Artificial intelligence driven power grid production management organizational structure optimization method

Abstract

The application provides an artificial intelligence driven power grid production management organization structure optimization method, relates to the technical field of power grid production management, and comprises the following steps: constructing a global data sensing network and an edge computing infrastructure; establishing an intelligent execution terminal, performing automatic inspection operation on power grid management, building a track according to the planning state of the centralized area of power grid equipment, installing an inspection robot to periodically inspect along the preset track, and collecting power information of each power grid equipment; developing an intelligent decision deduction engine and a knowledge graph system; and establishing a dynamically self-adaptive hybrid decision execution unit; through the deep fusion of the global sensing network and the edge computing, the application solves the response lag problem caused by the traditional dependence on manual inspection and hierarchical reporting, adopts the inspection robot to complete the inspection task on the centralized area of power grid equipment, realizes the equipment inspection purpose in a high range, and can flexibly perform partial discharge detection and processing on power grid equipment at different positions.

Description

Technical Field

[0001] This invention relates to the field of power grid production management technology, specifically to an artificial intelligence-driven method for optimizing the organizational structure of power grid production management. Background Technology

[0002] Optimization of the power grid production management organizational structure refers to a systematic reform that improves the efficiency, safety, and economy of power grid operation by adjusting the management architecture, resource allocation, and process system of power companies in production and operation. With the large-scale integration of new energy sources and the acceleration of power marketization and digital transformation, the traditional hierarchical management model is gradually showing problems such as delayed response and insufficient coordination. Optimization measures focus on building a multi-dimensional collaborative mechanism, with the core objective of establishing an efficient management system adapted to the needs of the new power system, balancing safety, efficiency, and sustainable development, and providing a solid organizational guarantee for energy transition.

[0003] In existing technologies, power grid production management mainly relies on manual inspections and hierarchical reporting. This approach leads to response delays and low timeliness in fault location. On the other hand, inspections of areas with concentrated power grid equipment still mainly rely on manual labor. This approach carries high risks and is inefficient. Furthermore, conventional inspection robots are unable to avoid obstacles on the ground. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide an AI-driven method for optimizing the organizational structure of power grid production management, thereby solving the problems mentioned in the background. This invention, through the deep integration of a global sensing network and edge computing, reduces data acquisition and analysis latency from hours to milliseconds, resolving the response lag issues caused by traditional reliance on manual inspections and hierarchical reporting. It employs inspection robots to complete inspection tasks in concentrated areas of power grid equipment. With both ground-level and obstacle-crossing track structures, the inspection robot can move even in the presence of various obstacles on the ground, achieving a wide range of equipment inspection objectives. It can also flexibly perform partial discharge detection and processing on power grid equipment in different locations.

[0005] To achieve the above objectives, the present invention provides a method for optimizing the organizational structure of power grid production management driven by artificial intelligence, comprising the following steps:

[0006] S1. Construct a full-domain data sensing network and edge computing infrastructure, with multi-source data fusion as the core, and establish an intelligent sensing system covering the entire process of generation, transmission, transformation and distribution;

[0007] S2. Establish intelligent execution terminals to perform automated inspection operations for power grid management. Build tracks based on the planned status of concentrated power grid equipment areas, install inspection robots to conduct regular inspections along the preset tracks, and collect power information for each power grid device.

[0008] S3. Develop an intelligent decision-making inference engine and knowledge graph system, and optimize the construction of a dynamic decision-making center based on digital twins and game theory;

[0009] S4. Establish a dynamic and adaptive hybrid decision-making and execution unit, reconstruct the human-machine collaborative task execution mechanism, and divide the response system into multiple levels according to the event level;

[0010] S5. Reconstruct the flexible organizational structure and cross-departmental collaboration mechanism, build a matrix dynamic organization, establish an organizational agility assessment system, quantify departmental collaboration effectiveness through the collaboration response index, and directly link it with performance appraisal.

[0011] S6. Build a model sharing platform to implement and verify models, promote automated iterative upgrades of the system, and simultaneously establish a model performance degradation early warning mechanism.

[0012] Furthermore, in step S1, high-precision IoT sensors, inspection drone clusters, and meteorological satellite data receiving stations are deployed to form a multimodal data warehouse that is updated every minute. Edge computing devices are deployed at key nodes such as substations and new energy power plants to run lightweight AI models under the federated learning framework.

[0013] Furthermore, in step S3, a 1:1 virtual mirror system is built using the power grid physical information model, and an LSTM-GAN hybrid model is integrated to predict the load fluctuation and equipment failure probability distribution in the next 72 hours. The interests of power generators, transmission and transformation operators and dispatch centers are modeled as an incomplete information game using the subject collaborative optimization algorithm.

[0014] Furthermore, in step S4, the AI ​​agent directly dispatches the inspection robot to perform standardized operations, simultaneously generating digital work orders and archiving them in the blockchain evidence storage system. The augmented reality interface simultaneously presents a three-dimensional heat map of the fault impact domain, the distribution of spare parts inventory, and the real-time location of the emergency repair team. The AI ​​generates multiple disposal plans, including success rate estimation and cost accounting.

[0015] Furthermore, the track includes a bottom track and an obstacle-crossing track, both of which are used to provide movement guidance for the inspection robot. The obstacle-crossing track needs to be built according to the obstacles on the ground. If there are pipelines or small power grid devices blocking the ground, the bottom track is extended to one side of the obstacle, and then the obstacle-crossing track is built and supported by columns.

[0016] Furthermore, after the inspection robot moves along the bottom track and moves to one side of the obstacle, it is lifted from the bottom track onto the obstacle-crossing track via the docking platform, and the number of layers and height of the obstacle-crossing track are increased according to the height of the obstacle.

[0017] Furthermore, by activating the electric lifting rod, the transmission plate and the top plate are driven to move up and down. The top plate drives the bottom lifting plate to rise or fall through the connecting column. The lifting of the lifting plate raises the power system and the drive rollers until the lifting drive rollers are pushed from the docking platform surface of the bottom track to the bottom surface of the docking platform of the upper obstacle crossing track.

[0018] Furthermore, the electric lifting rod provides pressure to the drive rollers against the docking platform of the obstacle-crossing track. At this time, the power system is controlled to run in reverse, causing the drive rollers to roll along the docking platform of the obstacle-crossing track to the inside of the moving channel of the obstacle-crossing track. The electric telescopic rod is then controlled to retract, thereby lifting the entire inspection robot.

[0019] Furthermore, it also includes leakage monitoring of ground cable lines: after the first motor is started, the drive shaft drives the rotating arm at the end to rotate. The rotating arm controls the first motor, the pressing wheel, the conductive column and the connecting plate to rotate synchronously. The pressing wheel is used to press on the ground and make contact with the surface of the cable line it passes over, in order to detect and process the leakage status of the ground cable, and transmit the current signal to the partial discharge detector through the conductive column and the wire at the rear end of the conductive column.

[0020] Furthermore, it also includes the detection and correction process for the inspection robot: the first motor drives the rotating arm to a horizontal state, and the second motor controls the connecting plate to move towards the rear end until the positioning post is embedded in the interior of the positioning groove. The interior of the positioning groove is conical. By having the top of the positioning post abut against the innermost area of ​​the positioning groove, the entire connecting plate is positioned, which assists the first motor in controlling the entire rotating arm to reach a completely horizontal state. Each time, the discharge detection process of one of the power grid devices is completed. Through the cooperation of the positioning post and the positioning groove, the detection and correction process of the inspection robot is achieved.

[0021] The beneficial effects of this invention are:

[0022] 1. This AI-driven method for optimizing the organizational structure of power grid production management, through the deep integration of a global sensing network and edge computing, reduces the data acquisition and analysis latency from hours to milliseconds, solving the problem of response delays caused by traditional reliance on manual inspections and hierarchical reporting. Through a hybrid decision-making mechanism of AI agents and human experts, it forms an agile closed loop of prediction-decision-execution-feedback, ensuring the synergistic evolution of technological change and organizational capabilities.

[0023] 2. In this AI-driven optimization method for the organizational structure of power grid production management, inspection robots are used to complete the inspection tasks of concentrated areas of power grid equipment. With the combination of two track structures, namely the bottom track and the obstacle-crossing track, it can ensure that the inspection robot can move in the presence of various obstacles on the ground and achieve the purpose of equipment inspection over a wide range.

[0024] 3. The inspection robot used in this AI-driven power grid production management organizational structure optimization method can detect leakage current in the ground-laid lines and fixed areas through its end-of-line detection module. It can also flexibly perform partial discharge detection on power grid equipment in different locations and quickly and efficiently locate the monitoring position of the detection module itself. Attached Figure Description

[0025] Figure 1 This is a flowchart of the AI-driven power grid production management organizational structure optimization method of the present invention.

[0026] Figure 2 This is a diagram showing the composition of key deployed equipment used in the AI-driven power grid production management organizational structure optimization method of this invention.

[0027] Figure 3 This is a diagram of the inspection robot and its track connection according to the present invention;

[0028] Figure 4 This is a structural diagram of the track section of the present invention;

[0029] Figure 5 This is a schematic diagram of the inspection robot part of the present invention;

[0030] Figure 6 This is a schematic diagram of the drive mechanism of the inspection robot of the present invention;

[0031] Figure 7 This is a schematic diagram of the structure of the mobile trolley part of the present invention;

[0032] Figure 8 This is a schematic diagram of the detection module of the present invention;

[0033] Figure 9 for Figure 7 Enlarged view of region A in the middle;

[0034] In the diagram: 1. Bottom track; 2. Obstacle-crossing track; 3. Column; 4. Inspection robot; 5. Moving channel; 6. Front protruding plate; 7. Docking platform; 8. Lifting plate; 9. Moving trolley; 10. Drive mechanism; 11. Detection module; 12. Power system; 13. Drive roller; 14. Connecting column; 15. Top plate; 16. Transmission plate; 17. Support wheel; 18. Rear baffle; 19. Front end plate; 20. Partial discharge detector; 21. Electric lifting rod; 22. First motor; 23. Drive shaft; 24. Rotating arm; 25. Second motor; 26. Lead screw; 27. Pressing wheel; 28. Conductive ring; 29. ​​Electrode column; 30. Conductive column; 31. Connecting plate; 32. Conduit; 33. Threaded sleeve; 34. Positioning column; 35. Extension plate; 36. Positioning groove. Detailed Implementation

[0035] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0036] Please see Figures 1 to 9 This invention provides the following technical solution: an artificial intelligence-driven method for optimizing the organizational structure of power grid production management, comprising the following steps:

[0037] S1. Construct a full-domain data perception network and edge computing infrastructure. With multi-source data fusion as the core, establish an intelligent perception system covering the entire process of generation, transmission, transformation, and distribution. Deploy high-precision IoT sensors, inspection drone clusters, and meteorological satellite data receiving stations to form a multimodal data warehouse that is updated in minutes. Deploy edge computing devices at key nodes such as substations and new energy power stations to run lightweight AI models under the federated learning framework.

[0038] S2. Establish intelligent execution terminals to perform automated inspection operations for power grid management. Based on the planning status of the concentrated area of ​​power grid equipment, build tracks and install inspection robots 4 to conduct regular inspections along the preset tracks and collect power information of each power grid device.

[0039] S3. Develop an intelligent decision-making inference engine and knowledge graph system. Based on digital twins and game theory, optimize and construct a dynamic decision-making center. Build a 1:1 virtual mirror system using the power grid physical information model. Integrate the LSTM-GAN hybrid model to predict the load fluctuation and equipment failure probability distribution in the next 72 hours. Use the subject collaborative optimization algorithm to model the interests of power generators, transmission and transformation operators and dispatch centers as an incomplete information game.

[0040] S4. Establish a dynamic and adaptive hybrid decision-making and execution unit, reconstruct the human-machine collaborative task execution mechanism, and divide the response system into multiple levels according to the event level. The AI ​​agent directly dispatches the inspection robot 4 to perform standardized operations, synchronously generates digital work orders and archives them in the blockchain evidence storage system, and synchronously presents a three-dimensional heat map of the fault impact domain, the distribution of spare parts inventory, and the real-time location of the repair team through an augmented reality interface. The AI ​​generates multiple disposal plans, which include success rate estimation and cost accounting.

[0041] S5. Reconstruct the flexible organizational structure and cross-departmental collaboration mechanism, build a matrix dynamic organization, establish an organizational agility assessment system, quantify departmental collaboration effectiveness through the collaboration response index, and directly link it with performance appraisal.

[0042] S6. Build a model sharing platform to implement and verify models, promote automated iterative upgrades of the system, and simultaneously establish a model performance degradation early warning mechanism.

[0043] In this embodiment, an inspection robot 4 used in step S2 above and a track structure applicable to the inspection robot 4 are also provided. The track structure includes a bottom track 1 and an obstacle crossing track 2. Both the bottom track 1 and the obstacle crossing track 2 include a moving channel 5, a front protruding plate 6 and a docking platform 7. The docking platform 7 is located at both ends of the bottom track 1 and the obstacle crossing track 2. The bottom track 1 and the obstacle crossing track 2 are provided with a moving channel 5 inside. The front end of the moving channel 5 is welded with a front protruding plate 6. Columns 3 are welded to the back of both ends of the obstacle crossing track 2.

[0044] Specifically, in this embodiment, both the bottom track 1 and the obstacle crossing track 2 are used to provide movement guidance for the inspection robot 4. The obstacle crossing track 2 needs to be built according to the obstacles on the ground. For example, if there are pipelines or small power grid devices blocking the ground, the bottom track 1 can be extended to one side of the obstacle. Then, the obstacle crossing track 2 is built and supported by the column 3. After the inspection robot 4 moves along the bottom track 1 and moves to one side of the obstacle, it is lifted from the bottom track 1 onto the obstacle crossing track 2 through the docking platform 7. The number of layers and the height of the obstacle crossing track 2 are increased according to the height of the obstacle.

[0045] The inspection robot 4 includes a drive mechanism 10, a mobile trolley 9, and a detection module 11. The drive mechanism 10 includes a lifting plate 8, a connecting column 14, and a top plate 15. The connecting column 14 is welded to the surface of the lifting plate 8, and the top plate 15 is welded to the top of the connecting column 14. A power system 12 is screwed to the surface of the lifting plate 8, and a drive roller 13 is inserted into the rear end of the power system 12. A transmission plate 16 is welded to the side of the top plate 15. An electric lifting rod 21 is welded to the top of the mobile trolley 9, and the top of the electric lifting rod 21 is screwed to the bottom of the transmission plate 16.

[0046] By activating the electric lifting rod 21, the transmission plate 16 and the top plate 15 are driven to move up and down. The top plate 15 drives the bottom lifting plate 8 to rise or fall through the connecting column 14. The lifting plate 8 raises the power system 12 and the drive roller 13 until the drive roller 13 is lifted from the surface of the docking platform 7 of the bottom track 1 to the bottom surface of the docking platform 7 of the upper obstacle crossing track 2. The electric lifting rod 21 provides pressure for the drive roller 13 to press against the docking platform 7 of the obstacle crossing track 2. At this time, the power system 12 is controlled to run in reverse so that the drive roller 13 rolls along the docking platform 7 of the obstacle crossing track 2 to the inside of the moving channel 5 of the obstacle crossing track 2. The electric telescopic rod is then controlled to retract, which is used to lift the entire inspection robot 4.

[0047] A front end plate 19 and a rear end plate are welded to the surface of the mobile trolley 9. An extension plate 35 is integrally formed at the end of the rear end plate, and a positioning groove 36 is formed at the end of the extension plate 35. The detection module 11 includes a first motor 22, a drive shaft 23, a rotating arm 24, a second motor 25, a pressing wheel 27, a conductive post 30, and an electrode post 29. The first motor 22 is screwed onto the surface of the rear end plate. The output end of the first motor 22 is inserted into the drive shaft 23. The end of the drive shaft 23 is screwed into the rotating arm 24. The end of the rotating arm 24 is axially connected to the pressing wheel 27. A conductive ring 28 is embedded on the side of the pressing wheel 27, and a conductive rod is inserted in the middle of the pressing wheel 27. The conductive post 30 has an electrode post 29 installed at its front end and a connecting plate 31 installed at its rear end. The end of the connecting plate 31 has a threaded sleeve 33. A second motor 25 is screwed onto the surface of the rotating arm 24. A lead screw 26 is inserted into the output end of the second motor 25 and passes through the inside of the threaded sleeve. A positioning post 34 and a conduit 32 are welded onto the back of the connecting plate 31. A partial discharge detector 20 is screwed onto the surface of the moving trolley 9. The surface of the conductive post 30 is partially connected to the conductive ring 28 through a circuit. The conductive post 30 is partially connected to the partial discharge detector 20 after passing through the conduit 32 with an electric wire.

[0048] After the first motor 22 is started, the drive shaft 23 drives the rotating arm 24 at the end to rotate. The rotating arm 24 controls the first motor 22, the pressing wheel 27, the conductive post 30 and the connecting plate 31 to rotate synchronously. The pressing wheel 27 is used to press on the ground and make contact with the surface of the cable line it passes over, in order to detect and process the leakage status of the ground cable, and transmit the current signal to the partial discharge detector 20 through the conductive post 30 and the wire at the rear end of the conductive post 30.

[0049] After the rotating arm 24 is rotated by the first motor 22, the second motor 25 is started. The second motor 25 drives the lead screw 26 to rotate. The lead screw 26, in conjunction with the threaded sleeve 33, drives the entire connecting plate 31 to move in translation. The connecting plate 31 pulls the conductive post 30 at one end to move in extension and retraction at the middle position of the pressing wheel 27, so that the electrode post 29 can abut against the surface of the adjacent power grid equipment. The partial discharge detector 20 at the rear end detects the current state of the contacted power grid equipment surface through the wire, the conductive post 30 and the electrode post 29.

[0050] By controlling the first motor 22 to drive the rotating arm 24 to a horizontal state, the second motor 25 controls the connecting plate 31 to move towards the rear end until the positioning post 34 is embedded into the interior of the positioning groove 36. The interior of the positioning groove 36 is conical. By the positioning post 34 abutting against the innermost area of ​​the positioning groove 36, the entire connecting plate 31 is positioned, thereby assisting the first motor 22 in controlling the entire rotating arm 24 to reach a completely horizontal state. Each time the discharge detection process of one of the power grid devices is completed, the detection and correction process of the inspection robot 4 is achieved through the cooperation of the positioning post 34 and the positioning groove 36.

[0051] The foregoing has shown and described the basic principles and main features of the present invention and its advantages. It will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention.

[0052] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An artificial intelligence driven power grid production management organizational structure optimization method, characterized in that, Comprise the following steps: S1, construct global data perception network and edge computing infrastructure, multi-source data fusion as the core, to establish a full link intelligent sensing system covering the generation, transmission, transformation, distribution; S2, establish intelligent execution terminal, automatic inspection operation for power grid management, according to the power grid equipment centralized area planning state to build track, and install inspection robot along the preset track for regular inspection, and collect the power information of each power grid equipment; S3, develop intelligent decision deduction engine and knowledge graph system, based on digital twin and game theory optimization to build dynamic decision center; S4, establish dynamic adaptive hybrid decision execution unit, reconfigure the task execution mechanism of man-machine cooperation, according to the event level division multi-level response system; S5, reconfigure the elastic organizational structure and cross-department cooperation mechanism, build matrix dynamic organization, establish organization agility evaluation system, quantify the department cooperation efficiency through collaborative response index, and directly link with performance evaluation; S6, build model sharing platform, implement verification on the model, promote system automatic iteration and upgrade, and simultaneously establish model performance attenuation early warning mechanism; In step S3, a 1:1 virtual mirror system is built by using power grid physical information model, an LSTM-GAN hybrid model is integrated to predict future 72-hour load fluctuation and equipment failure probability distribution, and the interests of power suppliers, transmission and transformation operators and dispatching centers are modeled as incomplete information game by means of subject collaborative optimization algorithm; The track comprises a bottom track and an obstacle track, and the bottom track and the obstacle track are both used to provide the function of moving guide for the inspection robot, and the obstacle track needs to be built according to the obstacles existing on the ground. After the ground is blocked by pipelines or small power grid equipment, the bottom track is extended to one side of the obstacle, and then the obstacle track is built and supported by relying on the stand. After the inspection robot moves along the bottom track and moves to one side of the obstacle, it is lifted from the bottom track to the obstacle track by the docking platform, and the number of layers and the height of the obstacle track are increased according to the height of the obstacle.

2. The artificial intelligence driven power grid production management organizational structure optimization method of claim 1, wherein: In step S1, high-precision IoT sensors, inspection unmanned aerial vehicle clusters and meteorological satellite data receiving stations are deployed to form a minute-level update multi-modal data warehouse. Edge computing devices are deployed at key nodes of substations and new energy stations to run lightweight AI models under the federated learning framework.

3. The artificial intelligence driven power grid production management organizational structure optimization method of claim 1, wherein: In step S4, the AI agent directly schedules the inspection robot to execute standardized operations, synchronously generates digital work orders and archives them in the blockchain storage system, synchronously presents the three-dimensional heat map of the fault influence domain, the distribution of spare parts and spare parts inventory, and the real-time positioning of the repair team through the augmented reality interface, and the AI generates multiple disposal schemes including success rate estimation and cost accounting.

4. The artificial intelligence driven power grid production management organizational structure optimization method of claim 1, wherein: The power lifting rod is started to drive the transmission plate and the top plate to move up and down. The top plate drives the lifting plate at the bottom to lift or drop through the connecting column. The power system and the driving roller are lifted through the lifting of the lifting plate until the driving roller is lifted from the surface of the docking platform of the bottom track to the bottom surface of the docking platform of the upper obstacle track.

5. The artificial intelligence driven power grid production management organizational structure optimization method of claim 4, wherein: The driving roller is provided with pressure by the electric lifting rod, and the power system is reversely controlled to drive the driving roller to roll along the docking platform of the barrier track to the inside of the moving channel of the barrier track, and the electric telescopic rod is controlled to retract to lift the whole inspection robot.

6. The artificial intelligence driven power grid production management organizational structure optimization method of claim 4, wherein, The ground cable line leakage monitoring is also included: after the first motor is started, the rotating arm at the end is rotated by the driving shaft, the rotating arm controls the first motor, the pressing wheel, the conductive column and the connecting plate to rotate synchronously, the pressing wheel is used to press on the ground and contact the surface of the cable line passed to detect the leakage state of the ground cable and transmit the current signal to the partial discharge detector through the conductive column and the wire at the rear end of the conductive column.

7. The artificial intelligence driven power grid production management organizational structure optimization method of claim 5, wherein, The detection and correction process of the inspection robot is also included: the rotating arm is driven by the first motor to reach the horizontal state, the connecting plate is moved towards the rear end by the second motor until the positioning column is embedded in the inside of the positioning groove, the inside of the positioning groove is in a conical state, the positioning column abuts against the innermost area of the positioning groove to position the whole connecting plate, the first motor controls the rotating arm to reach the completely horizontal state, and the detection and correction of the inspection robot are achieved through the cooperation of the positioning column and the positioning groove after the discharge detection process of each power grid equipment is completed.

Images