status changes from different manufacturers, different equipment types, different voltage levels, different operating years, different operating environments, and different operating seasons, and provides evaluation model parameters accordingly. , Judgment threshold personalized adjustment method, statistical analysis of equipment individual attribute information to obtain state evaluation correction index, establish a differentiated evaluation model of equipment state evaluation. The following takes transformer equipment, switches and GIS equipment, and transmission line (cable) equipment as examples to illustrate their specific technical routes. Through the combination of DS evidence reasoning theory and rules, the basic probability assignment function, reliability function and likelihood function based on the joint action of multiple evidence bodies are obtained, as shown in Figure 5, forming a multi-factor state evaluation analysis algorithm.

[0072] In other words, in an embodiment of this aspect, the evaluation module 130 is configured to adopt a multi-factor state evaluation analysis algorithm, as shown in FIG. 5, which specifically includes:

[0073] 1) Analyze the decision-making problem and construct a systematic proposition set, such as the GIS ontology evaluation module, that is, the system's identification framework

Ω = {A1,A2,......, Ak};

[0074] 2) For the target information system, construct an evidence body Ei (i = 1, 2, ..., m) based on the recognition framework, and specific detection methods, such as partial discharge, SF6 humidity, etc.;

[0075] 3) According to the collected data of each evidence body-global full data, combined with the characteristics of each proposition set in the recognition framework, determine the basic credibility distribution mi (Aj) of each evidence body, j = 1, 2,......, K, that is, the ability to respond to the status of the equipment with different status information;

[0076] 4) According to the basic credibility allocation mi (Aj), respectively calculate the reliability interval of each proposition in the recognition framework under the action of a single body of evidence [Beli, Pli];

[0077] 5) Calculate the basic credibility distribution m (Aj) and the credibility interval [Bel, Pl] under the combined action of all the evidence bodies by using DS synthesis rules;

[0078] 6) Construct corresponding decision rules according to specific problems;

[0079] 7) Draw a decision conclusion according to the decision rule.

[0080] Further, in an embodiment of the present invention, in conjunction with FIG. 6, the differential state evaluation process of transformer equipment is described as follows: first analyze and summarize the relevant parameters of the transformer state evaluation, determine the data type of the relevant parameters, and compare Data statistical analysis, classification and recognition methods, and proposed feature extraction methods for unstructured data such as images, videos, texts, etc., and then use principal component analysis, association analysis and other big data core mining and analysis methods to determine the status of the transformer The characteristic parameter and the coupling relationship between it and the equipment state improve the existing characteristic parameter set. Finally, for the specified parameters or parameter sets, use methods such as multivariate statistical analysis and multi-dimensional correlation analysis to determine the evaluation criteria and evaluation models for transformer status evaluation, and establish a complete evaluation index system; through regular or irregular data analysis, the realization Dynamic maintenance of the index system; statistical analysis of the error of the transformer state judgment under the conditions of equipment attributes, special working conditions, different structures, etc., determine the correction index under the corresponding conditions, use artificial neural networks, fuzzy clustering and other methods to establish the individualized state of the transformer The evaluation model realizes the differentiated evaluation of the status of transformer equipment.

[0081] The process of GIS/circuit breaker equipment differentiation status evaluation is described as follows: extract relevant parameters of the operating conditions of power distribution equipment from the big data comprehensive analysis platform, comprehensive meteorological environment information, operating conditions information, online monitoring information, and pre-tests Information such as regular inspection information, manual inspections, and equipment individualized data are analyzed using a systematic hierarchical clustering method to analyze the relationship between the above parameters

Establish the evaluation model of key performance such as partial discharge of GIS/circuit breaker and breaking short-circuit current based on the dependence relationship between the parameters and the state of GIS/circuit breaker. Use sequence rule mining method to analyze the impact of individual equipment differences (operating years, manufacturer models, operating conditions) and GIS/circuit breaker key performance degradation, expand the influencing factors of the evaluation model, and form a GIS/circuit breaker based on data state dependence Personalized and differentiated evaluation methods for key performance.

[0082] The differentiated state evaluation process of distribution lines is described as follows: extract relevant parameters from the big data platform and the operating conditions of transmission lines and cables, including meteorological environment, operating conditions, online monitoring, manual inspections, and pre-test and regular inspections. And other data, using the systematic hierarchical clustering method to analyze the dependence between the above parameters and between the parameters and the status of the distribution line, and establish the key performance of the distribution line (cable) such as ice coating, sag, insulator contamination, lightning protection level, etc. Evaluation model. Use sequence rule mining method to analyze the influence of individual equipment differences (operating years, manufacturer models, operating conditions) and key performance degradation of distribution lines (cables), expand the influencing factors of the evaluation model, and form power distribution based on data state dependence Individualized and differentiated evaluation methods for key performance of lines.

[0083] The following describes the rapid assessment method of the device status with reference to FIG. 7. Specifically, with the support of the big data hardware platform, using predictive models, isolated point analysis, clustering and partitioning, etc., a method for rapid detection and early warning of abnormal conditions based on real-time data stream mining technology of state information is proposed to achieve rapid abnormal conditions. Detection and early warning to improve the timeliness of evaluation. The state information data stream is an ordered time sequence composed of a large number of continuously arriving, potentially infinitely long, and constantly changing multi-source state information data. With the improvement and perfection of the power distribution equipment status monitoring system and production management system, as well as the real-time integration of power grid information and environmental weather information, the status information related data of power distribution equipment presents the characteristics of a large number of data streams and continuity, which can be quickly mined and inspected. The abnormal state in the output data stream can provide early warning, state evaluation and decision support for power distribution equipment. The rapid mining and early warning research of the abnormal value of the real-time data stream of the status information requires real-time mining of the abnormal status of a large number of structured data streams of the equipment, and on the other hand, the image, video, vibration (waveform, fingerprint), partial discharge (waveform, Quick extraction of unstructured data feature values such as map) and test reports.

[0084] Specifically, with reference to FIG. 7, first, a summary operation is performed on the device status information data stream, that is, a fixed-length window is set, and the window contains all data in a limited collection period. There are three outlier detection methods for the aggregated data, which are the threshold value, trend analysis method and time series transfer function model in the comparative state evaluation guidelines. These three methods can detect three types of abnormal values, including abnormal values that exceed the state threshold, horizontal migration abnormal values caused by external interference, and trend change abnormal values caused by potential faults. The specific steps include the following:

[0085] A) According to the relevant requirements in the power distribution equipment state evaluation guidelines, scan the data one by one corresponding to each state quantity threshold in the guidelines, and when any one of the data exceeds the threshold defined in the guidelines, the data is marked as abnormal Value, separated from the original data.

[0086] B) Transform the data into a multivariate time series, calculate the cross-covariance function and cross-correlation function of each one-dimensional time series, so as to obtain the order and delay parameters of the transfer function numerator, denominator polynomial, and then fit the transfer function Model, finally according to the ACF test of the model residual sequence to determine the interference time and the abnormal data generated. Specifically, at a certain moment in the operation of power distribution equipment, the system may be affected by external interference and affect the distribution of state data (for example, when the transformer suffers a short-circuit impact, the oil temperature will rise for a short time), in this case , The data will have a certain initial migration effect when the interference occurs at time T, and then permanent horizontal migration or temporary horizontal migration will occur due to the interference reason and the difference in state quantity attributes. Such outliers can be quickly detected through the transfer function model of the time series, that is, the data is first transformed into a multivariate time series, and the cross-covariance function and cross-correlation function of each one-dimensional time series are calculated, so as to obtain the transfer function numerator, The order and delay parameters of the denominator polynomial are then fitted to the transfer function model, and finally the interference time and the abnormal data generated are determined according to the ACF test of the model residual sequence.

[0087] C) Based on incremental recursive least squares regression parameter estimation and generalized likelihood ratio change point detection, the incremental mechanism is used to determine the data sequence regression model parameters and segmentation points, and the data trend characteristics are extracted in real time to change the trend The data is marked as abnormal data. This is because the insulation aging and mechanical defects may occur during the operation of the power distribution equipment, and the state quantity data may have a trend change (such as the accelerated deterioration of the oil-paper insulation of the transformer will cause the oil dielectric loss and the increase of C0 and C02 gas in the oil. The trend is strengthened), so the separation of abnormal values of such trend changes is of great significance for detecting potential failures of power distribution equipment. In the embodiment of the present invention, the detection method of this type of outlier is based on incremental recursive least squares regression parameter estimation and generalized likelihood ratio change point detection. The algorithm uses an incremental mechanism to determine the data sequence regression model parameters and Divide points, extract data trend characteristics in real time, and mark data with trend changes as abnormal data.

[0088] Further, in order to establish a more accurate model of the outage probability of power distribution equipment, based on historical accident records of electric power companies and equipment real-time status monitoring information, reveal the internal relationship between the internal state of the equipment, the evolution of the external environment and the forced outage of the equipment, Establish an "accident learning-event-driven" spatio-temporal state model to reinforce the learning system. Uncertainty theory, such as credibility theory, cloud model and other methods are used to give a switchable time-varying equipment outage model in the absence of collected data. Establish a parameter learning library to enable the equipment shutdown model to have adaptive feedback correction and safety check functions. The forced outage rate of equipment is mainly affected by time and space factors. The time factor is mainly reflected in the aging of the equipment, and the space factor is mainly reflected in the different locations of the equipment in the power system and the different meteorological environment around it. For modeling of transformer failure rate based on time-space state analysis, the model should have strong generalization ability and universality. The failure rate model of power distribution equipment that considers big data can not only characterize the general aging of the equipment, but also consider the impact of big data and multiple factors on the equipment, and can quantify the impact of internal and external covariates on the failure rate, such as certain equipment Detect information, the external environment of equipment operation, meteorological conditions and system conditions. The model also needs to consider the random process of state transition. Considering the random process of specific equipment will make the model more accurate and specialized, and closer to reality. In addition to general models, different devices also have many features under different conditions. Body model. The main reason for the aging failure of transmission lines is the loss of wire tensile strength, which is a gradual accumulation and irreversible process. Theoretical analysis and experimental results show that annealing of high-temperature conductors is the main reason for the loss of tensile strength of wires. The temperature of overhead transmission wires mainly depends on the wire current, ambient temperature, wind speed, wind direction, and solar heat.

[0089] In an embodiment of the present invention, the failure rate of the distribution line under different weather conditions is the number of failures when the unit of time is converted into a year, and the average value of the failure rate when a unit of 1 calendar year is 2 It can be expressed as:

[0090]

[0091] where N is the expected duration of normal weather, S is the expected duration of severe weather, and humans indicate that the expected value of the component failure rate in normal weather is the expected value of the component failure rate in severe weather.

[0092] Most of the transformers used in the power grid are oil-immersed transformers. The main reason for the aging failure of the transformer is the loss of the mechanical strength of the insulating paper, which is a gradual accumulation and irreversible process. The insulation failure of a transformer is related to its operating temperature. It is generally considered that the hot spot temperature of the transformer is the highest temperature suffered by the transformer insulation system, and the hot spot is near the top of the transformer's high-voltage or low-voltage winding. The aging process of transformers is often described by the Weibull distribution, which is the long-term failure Arrhenius-Weibull model. Therefore, the transformer's failure rate and cumulative probability distribution function can be expressed as:

[0093] Er Yi Shao)=-(-two¹

[0094] -15

CN 108564254 Β

[0095] It should be noted that the model assumes that parameters 8 and C remain unchanged at different temperatures. With enough samples, these parameters can be estimated by the least square method or the maximum likelihood method. The two-state weather model is used to describe the accidental failure mode failure rate of the transformer, and the two-state weather model is used to describe the accidental failure mode failure rate of the transformer. The expression is:

[0096] 4, (Sichuan) =,

2=2(1-production)Wow=0

N ^Λ W+5 λ-----corpse "=]

S

[0097] Among them, 2 is the statistical average of accidental failures of the transformer, N is the duration of normal weather, S is the duration of severe weather, F is the proportion of faults that occur in severe weather, and w is the current weather of the transformer. Conditions, normal weather w = 0, severe weather w = l.

[0098] Further, the evaluation module 130 is also used to evaluate the importance of equipment according to the equipment status and system risk. The reliability of the operation of power distribution equipment is directly related to the safety and stability of the power system. With the continuous expansion of power scale and the introduction of the market economy, power companies have put forward higher requirements for the safe and economic operation of power distribution equipment; the completeness of maintenance plans and the formulation of maintenance strategies directly determine the cost of power equipment during the use phase As well as the service life; therefore, the risk analysis of power equipment failure and the evaluation of the importance of the system risk are not only conducive to formulating an appropriate maintenance plan, improving the reliability of the system operation, but also can better avoid the maintenance insufficiency and the maintenance of the traditional preventive maintenance program. Overmaintenance and other problems, reduce maintenance costs and operating costs, and effectively improve the reliability and economy of power equipment operation.

[0099] In an embodiment of the present invention, performing equipment importance assessment according to equipment status and system risk specifically includes the following steps:

[0100] a) Using the PHM model to calculate the real-time failure probability of the system components considering the big data according to the evaluation results of the big data status, operating information, and micro-meteorological data;

[0101] b) Use the enumeration method to select the system state, enumerate to the third-order fault, form an expected fault event, and calculate the probability of the occurrence of the fault event;

[0102] c) Perform a static safety analysis on the selected system state, and use the optimal power flow to calculate whether the system state meets the sufficiency. If load shedding is required, then the system state is an emergency state, go to step d), if load shedding is not required, then If the system status is alert or healthy, perform N-1 verification on the system, if it meets the safety criteria, it is healthy, and return to step b), otherwise it is alert, and go to step d);

[0103] d) Calculate the emergency index or the number of alarm rings in the system state, and use the risk tracking model to calculate the contribution value of each faulty component in the state;

[0104] e) Return to step b) until all failure events in the expected failure set are traversed;

[0105] f) Calculate the system's total emergency index and total number of warning rings, and calculate the component's emergency importance index and alert importance index, and sort according to the importance index to determine the system's weak equipment.

[0106] In summary, according to the big data-based power distribution equipment status visualization platform of the embodiment of the present invention, relatively mature software engineering development technology is used in the system development process to perform demand analysis on system functions and construct a network application system. At the same time, a safe operation mechanism is established to ensure the continuous and reliable operation of the system, and a reasonable work process is formulated according to relevant management specifications, which mainly involves power supply analysis of distribution network equipment, state maintenance, life cycle management of distribution network equipment, and power consumption information collection business modules . The platform uses information technology to establish a stable and efficient operation and maintenance and data verification system, promote the practicality and deepening of the application of the system, and ensure the sustainable development of each system; realize the power distribution of the power distribution equipment within the jurisdiction of the power supply company

Analysis and information early warning requirements. This big data-based visualization platform for the status of power distribution equipment is mainly improved from the following aspects:

[0107] (1) More powerful equipment status monitoring function.

[0108] Fully meet the future PMS2.0 data structure requirements of State Grid Corporation, and centrally display various equipment status information. Use visualization technology to convert numbers and words into graphics, and use dynamic curves, statistical graphs, lists and other friendly methods to show individual differences and development trends of equipment status. Develop the "equipment I care about" module to facilitate the operation and inspection personnel of different levels and different units to dynamically grasp the status of the equipment in real time according to their needs, and realize visual monitoring.

[0109] Based on big data analysis, panoramic display of equipment status function.

[0110] The platform will access PMS2.0 system data and power distribution automation system data, mine core state parameters closely related to equipment load capacity, and establish equipment load capacity under different service environments, different equipment aging states, and different equipment defects Short-term, mid-term, and long-term multi-scale dynamic evaluation and prediction models, study the overload operation of power distribution equipment and the relationship with the health and life of the equipment, and propose a real-time dynamic capacity expansion control strategy and a dynamic check method for the equipment load safety margin . Multi-dimensional display of equipment status information improves the refined management of distribution network equipment.

[0111] More abundant equipment fault diagnosis functions.

[0112] Highly integrate information from different application systems, equipped with advanced diagnostic modules such as status detection, maintenance work, auxiliary work, power supply quality analysis, risk warning, etc., conduct in-depth mining and multi-angle analysis of massive data, combined with the status of power distribution equipment Monitoring information and equipment maintenance test information in the PMS, online diagnosis of the status of power distribution equipment, and use of typical faults, defect case libraries and equipment standard libraries to improve the real-time and accuracy of status diagnosis, and support managers to make quick and accurate decisions.

[0113] More comprehensive operational risk early warning function.

[0114] Perform continuous real-time scanning of the weak points of the power grid according to established rules, automatically find power distribution equipment with serious defects, state degradation and other fault risks, and display through risk information summary, inspection defects, inspection hidden dangers, monitoring abnormalities and other modules, and Automatically issue warning information.

[0115] On the one hand, the platform uses a power distribution equipment evaluation method based on power big data analysis, extracts relevant parameters of substation equipment operating conditions from the big data comprehensive analysis platform, and uses sequence rule mining methods to analyze individual differences in equipment and equipment The impact of the deterioration of key performances forms a differentiated evaluation method of equipment key performance based on the dependence of data status.

[0116] On the other hand, the platform uses the minimum covariance determinant MCD robust multivariate detection method for the abnormal values of the three types of characteristic gases of oil chromatography H2, CO and total hydrocarbons. Using iteration and Mahalanobis distance ideas to construct a robust covariance estimator to detect outliers, it strengthens the statistical law of oil chromatographic data; through the tracking and evaluation of the outlier interval, the changes in the operating state of the transformer are more clearly reflected.

[0117] Further, the platform adopts the transformer partial discharge pattern recognition method based on the boosted dual-tree complex wavelet transform and the BP neural network, and designs the transformer partial discharge detection system based on the ultra-high frequency method, and applies the boosted dual-tree complex wavelet transform De-noise the collected partial discharge envelope signals, and use BP neural network to perform pattern recognition on the type of transformer discharge, effectively eliminating the interference of spatial electromagnetic waves and hardware circuit noise in the original signal, and improving the accuracy of partial discharge type recognition. [0118] According to an embodiment of the present invention, a big data-based power distribution equipment status visualization platform uses multi-dimensional visualization display functions to maximize the reuse of established various power distribution equipment basic data and maintenance management resources, and use data integration And data mining technology to achieve comprehensive equipment analysis, carry out state evaluation, fault diagnosis, risk assessment, life prediction and maintenance decision-making work for transmission and transformation equipment, assist with technical supervision and intelligent report management, and provide a comprehensive solution for the operation and maintenance department.

Scenery, real-time, multi-dimensional, and intelligent equipment management and control information platform, so that a large number of equipment with different conditions, new and old, and in different states are always under strict monitoring, to prevent equipment operation risks in advance, and improve the speed of response to emergencies. The business needs of intensive development and lean management, improve equipment multi-source information interaction and integration capabilities, realize panoramic real-time perception, multi-dimensional intelligent monitoring and control of power distribution equipment, and make the life cycle management of equipment transparent and efficient.

[0119] In this description, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. means to describe in conjunction with the embodiment or example Specific features, structures, materials or characteristics are included in at least one embodiment or example of the present invention. In this context, the schematic representations of the above-mentioned terms do not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0120] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, and substitutions can be made to these embodiments without departing from the principle and purpose of the present invention. And variations, the scope of the present invention is defined by the claims and their equivalents.

CN 108564254 Β