Method and device for identifying future-state power grid maintenance risks based on a power grid knowledge graph

Abstract

The present invention relates to a method and device for identifying future-state power grid maintenance risks based on a power grid knowledge graph. The method includes constructing a power grid risk quantification index evaluation model; training the preset power grid knowledge graph to associate power grid maintenance risk identification logic data with the preset power grid knowledge graph, and constructing a future-state power grid maintenance risk identification model based on the power grid knowledge graph; obtaining equipment information of power outage requirements and inputting it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result, and generating a risk warning notice. Based on the existing power outage maintenance knowledge graph, the present invention stores the power grid safety risk quantification evaluation standard, constructs a knowledge system for future-state risk identification of power grid outage maintenance in the knowledge graph, and uses visualization technology to describe the mining, analysis, construction, and drawing of the power grid outage maintenance risk information by the knowledge graph, so as to output the result of the risk warning notice.

Description

Technical Field

[0001] The present invention belongs to the technical field of power markets, and particularly relates to a method and device for identifying future-state power grid maintenance risks based on a power grid knowledge graph. Background Art

[0002] With the continuous deepening of industrial transformation and upgrading in the energy and power field in China and the continuous advancement of power market reform, a large number of new energy sources and emerging loads are connected. The scale of power grid infrastructure commissioning, major repairs, technical renovations, and line pole relocations has expanded unprecedentedly, and the form and operating characteristics of the power system have become increasingly complex. The power grid has gradually evolved into a new type of network with multiple factors such as power sources, loads, energy storage, and people being random and real-time uncertain. With the change of the power grid structure, challenges are faced in aspects such as dispatching operation methods, power grid carrying capacity, and risk carrying capacity. Traditional methods can no longer meet the diverse requirements of power grid stability maintenance, and power grid maintenance risk identification based on artificial intelligence has become one of the important means to predict the future state of the power grid.

[0003] How to introduce a power grid risk identification method based on artificial intelligence in the power maintenance plan scheduling work to effectively control various risk factors has theoretical significance and practical value for reducing the accident rate and improving power supply reliability. The power grid risks caused by power outages for maintenance are as follows: (1) Overlapping maintenance will weaken the electrical connection between substation nodes and the power grid system; (2) Simultaneous maintenance will lead to the risk of substation voltage loss or becoming an electrical island; (3) Electrical islands and substation voltage loss will lead to the risk of regional load loss; (4) The power grid operation risk caused by the N-1 fault set when arranging maintenance; (5) The load overload risk of single transformers, single busbars, etc. caused by load transfer due to N-1 faults.

[0004] In summary, most of the existing risk identification methods rely on manual experience judgment in combination with the network topology structure. However, the subjective influencing factors in manual experience judgment account for a large proportion. When a large number of power outage requirements need to be scheduled, the manual identification efficiency can no longer adapt to the increasingly complex power grid form and operating characteristics, and there are loopholes in aspects such as risk identification accuracy and power grid operation rationality. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to overcome the deficiencies of the prior art and provide a method and device for identifying future-state power grid maintenance risks based on a power grid knowledge graph to solve the problems of loopholes in aspects such as risk identification accuracy and power grid operation rationality in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solutions: A method for identifying future-state power grid maintenance risks based on a power grid knowledge graph, including:

[0007] Constructing a power grid risk quantification index evaluation model; a risk level corresponding to a power grid risk value is set in the power grid risk quantification index evaluation model;

[0008] Incorporate the power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model based on a preset power grid knowledge graph, and train the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, so as to construct a future-state power grid maintenance risk identification model based on the power grid knowledge graph;

[0009] Obtain the device information of the power outage demand and input it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result;

[0010] Generate a risk warning notice according to the risk identification result.

[0011] Furthermore, the power grid maintenance risk identification logic data includes:

[0012] Quantification index of the severity of power grid risk hazards, quantification index of social impact factors, quantification index of lost load or user quality, device type factor, and historical statistics factor.

[0013] Furthermore, the training of the preset power grid knowledge graph includes:

[0014] Knowledge modeling, data extraction, knowledge extraction, graph construction, storage optimization, knowledge reasoning, and scenario application.

[0015] Furthermore, the obtaining of the device information of the power outage demand and inputting it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result includes:

[0016] Input the device information of the power outage demand into the power grid maintenance risk identification model of the power grid knowledge graph;

[0017] The power grid maintenance risk identification model of the power grid knowledge graph performs power grid topology structure analysis on the device information and then conducts risk identification on the power outage demand;

[0018] Obtain a risk identification result.

[0019] Furthermore, the risk identification of the power outage demand includes:

[0020] Statistical analysis of power outage users, identification of overlapping maintenance risks, identification of electrical island risks, identification of substation voltage loss risks, identification of overload risks in load transfer, and risk identification of power grid risk points analysis.

[0021] Furthermore, the generating of the risk warning notice according to the risk identification result includes:

[0022] Input the risk identification result into a list.

[0023] Furthermore, the risk early warning notice includes:

[0024] Risk name, risk number, risk level, risk value, accident / incident level, risk start time, risk end time, and recommended measures.

[0025] The embodiment of the present application provides a future-state power grid maintenance risk identification device based on a power grid knowledge graph, including:

[0026] A first construction module for constructing a power grid risk quantification index evaluation model; a risk level corresponding to the power grid risk value is set in the power grid risk quantification index evaluation model;

[0027] A second construction module for incorporating power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model based on a preset power grid knowledge graph, and training the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, thereby constructing a future-state power grid maintenance risk identification model based on the power grid knowledge graph;

[0028] A calculation module for obtaining device information of the power outage requirement and inputting it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result;

[0029] A generation module for generating a risk early warning notice according to the risk identification result.

[0030] The beneficial effects that can be achieved by the present invention adopting the above technical solutions include:

[0031] The present invention provides a future-state power grid maintenance risk identification method and device based on a power grid knowledge graph. The present invention proposes to adopt a future-state power grid maintenance risk identification method based on a power grid knowledge graph to assist power grid staff in identifying future-state power grid risks in the preparation of maintenance plans. Based on the local dispatching power outage maintenance knowledge graph, the power grid safety risk quantification evaluation standard is stored, and a knowledge system for future-state risk identification of power grid outage maintenance in the knowledge graph is constructed. Visualization technology is used to describe the mining, analysis, construction, and drawing of power outage maintenance risk information by the knowledge graph, so as to output the result of the risk early warning notice. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0033] Figure 1 Schematic diagram of the steps of the method for identifying future power grid maintenance risks based on the power grid knowledge graph of the present invention;

[0034] Figure 2 Schematic diagram of the process of the method for identifying future power grid maintenance risks based on the power grid knowledge graph of the present invention;

[0035] Figure 3 Schematic diagram of the structure of the device for identifying future power grid maintenance risks based on the power grid knowledge graph of the present invention. Detailed implementation manners

[0036] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other implementation manners obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope protected by the present invention.

[0037] The following introduces a specific method and device for identifying future power grid maintenance risks based on the power grid knowledge graph provided in the embodiments of the present application with reference to the accompanying drawings.

[0038] As Figure 1 shown, the method for identifying future power grid maintenance risks based on the power grid knowledge graph provided in the embodiments of the present application includes:

[0039] S101, constructing an evaluation model for power grid risk quantification indicators; a risk level corresponding to the power grid risk value is set in the evaluation model for power grid risk quantification indicators;

[0040] According to the "Power Grid Safety Risk Quantification Evaluation Standard", the power grid risk value = max{(risk hazard value) × (risk probability value)}, where: risk hazard value = (hazard severity score) × (social impact factor) × (loss of load or user nature factor), and risk probability value = (equipment type factor) × (fault category factor) × (historical data statistics factor). According to the magnitude of the power grid risk value, the power grid risk is divided into six levels, level I risk (red: risk value ≥ 1500), level II risk (orange: 800 ≤ risk value < 1500), level III risk (yellow: 120 ≤ risk value < 800), level IV risk (blue: 20 ≤ risk value < 120), level V risk (white: 5 ≤ risk value < 20), and level VI risk (2 < risk value < 5).

[0041] S102. Incorporate the power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model based on a preset power grid knowledge graph, and train the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, thereby constructing a future-state power grid maintenance risk identification model based on the power grid knowledge graph.

[0042] Preferably, the power grid maintenance risk identification logic data includes:

[0043] Quantification index of the severity of power grid risk hazards, quantification index of social impact factors, quantification index of lost load or user quality, equipment type factor, and historical statistical factor.

[0044] This application incorporates the following power grid maintenance risk identification rule data into the database based on the topological structure and data relationship of the existing power grid knowledge graph:

[0045] Quantification index of the severity of power grid risk hazards. According to the threat that the risk may pose to the safety of the power grid and the degree of load loss, the severity of hazards is divided into twelve levels, and each level of hazard corresponds to the levels of power safety accidents and events stipulated in the relevant regulations.

[0046] Quantification index of social impact factors. The quantification index of social impact factors is rated during the power supply guarantee period and is divided into: normal period (1), power supply guarantee during special period (1.2), secondary power supply guarantee (1.4), primary power supply guarantee (1.6), and special-grade power supply guarantee (2).

[0047] Quantification index of lost load or user quality. Based on Baidu Map, incorporate the area where the power grid equipment is located and the user information data affected into the power grid risk quantification index evaluation model. The quantification index of lost load or user quality is divided into: county-level and suburban load (1.2), urban load (1.5), secondary important users (2.1), important users (2.3), and special-grade users (2.5).

[0048] Equipment type factor: including electrical primary equipment type factor, as shown in Table 1.

[0049]

[0050]

[0051] Communication equipment type factor, as shown in Table 2.

[0052]

[0053] Automation equipment type factor, as shown in Table 3.

[0054]

[0055] Fault category factor: Fault category factors for primary equipment, protection, stability control and other components ① The fault categories are selected according to the requirements of the "Power System Safety and Stability Guide". ② When evaluating the baseline risk of the power grid, the first-level, second-level and third-level faults need to be considered. ③ When evaluating the problem-based risk, the first-level and second-level faults specified in the "Power System Safety and Stability Guide" and the unconventional faults with a probability score not lower than 0.1 should be considered.

[0056] The specific grading is shown in Tables 4, 5, 6 and 7.

[0057] Table 4 Fault category factor

[0058] Type First-level fault Second-level fault Third-level fault Score 0.8～1.2 0.1～0.6 0～0.2

[0059] Table 5 Fault categories and values of primary, protection and stability control components

[0060]

[0061]

[0062] Table 6 Fault category factor of communication equipment

[0063] Type First type of fault Second type of fault Third type of fault Fourth type of fault Score 1～0.8 0.8～0.6 0.6～0.2 0～0.2

[0064] Table 7 Fault category factor of automation equipment

[0065]

[0066] Table 8 Historical statistics factor

[0067]

[0068] Historical statistics factor: Historical data statistics factor = 1 + number of faults of the same type of equipment occurred last year / number of such equipment. ① In this part, only the more serious yellow, orange and red warning levels of the new version of meteorological disaster warning signals are selected; ② Typhoon: yellow warning takes 1 - 2, orange warning takes 2 - 3, red warning takes 3 - 4; ③ Thunderstorm with strong wind: yellow warning takes 1 - 1.2, orange warning takes 1.2 - 1.5, red warning takes 1.5 - 2; ④ Forest fire risk: orange warning takes 1 - 1.2, red warning takes 1 - 1.5; ⑤ High temperature: orange warning takes 1.1, red warning takes 1.2; ⑥ Heavy fog: orange warning takes 1.1, red warning takes 1.2; ⑦ Ice formation: The value is taken according to the weather conditions and the ice coating situation of the line. The specific grading is shown in Appendix 8.

[0069] Preferably, the training of the preset power grid knowledge graph includes:

[0070] Knowledge modeling, data extraction, knowledge extraction, graph construction, storage optimization, knowledge reasoning and scenario application.

[0071] Among them, knowledge extraction includes entity extraction, relationship extraction and attribute extraction; knowledge graph construction includes data cleaning, data normalization, semantic disambiguation, logical verification, format conversion, and data ablation. After the above processing of the data, quality assessment and knowledge reasoning are carried out; data extraction includes: risk identification quantification criteria, Google Maps data, and relational database user data; storage optimization includes knowledge framework, fusion knowledge base, and function optimization. Among them, the knowledge framework includes regulatory documents, user information, work processes, and industry knowledge. Based on expert experience, the knowledge framework mining logic is constructed to generate a general framework, user framework, and industry framework, which are then stored in the fusion knowledge base. The fusion knowledge base includes triples, graph databases, and relational databases. Finally, function optimization is carried out, including inverted index, incremental update, permission management, logical operation, and disaster tolerance mechanism.

[0072] S103, input the device information of the power outage requirement into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result;

[0073] Input the device information of the power outage requirement, and adopt the N-1 and N-2 fault sets to traverse the power grid risk identification system of the power grid knowledge graph. According to the devices in the power outage plan, through the analysis of the power grid topology structure of the knowledge graph, the list of users affected by the power outage of this device and the important 10kV users with reduced power supply reliability are analyzed; through N-1 verification, the number and list of power outage users that may be caused after the N-1 fault of this power outage device are obtained. The risk identification accuracy rate of the model is verified according to manual experience. According to the analysis of the power grid topology structure, the risk of the power outage plan is automatically identified, and the output includes the substation voltage loss assessment result, risk level, risk category, and hazard point analysis content. The risk identification process is shown in the appendix Figure 2 。

[0074] S104, generate a risk warning notice according to the risk identification result.

[0075] The risk identification result includes: fault type, hazard name, risk description, possible risk consequences, lost load (MW), lost load ratio (%), number of lost users, lost user ratio (%), impact on important users (total power outage). According to the power outage plan, the power grid risk level of each power outage plan can be obtained. According to the South Grid risk quantification assessment method, the following can be obtained: hazard value (hazard severity score, social impact factor, lost load or user nature factor); probability value (device type factor, fault category factor, historical data statistics factor), risk value, risk level. The risk warning notice is shown in Table 9.

[0076]

[0077] The working principle of the future-state power grid maintenance risk identification method based on the power grid knowledge graph is as follows: The power grid knowledge graph preset in this application is established based on the existing power grid topology and has functions such as power grid equipment relationship construction, human-computer interaction data query, and hidden relationship reasoning. In addition, it also meets information extraction, knowledge fusion, and knowledge processing. The existing power grid knowledge graph satisfies the following functions: It can extract the specific information of each device in the power grid topology and the relationships between devices from various types of data sources, and form an ontological expression on this basis. After obtaining new knowledge (such as new device information), it can achieve self-consistency and eliminate the contradictions and ambiguities between new knowledge and historical data. For the newly fused knowledge, a data quality assessment is given, and the qualified part is incorporated into the knowledge base to ensure the quality of the knowledge graph.

[0078] The embodiment of this application provides a future-state power grid maintenance risk identification device based on the power grid knowledge graph, including:

[0079] The first construction module 301 is used to construct a power grid risk quantification index evaluation model; the power grid risk quantification index evaluation model is provided with risk levels corresponding to power grid risk values;

[0080] The second construction module 302 is used to incorporate the power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model based on the preset power grid knowledge graph, and train the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, and construct a future-state power grid maintenance risk identification model based on the power grid knowledge graph;

[0081] The calculation module 303 is used to obtain the device information of the power outage demand and input it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result;

[0082] The generation module 304 is used to generate a risk warning notice according to the risk identification result.

[0083] The working principle of the future-state power grid maintenance risk identification device based on the power grid knowledge graph provided by this application is as follows: The first construction module 301 constructs a power grid risk quantification index evaluation model; the risk level corresponding to the power grid risk value is set in the power grid risk quantification index evaluation model; the second construction module 302 incorporates the power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model based on the preset power grid knowledge graph, and trains the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, thereby constructing a future-state power grid maintenance risk identification model based on the power grid knowledge graph; the calculation module 303 obtains the device information of the power outage requirement and inputs it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result; the generation module 304 generates a risk warning notice according to the risk identification result.

[0084] An embodiment of this application provides a computer device, including a processor and a memory connected to the processor;

[0085] The memory is used to store a computer program, and the computer program is used to execute the future-state power grid maintenance risk identification method provided in any of the above embodiments;

[0086] The processor is used to call and execute the computer program in the memory.

[0087] In summary, the present invention provides a future-state power grid maintenance risk identification method and device based on a power grid knowledge graph. The method includes constructing a power grid risk quantification index evaluation model; training the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, thereby constructing a future-state power grid maintenance risk identification model based on the power grid knowledge graph; obtaining the device information of the power outage requirement and inputting it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result, and generating a risk warning notice. Based on the existing power outage maintenance knowledge graph, the present invention stores the power grid safety risk quantification evaluation standard, constructs a knowledge system for the future-state risk identification of the power grid outage maintenance in the knowledge graph, and uses visualization technology to describe the mining, analysis, construction, and drawing of the power grid outage maintenance risk information by the knowledge graph, so as to output the result of the risk warning notice.

[0088] It can be understood that the above-provided method embodiment corresponds to the above device embodiment, and the corresponding specific content can be referred to each other and will not be elaborated here.

[0089] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.

[0090] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0091] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction method that realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0092] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0093] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for identifying future-state power grid maintenance risks based on a power grid knowledge graph, characterized in that, it includes: Constructing a power grid risk quantification index evaluation model; in the power grid risk quantification index evaluation model, there is a risk level corresponding to the power grid risk value; Based on the preset power grid knowledge graph, incorporating power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model, and training the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, constructing a future-state power grid maintenance risk identification model based on the power grid knowledge graph; wherein, the power grid maintenance risk identification logic data includes: power grid risk severity quantification index, social impact factor quantification index, loss of load or user quality quantification index, equipment type factor, and historical statistics factor; The training of the preset power grid knowledge graph includes: knowledge modeling, data extraction, knowledge extraction, graph construction, storage optimization, knowledge reasoning, and scenario application; Obtaining the device information of the power outage demand and inputting it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result, including: inputting the device information of the power outage demand into the power grid maintenance risk identification model of the power grid knowledge graph; the power grid maintenance risk identification model of the power grid knowledge graph analyzes the power grid topology structure of the device information and then identifies the risk of the power outage demand; obtaining a risk identification result; Among them, the risk identification of the power outage demand includes: counting power outage users, identifying overlapping maintenance risks, identifying electrical island risks, identifying substation loss-of-voltage risks, identifying load transfer overload risks, and analyzing risk points of the power grid; Generating a risk warning notice according to the risk identification result.

2. The method according to claim 1, characterized in that, The generating a risk warning notice according to the risk identification result includes: Inputting the risk identification result into a list.

3. The method according to claim 2, characterized in that, The risk warning notice includes: Risk name, risk number, risk level, risk value, accident event level, risk start time, risk end time, and recommended measures.

4. A device for identifying future-state power grid maintenance risks based on a power grid knowledge graph, characterized in that, it includes: A first construction module for constructing a power grid risk quantification index evaluation model; in the power grid risk quantification index evaluation model, there is a risk level corresponding to the power grid risk value; The second construction module is used to incorporate the power grid maintenance risk identification logic data into the power grid risk quantification index evaluation model based on the preset power grid knowledge graph, and train the preset power grid knowledge graph to associate the power grid maintenance risk identification logic data with the preset power grid knowledge graph, so as to construct a future-state power grid maintenance risk identification model based on the power grid knowledge graph; wherein, the power grid maintenance risk identification logic data includes: power grid risk severity quantification index, social impact factor quantification index, loss of load or user quality quantification index, equipment type factor, and historical statistical factor; the training of the preset power grid knowledge graph includes: knowledge modeling, data extraction, knowledge extraction, graph construction, storage optimization, knowledge reasoning, and scenario application; The calculation module is used to obtain the equipment information of the power outage demand and input it into the power grid maintenance risk identification model of the power grid knowledge graph to obtain a risk identification result, specifically for inputting the equipment information of the power outage demand into the power grid maintenance risk identification model of the power grid knowledge graph; the power grid maintenance risk identification model of the power grid knowledge graph performs power grid topology structure analysis on the equipment information and then identifies the risk of the power outage demand; obtaining a risk identification result; wherein, the risk identification of the power outage demand includes: counting power outage users, overlapping maintenance risk identification, electrical island risk identification, substation voltage loss risk identification, load transfer overload risk identification, and power grid risk point analysis risk identification; The generation module is used to generate a risk warning notice according to the risk identification result.

Images