CN106815647A - A kind of high efficiency distribution network failure repairing system and method based on data analysis - Google Patents

Abstract

The present invention provides a high-efficiency distribution network fault repair system and method based on data analysis, including a real-time analysis unit for real-time display of fault density, power outage range, heavy overload of lines and equipment, and visual monitoring of repair vehicles and processes, The emergency repair analysis unit for analyzing faulty equipment, fault causes and work intensity, the emergency repair prediction unit for analyzing and grasping the distribution network fault rules, the high-quality service management unit for analyzing user complaint data and providing Internet power services, and the integrated The fault data provides the decision-making unit for auxiliary decision-making information of operation and maintenance; the present invention builds a data-based analysis platform by collecting and sorting out distribution network emergency repair related data, realizes lean, grid-based, and integrated management of distribution network emergency repair, and provides distribution network equipment management , high-quality service management, distribution network planning and construction and other work to provide basic data and effective support.

Description

A high-efficiency distribution network fault repair system and method based on data analysis

technical field

The invention relates to the field of emergency repair of distribution network faults, in particular to a high-efficiency distribution network fault repair system and method based on data analysis.

Background technique

In recent years, the public has higher and higher requirements and expectations for the power supply services of power supply enterprises, and the contradiction between supply and demand has become increasingly prominent. Among them, the level of power guarantee and power supply service level are the most concerned, most direct and most realistic interests of the general public. In power supply enterprises, the bridgehead for tackling the level of power supply guarantee, power supply service level and customer satisfaction is the emergency repair business of distribution network, which is related to the fundamental interests of millions of households and customers in urban areas and suburban areas of various cities. The company is an important position to embody social value and service concept. Therefore, the improvement of professional management and service level of distribution network emergency repair is particularly important for the improvement of customer satisfaction.

Currently there are mainly the following problems:

1. At present, the emergency repair work of the distribution network adopts the working method of dispatching 95558 orders, one order dispatching, and keeping the paper records of the emergency repair records. Real-time and predictive analysis and statistics cannot effectively grasp the status and laws of distribution network faults, and there are problems of passive emergency repairs and low efficiency, which may easily lead to timeouts for emergency repairs, low repair quality (high repair rate), low repair efficiency, and low power supply reliability. Some service issues that directly affect customer satisfaction.

2. At present, there is a problem that the emergency repair work of the distribution network is not well targeted. Only relying on the information provided by 95598 and the historical experience of emergency repair personnel, there is no relationship between the type and amount of faults and various related factors (temperature, load, wind, time, latitude and longitude, etc.) It is impossible to accurately grasp the types and causes of on-site failures under different circumstances, and it is impossible to predict and prepare emergency repair tools and materials before special circumstances (peak summer, stormy weather, etc.), or to send additional manpower, vehicles, and tools. Appliances, etc., or even targeted personnel training for certain types of faults that may break out in a large area in advance, etc., may easily lead to long repair times and rush repairs due to insufficient personnel and equipment, incomplete spare parts and materials, and unreasonable stations. The personnel were exhausted, and the power supply could not be restored for a long time, which led to a decrease in customer satisfaction and even complaints.

3. At present, after the emergency repair work is completed, only the receipt is completed in the 95598 system. There is a lack of collation and analysis of the emergency repair data, and it is impossible to summarize and provide the failure rate of stations/lines in the urban area at the same time, 10KV repeated tripping lines, screening of abnormal power consumption areas, The data directly related to the reliability and quality of the user's power supply, such as the analysis of user complaints and the analysis of multiple power outages at the opening and closing stations, cannot grasp the user's power usage in a timely manner, and it is easy to cause power outages due to normal reasons such as failure, maintenance, renovation and upgrading, resulting in short-term "Frequent power outages" seriously affect user experience and cause unnecessary complaints.

4. At present, emergency repair stations only rely on experience summaries and administrative regions to divide and arrange, which lack scientific basis, and are relatively fixed and rigid. They cannot be dynamically arranged according to the trend of fault volume, so as to maximize the efficiency of fault handling, reduce unnecessary running, and greatly reduce user power outage time .

5. At present, besides distribution network failure repair work orders, the current distribution network emergency repair work also involves marketing measurement work orders for low-voltage metering devices, power quality, and burning household appliances. Effective marketing analysis and communication cannot enable the company to quickly understand the user's electricity consumption situation and eliminate all hidden dangers in a timely manner, resulting in dissatisfaction among users or damage to the company's interests due to some small marketing issues.

6. At present, the management decisions related to the emergency repair business of the distribution network are all based on human speculation based on historical experience, weather and other factors, lacking scientific and accurate basis, and the decision-making management is often relatively broad, unable to refine to specific tasks, and there are deviations in implementation. The repair efficiency is not high, and the customer experience cannot reach a high level.

Contents of the invention

In order to solve the above problems, the present invention provides a high-efficiency distribution network fault repair system and method based on data analysis. By collecting and sorting out distribution network repair-related data, a data-based analysis platform is built to realize lean distribution network repair , grid-based, and integrated management, providing basic data and effective support for distribution network equipment management, high-quality service management, distribution network planning and construction, etc.

The technical scheme that the present invention adopts for solving the problems of the technologies described above is as follows:

Technical solution one:

A high-efficiency distribution network fault repair system based on data analysis, which includes a real-time analysis unit for real-time display of fault density, power outage range, heavy and overload conditions of lines and equipment, and visual monitoring of repair vehicles and processes for analyzing faults The emergency repair analysis unit for equipment, fault causes and work intensity, the emergency repair prediction unit for analyzing and grasping the fault rules of distribution network, the high-quality service management unit for analyzing user complaint data and providing Internet power services, and the integrated fault data for providing operation Decision-making unit of dimension auxiliary decision-making information.

Further, the real-time analysis unit includes a fault density analysis module, a power outage range distribution analysis module, a geographic location analysis module for emergency repair vehicles, a real-time emergency repair monitoring module, and a low-voltage station area display module.

Further, the emergency repair analysis unit includes a historical work order statistics module, a fault cause analysis module, a line equipment heavy overload analysis module, and a stagnation point work intensity analysis module.

Further, the emergency repair prediction unit includes a future fault amount prediction module, a fault type prediction and analysis module, and a fault amount prediction grid analysis module.

Further, the high-quality service management unit includes a mobile client emergency repair service module, a high-frequency fault distribution analysis module, a user complaint analysis module, and a standardized emergency repair process establishment module.

Further, the auxiliary decision-making unit includes an analysis module of the city's optimal stagnation point and an auxiliary project approval module.

Technical solution two:

A high-efficiency distribution network fault repair method based on data analysis, comprising the following steps:

1) In the real-time analysis unit, establish a geographic map of distribution network fault repairs, input repair work order information, planned power outage information, and fault power outage information data in real time, display the real-time fault density of the distribution network in the form of a heat map, and display real-time power outages in the form of a range map Scope, display the geographic location of emergency repair vehicles in the form of a positioning map, and display heavy overload lines and station areas in the form of a distribution map; to realize the visualization of fault density and power outage range, automatically lock the repair location, and automatically merge repeated repair work orders, according to the GPS intelligence of the emergency repair vehicle Locking and efficient scheduling, so as to achieve scientific allocation of emergency repair resources, control the overall situation of emergency repairs, and reduce the time for emergency repairs to arrive at the scene;

2) Relying on the 4G telecommunication network and remote terminals in the real-time analysis unit, realize the visual remote emergency repair monitoring of individual soldiers and vehicles, so that the emergency repair personnel can communicate with the command center;

3) In the emergency repair analysis unit, use charts to analyze the data of the whole process of emergency repair, the faulty equipment, the cause of the fault, and the heavy overload of the line and equipment;

4) In the emergency repair prediction unit, based on the support vector machine prediction algorithm and the multi-dimensional fault analysis system, analyze and grasp the fault rules of the distribution network;

5) Establish a complaint analysis model in the high-quality service management unit and rush to repair the user's mobile phone client APP, analyze user complaint data over the years, visually display high-frequency faulty areas and station areas, provide customers with Internet power services, and eradicate defects in advance;

6) In the decision-making unit, model and analyze the optimal station for emergency repairs in the city, and assist in setting the optimal static and dynamic station;

7) Combined with the above analysis results, based on the expert analysis algorithm in the decision-making unit, carry out project reserve or emergency project management according to the degree of urgency for the stations and lines with frequent faults and outstanding user feedback.

Further, step 1) is realized through the following steps: based on ARCgis software, and based on the standard electronic map, the buildings in the map are fixed, divided and partitioned, and the address of a single building is "fixed", and the building is assigned according to the station area "Blocking" and "zoning" according to the entire community, combined with the geographic information and line diagrams of distribution network towers, lines and station equipment in PMS2.0 [Equipment (Asset) Operation and Maintenance Lean Management System], as well as vehicle-mounted GPS real-time Locate the coordinates of emergency repair vehicles, form a geographic map of distribution network fault emergency repairs, and input data such as fault repair work orders, planned power outage information, and fault power outage information in real time; on the one hand, repair repair work orders are automatically assigned to "fixed points" according to geographical distribution The distribution is attributed to "blocks", and the distribution according to the district is attributed to "regions". On the other hand, the planned and faulty lines or equipment are attributed to the towers, lines, and station equipment in the geographic map.

The fault density is obtained by the following method: combining the power grid information with the geographical location, and displaying the density of various faults in the region in different colors through the repair request heat map (contour map), realizing the visualization of work order repair data , so as to automatically lock the address of the latest repair work order and merge the same fault repair work order, reducing duplicate dispatch orders, thereby greatly improving the efficiency of emergency repair.

The power outage range map is established by the following method: Based on the unprocessed power outage repair work orders, faults and planned power outage information, a city-wide power outage range map is constructed, so as to comprehensively, intuitively and finely display the power outage situation of the entire regional distribution network, and realize Overall control of emergency repair work.

The real-time tracking of emergency repair vehicles is realized in the following ways: according to the GPS feedback on the emergency repair vehicle, real-time positioning to the geographical map allows managers to control the actual working conditions of each emergency repair vehicle, and efficiently dispatches emergency repair resources based on the principle of dispatching work orders to the nearest location, reducing the round trip of emergency repairs , reducing the duration of presence.

Further, the step 2) is realized through the following steps: Relying on the "individual, vehicle-mounted visual remote emergency repair monitoring" subsystem (has been accepted by the national invention patent), through 4G network signals and remote terminals, real-time monitoring of the emergency repair site Monitoring, real-time transmission of on-site video and audio information of emergency repair, supervision and guidance of emergency repair progress and service level in the whole process.

Based on the geographic information map, the relationship between the substation and its low-voltage users is intuitively displayed, which enhances the refinement of low-voltage management.

Further, the statistical analysis of the work order is realized by the following method: based on the work order data in the past years, understand the fault situation in the whole city, form a year/month/day comparison chart of the emergency repair work order in the same period, and a month/week/day work order volume trend Figure 1. Real-time tracking and analysis of the progress of emergency repair work orders, so that managers can see the change trend of work orders and work status at a glance, and intelligently form decision-making work orders.

Based on the work order data over the years, the causes of all work order failures are classified and compared to realize failure cause analysis.

Based on the work order data over the years, all faulty equipment in repair work orders are classified and compared to realize faulty equipment analysis.

Extract the relevant heavy overload information in the "integrated data analysis platform", perform geographical positioning, and distinguish it according to different gray levels, and form corresponding decision-making work orders to realize heavy overload analysis of lines and stations.

Extract the work order processing results, and display the work intensity of each station in real time with charts, so that managers can keep abreast of the working status of each staff, rationally deploy personnel, make management more lean, and realize work intensity analysis of stations.

Further, the step 4) is realized through the following steps: taking historical work order quantity, weather (including average temperature, wind force, humidity) and load change as input quantity, outputting accurate prediction of short-term failure quantity through support vector machine prediction algorithm, Then assign weights based on the year of commissioning of the cells in the regional "grid" to predict the number of work orders in each grid; on the other hand, through cluster analysis of various types of faults and related factors, find fault rules, predict fault types, and Displayed in the form of 3D and 2D graphics. In turn, it realizes the advance allocation of resources such as personnel, vehicles, and equipment, assists in the "grid" dynamic layout, and improves the efficiency of emergency repairs.

The work order volume prediction is performed on three different prediction algorithms: gray scale prediction, neural network prediction, support vector machine prediction, and short-term failure volume prediction with different input parameters (work order, weather, load, etc.), and analysis The advantages and disadvantages of various algorithms, the most accurate algorithm to find out the short-term fault quantity prediction is the support vector machine prediction algorithm. By accurately predicting the amount of failures on the next day, corresponding preparations are made in advance to prevent shortage of repair manpower and long waiting time of customers, and it is also used as the decision-making condition of the decision-making analysis module.

The grid work order volume forecast divides the jurisdiction into 30 grids by default, and calculates the average weight (probability) with the weighted value of the commissioning year of the community in each "grid". The weight distribution is carried out by grid, so as to rationally allocate emergency repair resources according to the forecast situation in each grid and change passive emergency repair to active inspection, which greatly reduces the number of work orders and improves the reliability of power supply.

The fault type prediction is displayed in the form of three-dimensional and two-dimensional analysis graphs through the mining of emergency repair data over the years, so that the occurrence of a certain type of fault in a certain area under a certain temperature and a certain load can be clearly obtained, and then the faults of the distribution network can be clearly grasped Occurrence rules, to achieve advanced prediction of faults and high-occurrence periods of various types of faults, to achieve the goal of customer service initiative.

Further, the step 5) is realized by the following methods: the business of receiving orders, arriving on site, merging orders, and canceling orders is concentrated on the handheld mobile terminal, and connected with the emergency repair management function of PMS2.0; through mining and sorting out emergency repair work orders, Extract the location of frequent power outages, station areas, and the number of power outages, distinguish them from large to small on the geographical map of distribution network fault repair, visualize them, and automatically warn users with high-risk complaint tendencies; analyze the types and reasons of user complaints, Establish a standardized emergency repair system.

The handheld mobile terminal is preferably a mobile phone APP, which has the following functions: 1. Customers can link to the PMS2.0 system through the APP one-key repair module to conveniently and efficiently report faults and repairs, and broaden the repair channels; 2. The user can use the APP server according to The repair work order number can be used to inquire the location of the emergency repair vehicle and the progress of the work order as well as the video data of the repair process in real time, and evaluate the service quality of the repair personnel. , and establish a corresponding star system, which is convenient for users to choose, and provides some simple troubleshooting methods and common sense of safe and energy-saving electricity consumption, establishes a discussion area for emergency repairs, grasps what users think in time, and provides users with the most considerate service.

By digging and sorting out emergency repair work orders, extract the locations of communities and station areas with frequent power outages and the number of power outages, and distinguish them from large to small in different gray levels on the geographical map of distribution network fault repairs, visually display them, and provide automatic warnings Users with high-risk complaint tendencies can eradicate related frequent power failure defects in advance, thereby improving customer satisfaction, and assisting the company in setting up projects to manage related station equipment and forming management decisions.

Analyze user complaints, analyze the types and reasons of user complaints, display complaints in charts, and build contact files for sensitive users and property management. Through the marketing 186 system, users will be notified by text messages according to the scope of power outages, and sensitive users and property management will be communicated according to The APP server feeds back information, understands and eases user emotions in a timely manner, and avoids user complaints.

According to the standardized emergency repair process of State Grid Corporation and combined with the actual work, 43 standardized emergency repair processes have been established, a standard emergency repair time map has been formed, standardized emergency repair vehicle configuration has been unified, and 13 standardized emergency repair training videos have been self-directed and self-acted, which has greatly improved the soft and hard power of emergency repair. And according to the whole process tracking and analysis of emergency repair work orders, real-time comparison and timely supervision and inspection of emergency repair work orders that do not meet the regulations.

Further, the step 6) is realized by the following method: constructing a repair station setting model, setting the optimal static station according to the four seasons of the region, historical failure density, traffic conditions, personnel conditions, work intensity and important user distribution, and setting the optimal static station according to a week Fault volume prediction and special event setting dynamic staging points, automatically forming decision-making work orders, realizing scientific and efficient allocation of emergency repair resources, and greatly shortening the time for emergency repair on-site; through real-time analysis, emergency repair analysis, fault prediction, high-quality service management and data integration, Guide the project approval, planning and construction of distribution network.

The expert algorithm is an expert algorithm, which refers to the analysis and calculation of abnormal data of distribution network equipment, team operation and maintenance, fault handling matching and stagnation point distribution based on the experience and knowledge of experts.

The present invention has following technical effect:

The present invention builds a data-based analysis platform by collecting and arranging data related to emergency repairs of distribution networks, and then accurately grasps the fault rules of distribution networks, predicts high-frequency fault areas, time periods, and fault causes in advance, scientifically allocates emergency repair resources, and realizes dynamic layout. Accurately analyze the data trends of various marketing and distribution indicators, form efficient and accurate auxiliary decision-making information, and then assist in improving the efficiency and pertinence of emergency repair work, and realize lean, grid-based, and integrated management of distribution network emergency repairs. Provide basic data and effective support for management, high-quality service management, distribution network planning and construction, etc.;

The invention practically solves a series of problems existing in emergency repair management, such as "kicking the ball", inconsistent emergency repair process, inconsistent on-site safety control, weak high-quality service monitoring means, and loose management of outsourcing teams; more in-depth excavation and analysis of distribution network The relationship between emergency repair business and high-quality service, improvement of power supply reliability, efficient management decision-making and reasonable construction and planning of distribution network in the future can better serve the masses of customers, improve customer satisfaction, and establish a service brand image.

Description of drawings

Accompanying drawing 1 is flow chart of the present invention;

Accompanying drawing 2 is the fault density equation diagram;

Accompanying drawing 3 is a power outage range map;

Attached Figure 4 is a comparative analysis chart of work order volume in each time period;

Accompanying drawing 5 is the fault cause analysis diagram;

Accompanying drawing 6 is the breakdown equipment type analysis diagram;

Accompanying drawing 7 is the heavy overload analysis diagram of line and station area;

Accompanying drawing 8 is the team and group work intensity analysis diagram;

Accompanying drawing 9 is a fault prediction diagram;

Figure 10 is a three-dimensional analysis diagram of related factors.

detailed description

The technical solutions of the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments.

This embodiment is the breakdown repair situation within one week in Shijiazhuang City.

As shown in Figure 1, a high-efficiency distribution network fault repair system based on data analysis, which includes a real-time analysis unit for real-time display of fault density, power outage range, heavy overload of lines and equipment, and visual monitoring of repair vehicles and processes , the emergency repair analysis unit used to analyze the faulty equipment, fault cause and work intensity, the emergency repair prediction unit used to analyze and grasp the fault rules of the distribution network, the high-quality service management unit used to analyze user complaint data and provide Internet power services, and the A decision-making unit that integrates fault data to provide auxiliary decision-making information for operation and maintenance.

The real-time analysis unit includes a fault density analysis module, a power outage range distribution analysis module, a geographic location analysis module for repair vehicles, a real-time repair monitoring module, and a display module for low-voltage station areas.

The emergency repair analysis unit includes a historical work order statistics module, a fault cause analysis module, a line equipment heavy overload analysis module, and a stagnant point work intensity analysis module.

The emergency repair prediction unit includes a future fault amount prediction module, a fault type prediction and analysis module, and a fault amount prediction grid analysis module.

The high-quality service management unit includes a mobile client emergency repair service module, a high-frequency fault distribution analysis module, a user complaint analysis module, and a standardized emergency repair process establishment module.

The auxiliary decision-making unit includes an analysis module of the city's optimal stagnation point and an auxiliary project reconstruction module.

A high-efficiency distribution network fault repair method based on data analysis, comprising the following steps:

1) In the real-time analysis unit, establish a geographic map of distribution network fault repairs, input repair work order information, planned power outage information, and fault power outage information data in real time, display the real-time fault density of the distribution network in the form of a heat map, and display real-time power outages in the form of a range map Scope, display the geographic location of emergency repair vehicles in the form of a positioning map, and display heavy overload lines and station areas in the form of a distribution map; to realize the visualization of fault density and power outage range, automatically lock the repair location, and automatically merge repeated repair work orders, according to the GPS intelligence of the emergency repair vehicle Locking and efficient scheduling, so as to achieve scientific allocation of emergency repair resources, control the overall situation of emergency repairs, and reduce the time for emergency repairs to arrive at the scene;

The gridded emergency repair station is established by the following method: using the standard electronic map as a blueprint, the buildings in the map are fixed, divided and partitioned, "fixed" according to the address of a single building, and "categorized" according to the station area of the building. "Block", according to the "division" of the entire community, combined with the geographical information and line diagram of distribution network towers, lines and station equipment in PMS2.0, and the coordinates of vehicles for real-time positioning and repairing by vehicle-mounted GPS, a geographical map of distribution network fault repairing is formed Real-time input of fault repair work order information, planned power outage information, fault power outage information and other data; on the one hand, repair work orders are automatically assigned to "fixed points" according to geographical distribution, "blocks" according to station area distribution, and "partitions" according to district distribution "; On the other hand, the planned and faulty lines or equipment are classified as pole towers, lines and station equipment in the geographic map.

As shown in Figure 2, the fault density is obtained by establishing a fault density map: combining the grid information with the geographical location, and displaying the density of various faults in the region in different colors through the heat map (contour map) of repairs , realize the visualization of work order repair data, so as to automatically lock the address of the latest repair work order and merge the same fault repair work order, reduce repeated dispatching orders, and greatly improve the efficiency of emergency repair.

As shown in Figure 3, the power outage range map is established by the following method: Based on the unprocessed power outage repair orders, faults and planned power outage information, a city-wide power outage range map is constructed, so as to comprehensively, intuitively and finely display the power distribution of the entire region In the event of network outages, the overall control of emergency repair work is realized.

2) Relying on the 4G telecommunication network and remote terminals in the real-time analysis unit, realize the visual remote emergency repair monitoring of individual soldiers and vehicles, so that the emergency repair personnel can communicate with the command center;

The real-time tracking of emergency repair vehicles is realized in the following ways: according to the GPS feedback on the emergency repair vehicle, real-time positioning to the geographical map allows managers to control the actual working conditions of each emergency repair vehicle, and efficiently dispatches emergency repair resources based on the principle of dispatching work orders to the nearest location, reducing the round trip of emergency repairs , reducing the duration of presence.

The monitoring of the emergency repair process is realized in the following ways: individual and vehicle-mounted visual remote emergency repair monitoring, real-time monitoring of the emergency repair site through 4G network signals, real-time transmission of video and audio information on the emergency repair site, and supervision and guidance of the emergency repair progress and service level throughout the process.

Based on the geographic information map, the relationship between the substation and its low-voltage users is intuitively displayed, which enhances the refinement of low-voltage management.

3) In the emergency repair analysis unit, use charts to analyze the data of the whole process of emergency repair, the faulty equipment, the cause of the fault, and the heavy overload of the line and equipment; realize the visual remote emergency repair monitoring of individual soldiers and vehicles, build a standardized emergency repair system, and form a "power distribution Network standardization emergency repair operation improvement", enhance the soft and hard power of emergency repair, achieve the goal of lean emergency repair management, and improve emergency repair efficiency;

Chart analysis includes work order statistical analysis, fault cause analysis, fault equipment analysis, heavy overload analysis of lines and station areas, and stagnation point work intensity analysis.

As shown in Figure 4, the work order statistical analysis is realized by the following methods: According to the work order data in the past years, understand the fault situation in the whole city, form the year/month/day comparison chart of the rush repair work order in the same period, and the month/week/day work order Order quantity trend chart, real-time tracking and analysis of the progress of emergency repair work orders, so that managers can see the change trend of work orders and work status at a glance, and intelligently form decision-making work orders.

As shown in Figure 5, based on the work order data over the years, the causes of all work order failures are classified and compared, and a failure cause analysis diagram is formed to realize failure cause analysis.

As shown in Figure 6, based on the work order data over the years, all faulty equipment in the repair work order is classified and compared to form a faulty equipment type analysis diagram, from which the type of faulty equipment can be seen at a glance.

As shown in Figure 7, the relevant heavy overload information in the "integrated data analysis platform" is extracted, geographically positioned, and differentiated according to different gray levels, and corresponding decision-making work orders are formed to form heavy overload analysis diagrams of lines and stations to realize Line and station heavy overload analysis.

As shown in Figure 8, the work order processing results are extracted, and the work intensity of each station is displayed in real time with the heavy overload analysis diagram of the line and station area, so that the manager can keep abreast of the working status of each staff, rationally deploy personnel, and make management more lean. In order to realize stagnation point work strength analysis.

4) In the emergency repair prediction unit, analyze and grasp the fault rules of the distribution network, based on the support vector machine prediction algorithm; realize the advanced prediction of the fault amount, high-incidence area, time period and fault type, and actively prepare in advance to achieve the goal of customer service initiative;

The prediction algorithm includes work order quantity prediction, grid work order quantity prediction and fault type prediction.

As shown in Figure 9, the work order quantity prediction is carried out in short-term by using different input parameters (work order, weather, load, etc.) for three different prediction algorithms: gray scale prediction, neural network prediction, and support vector machine prediction. Prediction of faults, forming a fault prediction map, analyzing the advantages and disadvantages of various algorithms, and finding the most accurate algorithm for short-term fault prediction is the support vector machine prediction algorithm. By accurately predicting the faults of the next day, make corresponding preparations in advance Work to prevent shortage of emergency repair manpower and long waiting time for customers, and also serve as the decision-making condition of the decision-making analysis module.

The grid work order volume forecast divides the jurisdiction into 30 grids by default, and calculates the average weight (probability) with the weighted value of the commissioning year of the community in each "grid". The weight distribution is carried out by grid, so as to rationally allocate emergency repair resources according to the forecast situation in each grid and change passive emergency repair to active inspection, which greatly reduces the number of work orders and improves the reliability of power supply.

The fault type prediction is displayed in the form of three-dimensional and two-dimensional analysis diagrams through the mining of emergency repair data over the years, forming a three-dimensional analysis diagram of related factors as shown in Figure 10, from which it can be clearly obtained that a certain temperature and a certain load under a certain temperature can be clearly obtained. The occurrence of various types of faults in a certain area, and then clearly grasp the occurrence rules of distribution network faults, realize advanced prediction of faults and high-occurrence periods of each fault type, and achieve the goal of customer service initiative.

5) Establish a complaint analysis model in the high-quality service management unit and rush to repair the user's mobile phone client APP, analyze the user complaint data over the years, visually display high-frequency fault areas and station areas, provide customers with Internet power services, and eradicate defects in advance; avoid frequent power outages, rely on The mobile client enables customers to report for repairs with one click, track the repaired vehicles, view the video of the repair process in real time, and report for faults in the user's property rights with one click, achieving the most intimate goal of serving users.

It is mainly realized through the following methods: the business of receiving orders, arriving at the scene, merging orders, and canceling orders is concentrated on the handheld mobile terminal, which is connected with the PMS2. The location of the station area and the number of power outages are distinguished from large to small on the geographical map of distribution network fault repairs, visualized, and users with high-risk complaint tendencies are automatically warned; the types and reasons of user complaints are analyzed, and a standardized repair system is established.

The handheld mobile terminal in this embodiment is the "Handheld Electric Power" APP of Hebei Province, and the "Mobile Emergency Repair Service Management Terminal" has been developed. It has the following functions: 1. Customers link to the PMS2.0 system through the APP one-key repair module, which is convenient and efficient. Apply for repairs and broaden the channels for repairing; 2. Users can use the APP server to inquire in real time the location of the vehicle receiving the repair order, the progress of the work order and the video data of the repair process according to the repair order number, and evaluate the service quality of the repair personnel. 3. For For internal failures of customers, the APP user server provides qualified electricians or construction companies, and establishes a corresponding star system to facilitate users to choose, and provides some simple troubleshooting methods and common sense of safe and energy-saving electricity use, and establishes a discussion area for emergency repairs. Grasp what users think in time and provide users with the most intimate service.

By digging and sorting out emergency repair work orders, extract the locations of communities and station areas with frequent power outages and the number of power outages, and distinguish them from large to small in different gray levels on the geographical map of distribution network fault repairs, visually display them, and provide automatic warnings Users with high-risk complaint tendencies can eradicate related frequent power failure defects in advance, thereby improving customer satisfaction, and assisting the company in setting up projects to manage related station equipment and forming management decisions.

Analyze user complaints, analyze the types and reasons of user complaints, display complaints in charts, and build contact files for sensitive users and property management. Through the marketing 186 system, users will be notified by text messages according to the scope of power outages, and sensitive users and property management will be communicated according to The APP server feeds back information, understands and eases user emotions in a timely manner, and avoids user complaints.

According to the standardized emergency repair process of State Grid Corporation and combined with the actual work, 43 standardized emergency repair processes have been established, a standard emergency repair time map has been formed, standardized emergency repair vehicle configuration has been unified, and 13 standardized emergency repair training videos have been self-directed and self-acted, which has greatly improved the soft and hard power of emergency repair. And according to the whole process tracking and analysis of emergency repair work orders, real-time comparison and timely supervision and inspection of emergency repair work orders that do not meet the regulations. This function has won the "First Prize of Hebei Provincial Electric Power Company's 2014 Outstanding Employee Innovation Achievement".

6) In the decision-making unit, model and analyze the optimal station for emergency repairs in the city, and assist in setting the optimal static and dynamic station;

Further, the step 6) is realized by the following method: constructing a repair station setting model, setting the optimal static station according to the four seasons of the region, historical failure density, traffic conditions, personnel conditions, work intensity and important user distribution, and setting the optimal static station according to a week Fault volume prediction and special event setting dynamic staging points, automatically forming decision-making work orders, realizing scientific and efficient allocation of emergency repair resources, and greatly shortening the time for emergency repair on-site; through real-time analysis, emergency repair analysis, fault prediction, high-quality service management and data integration, Guide the project approval, planning and construction of distribution network.

7) Combined with the above analysis results, based on the expert analysis algorithm in the decision-making unit, for the stations and lines with frequent faults and outstanding user feedback, carry out project reserve or emergency project management according to the degree of urgency, and display them in the form of pending work orders. A single decision is processed at one time, and projects are established to transform communities with frequent power outages, realizing intelligent emergency repairs and improving the company's investment accuracy.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. a kind of high efficiency distribution network failure based on data analysis rushes to repair system, it is characterised in that it is included for aobvious in real time Show fault density, power failure range, circuit and equipment weight overload situations and visual control recovery vehicle and process real-time point Analysis unit, the repairing analytic unit for analyzing faulty equipment, failure cause and working strength grasps power distribution network event for analyzing Hinder the repairing predicting unit of rule, for the good service management list analyzed customer complaint data, provide internet electrical power services Unit and the decision package for integrating fault data offer O＆M decision-making assistant information.

2. a kind of high efficiency distribution network failure based on data analysis according to claim 1 rushes to repair system, and its feature exists In the real-time analytic unit includes fault density analysis module, power failure range distributional analysis module, recovery vehicle geographical position Analysis module, repairing real-time monitoring module and low-voltage platform area display module.

3. a kind of high efficiency distribution network failure based on data analysis according to claim 1 rushes to repair system, and its feature exists In the repairing analytic unit includes history trouble ticket statistical module, failure reason analysis module, line facility weight overloading analysis mould Block and stationary point working strength analysis module.

4. a kind of high efficiency distribution network failure based on data analysis according to claim 1 rushes to repair system, and its feature exists In the repairing predicting unit includes the prediction of future malfunction amount prediction module, fault type forecast analysis module and failure amount Gridding analysis module.

5. a kind of high efficiency distribution network failure based on data analysis according to claim 1 rushes to repair system, and its feature exists In the good service administrative unit includes that mobile client repairing service module, high frequency fault distributional analysis module, user throw Tell that analysis module and standardization repairing flow set up module.

6. a kind of high efficiency distribution network failure based on data analysis according to claim 1 rushes to repair system, and its feature exists In the aid decision unit includes the optimal stationary point analysis module in city, auxiliary project verification transformation module.

7. a kind of high efficiency distribution network failure emergency repair method based on data analysis as claimed in claim 1, it is characterised in that It comprises the following steps：

Real-time analytic unit set up distribution network failure repairing geographical map, real-time typing report for repairment work order information, scheduled outage information, Fault outage information data, power distribution network real time fail density is shown with heating power diagram form, is had a power failure in real time with the displaying of scope diagram form Scope, positioning diagram form displaying recovery vehicle geographical position, overload circuit and Tai Qu again with the displaying of distribution map form；

4G communication networks and remote terminal are relied in real-time analytic unit, individual soldier, vehicle-mounted visual remote repairing monitoring is realized, Repair personnel is set to be interconnected with command centre；

In repairing analytic unit, the weight of overall process data, faulty equipment, failure cause and circuit and equipment is rushed to repair with graphic analyses Overload situations；

In repairing predicting unit, based on SVM prediction algorithm and various dimensions trouble analysis system, distribution is grasped in analysis Net fault observer；

Set up in good service administrative unit and complain analysis model and repairing user mobile phone client APP, analyzed user over the years and throw Data are told, visual presentation high frequency fault cell, Tai Qu, there is provided customers Internet electrical power services eradicate defect in advance；

Optimal stationary point is rushed to repair in decision package modeling analysis city, auxiliary sets optimal static and dynamic stationary point；

With reference to above-mentioned analysis result, analysis expert algorithm is based in decision package, the platform area prominent to Frequent Troubles, user's reflection And circuit, project deposit is carried out according to urgency level or urgent project verification is administered.

8. a kind of high efficiency distribution network failure emergency repair method based on data analysis described in claim 7,

Characterized in that, the step 1）Realized by following steps：Based on ARCgis softwares, with standard electronic map as blue This, building in map is pinpointed, piecemeal and subregion, with reference to power distribution network shaft tower, circuit and station kind equipment in PMS2.0 Geography information and line map, and vehicle GPS position recovery vehicle coordinate in real time, form distribution network failure repairing geographical map, real When typing troublshooting work order information, scheduled outage information, fault outage information data.

9. a kind of high efficiency distribution network failure emergency repair method based on data analysis according to claim 7, its feature exists In the step 4）Realized by following steps：With history trouble ticket amount, weather and load variations situation as input quantity, by branch Vector machine prediction algorithm output precisely prediction short duration failure amount is held, then with the time distribution power that puts into operation of cell in regional " grid " Weight, predicts work order amount in each grid；On the other hand by the cluster analysis to all types of failures and correlative factor, failure rule are searched Rule, predicts fault type, and represent in the form of three-dimensional, X-Y scheme.

10. a kind of high efficiency distribution network failure emergency repair method based on data analysis according to claim 7, its feature exists In the step 6）It is realized by the following method：Build repairing stationary point and model is set, according to regional period in the four seasons, historical failure The distribution of density, traffic conditions, personnel's situation, working strength and responsible consumer set optimal static stationary point, according to one week failure amount Prediction and special event set dynamic stationary point, automatically form decision-making work order, realize science, high-efficient disposition repairing resource, significantly contract It is short to rush to repair duration of showing up；Managed by analysis in real time, repairing analysis, failure predication, good service and the integration to data, instructed The project verification transformation of power distribution network, planning construction.