Transformer substation intelligent inspection system based on spatial vision accurate positioning technology

Abstract

The invention belongs to the technical field of transformer substation inspection, and particularly relates to a transformer substation intelligent inspection system based on a spatial vision precise positioning technology, which comprises a spatial positioning module used for realizing centimeter-level positioning of inspection equipment through a multi-source fusion positioning algorithm; the inspection execution module is used for executing dynamic path planning and multi-modal data acquisition; the data analysis module is used for equipment defect identification and predictive maintenance; the positioning accuracy of the inspection system is high: the positioning error in a dynamic environment is less than or equal to 3cm, which is superior to that in the prior art (10cm level); the detection efficiency is high, the defect identification speed is increased by 40%, and the false alarm rate is reduced to 2% or below; the adaptability is wide, and stable operation in the temperature range of-20 DEG C to 50 DEG C and in rainy and foggy weather is supported. Through multi-sensor cooperative positioning, dynamic path planning and a lightweight AI algorithm, the problems of low inspection precision and insufficient efficiency in a complex environment are solved, and the intelligent level of operation and maintenance of power equipment is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of substation inspection technology, and in particular to a substation intelligent inspection system based on spatial vision precise positioning technology. Background Art

[0002] A substation is a place in the power system where voltage and current are transformed, electrical energy is received, and electrical energy is distributed. The substation in a power plant is a step-up substation, whose function is to step up the voltage of the electrical energy generated by the generator and feed it into the high-voltage power grid. During the use of the substation, staff must conduct daily inspections of the substation to prevent safety hazards.

[0003] At present, with the development of the economy, the number of substations is increasing, and the equipment is also increasing. The operation team operators are centrally managed and controlled, and all unmanned substations have been realized. However, daily equipment inspections are still traditional on-site inspections, manual recording and filling out defect reports, and re-entering the report data into the computer for storage. Manual operations are performed to synthesize equipment inspections, weekly, monthly and annual reports on equipment defects, etc.

[0004] Spatial visual precision positioning technology is a technology that uses visual sensors (such as cameras, lidar, etc.) as the core, and achieves high-precision position and posture determination of target objects or devices in three-dimensional space through processes such as image acquisition, feature extraction and matching, and positioning algorithm calculation.

[0005] However, the existing substation inspection system is relatively simple. The system cannot conduct comprehensive inspections of substations, and cannot analyze and process the inspection data in a timely manner, thus affecting the application effect of the inspection system. Summary of the Invention

[0006] 1. Technical problems to be solved

[0007] The purpose of the present invention is to solve the problem that the system in the existing technology cannot comprehensively inspect the substation and cannot analyze and process the inspection data in time, and to propose an intelligent substation inspection system based on spatial vision precise positioning technology.

[0008] 2. Technical solution

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] The substation intelligent inspection system based on spatial visual precision positioning technology includes a spatial positioning module, which is used to achieve centimeter-level positioning of inspection equipment through a multi-source fusion positioning algorithm; an inspection execution module, which is used to perform dynamic path planning and multimodal data collection; and a data analysis module, which is used to identify equipment defects and predictive maintenance.

[0011] Preferably, the spatial positioning module includes: a binocular camera, a UWB base station and a lidar, which dynamically fuses visual SLAM feature points, UWB ranging data and laser point cloud through a Kalman filter, with a positioning error of ≤3cm; a dynamic environment modeling unit, which updates the three-dimensional map of the substation in real time to cope with equipment occlusion and lighting changes.

[0012] Preferably, the inspection execution module includes an autonomous navigation unit that uses a reinforcement learning (DQN) algorithm to generate a dynamic inspection path; a multimodal data acquisition unit that integrates an infrared thermal imager, a high-definition camera, a partial discharge sensor and a voiceprint sensor.

[0013] Preferably, the data analysis module includes an improved YOLOv7 model, which is compressed by knowledge distillation technology and deployed on embedded devices to detect equipment defects in real time; an LSTM neural network to predict equipment life and generate maintenance recommendations.

[0014] The present invention also proposes a substation digital twin system, which is based on the above-mentioned spatial positioning module and data analysis module, including a real-time three-dimensional modeling unit, which constructs a dynamic twin of the substation through lidar and visual SLAM; an AR interaction unit, which supports operation and maintenance personnel to view the internal parameters of the equipment through AR glasses and remotely mark fault points.

[0015] The present invention also proposes a multi-robot collaborative inspection path planning method, which is applied to the above-mentioned system. It includes a quantum annealing algorithm for solving the NP-hard problem of multi-device task allocation and collision avoidance; and an energy consumption optimization module for dynamically adjusting the inspection path according to the device battery status and task priority.

[0016] Preferably, in a scenario with more than 20 inspection devices, the quantum annealing algorithm takes ≤5 seconds for path planning and reduces energy consumption by 50%.

[0017] The present invention also proposes a method for storing inspection data, which is applied to the above-mentioned system, including a data chain unit, which generates a unique hash value from equipment defect images and sensor data and writes it into the blockchain; a multi-party collaborative verification unit, which supports the synchronous verification of data by nodes of power grid companies, insurance institutions and regulatory departments.

[0018] The present invention also proposes a method for deploying an inspection system in extreme environments, including loading a pulse heating device on the drone rotor to prevent icing in rainy and snowy weather; and using ATEX-certified explosion-proof inspection robots and intrinsically safe sensors in the explosion-proof area of ​​a chemical plant.

[0019] 3. Beneficial effects

[0020] Compared with the prior art, the advantages of the present invention are:

[0021] (1) In the present invention, the problems of low inspection accuracy and insufficient efficiency in complex environments are solved through multi-sensor collaborative positioning, dynamic path planning and lightweight AI algorithm, which significantly improves the intelligent level of power equipment operation and maintenance.

[0022] (2) In the present invention, the inspection system has high positioning accuracy: the positioning error in a dynamic environment is ≤3cm, which is better than the existing technology (10cm level); the detection efficiency is high: the defect recognition speed is increased by 40%, and the false alarm rate is reduced to less than 2%; the adaptability is wide: it supports stable operation in the temperature range of -20℃ to 50℃ and rainy and foggy weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic diagram of the composition of the substation intelligent inspection system based on spatial vision precise positioning technology proposed in this invention. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0025] Example 1:

[0026] Reference Figure 1 The intelligent substation inspection system based on spatial visual precision positioning technology includes a spatial positioning module for achieving centimeter-level positioning of inspection equipment through a multi-source fusion positioning algorithm. The spatial positioning module includes a binocular camera, a UWB base station, and a lidar. The module dynamically fuses visual SLAM feature points, UWB ranging data, and laser point clouds through a Kalman filter, achieving a positioning error of ≤3cm. A dynamic environment modeling unit updates the substation's three-dimensional map in real time to cope with equipment occlusion and lighting changes.

[0027] An inspection execution module, which is used to perform dynamic path planning and multimodal data collection. The inspection execution module includes an autonomous navigation unit that uses a reinforcement learning (DQN) algorithm to generate a dynamic inspection path; a multimodal data collection unit that integrates an infrared thermal imager, a high-definition camera, a partial discharge sensor, and a voiceprint sensor;

[0028] The data analysis module is used for equipment defect identification and predictive maintenance. The data analysis module includes an improved YOLOv7 model, which is compressed using knowledge distillation technology and deployed on embedded devices to detect equipment defects in real time; and an LSTM neural network to predict equipment life and generate maintenance recommendations.

[0029] In this embodiment, through multi-sensor collaborative positioning, dynamic path planning and lightweight AI algorithm, the problems of low inspection accuracy and insufficient efficiency in complex environments are solved, and the intelligence level of power equipment operation and maintenance is significantly improved.

[0030] In this embodiment, the inspection system has high positioning accuracy: the positioning error in a dynamic environment is ≤3cm, which is better than the existing technology (10cm level); the detection efficiency is high: the defect recognition speed is increased by 40%, and the false alarm rate is reduced to less than 2%; it has wide adaptability: it supports stable operation in a temperature range of -20℃ to 50℃ and rainy and foggy weather.

[0031] Example 2:

[0032] It has the implementation content of the above embodiment, wherein, for the specific implementation of the above embodiment, reference can be made to the above description, and the embodiment here will not be repeated in detail; and in the embodiment of the present application, it is different from the above embodiment in that:

[0033] In this embodiment, a digital twin system of a substation is based on the above-mentioned spatial positioning module and data analysis module, including a real-time three-dimensional modeling unit, which constructs a dynamic twin of the substation through lidar and visual SLAM; an AR interaction unit, which supports operation and maintenance personnel to view the internal parameters of the equipment through AR glasses and remotely mark fault points.

[0034] Example 3:

[0035] It has the implementation content of the above embodiment, wherein, for the specific implementation of the above embodiment, reference can be made to the above description, and the embodiment here will not be repeated in detail; and in the embodiment of the present application, it is different from the above embodiment in that:

[0036] In this embodiment, a multi-robot collaborative inspection path planning method is applied to the above-mentioned system, including a quantum annealing algorithm for solving the NP-hard problem of multi-device task allocation and collision avoidance; and an energy consumption optimization module for dynamically adjusting the inspection path based on the device battery status and task priority. In a scenario with more than 20 inspection devices, the quantum annealing algorithm achieves path planning time of ≤5 seconds and reduces energy consumption by 50%.

[0037] Example 4:

[0038] It has the implementation content of the above embodiment, wherein, for the specific implementation of the above embodiment, reference can be made to the above description, and the embodiment here will not be repeated in detail; and in the embodiment of the present application, it is different from the above embodiment in that:

[0039] In this embodiment, a method for storing inspection data is applied to the above-mentioned system, including a data chain unit, which generates a unique hash value from the equipment defect image and sensor data and writes it into the blockchain; a multi-party collaborative verification unit, which supports the synchronous verification of data by nodes of power grid companies, insurance institutions and regulatory departments.

[0040] Example 5:

[0041] It has the implementation content of the above embodiment, wherein, for the specific implementation of the above embodiment, reference can be made to the above description, and the embodiment here will not be repeated in detail; and in the embodiment of the present application, it is different from the above embodiment in that:

[0042] In this embodiment, a method for deploying an inspection system in extreme environments includes loading a pulse heating device on the drone rotor to prevent icing in rainy and snowy weather; and using ATEX-certified explosion-proof inspection robots and intrinsically safe sensors in the explosion-proof area of ​​a chemical plant.

[0043] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. The intelligent inspection system for substations based on spatial vision precise positioning technology is characterized by: It includes a spatial positioning module, which is used to achieve centimeter-level positioning of inspection equipment through a multi-source fusion positioning algorithm; an inspection execution module, which is used to perform dynamic path planning and multimodal data collection; Data analysis module for equipment defect identification and predictive maintenance.

2. The substation intelligent inspection system based on spatial vision precise positioning technology according to claim 1 is characterized in that: The spatial positioning module includes: a binocular camera, a UWB base station and a lidar, which dynamically fuses visual SLAM feature points, UWB ranging data and laser point clouds through a Kalman filter, with a positioning error of ≤3cm; a dynamic environment modeling unit that updates the substation 3D map in real time to cope with equipment occlusion and lighting changes.

3. The substation intelligent inspection system based on spatial vision precise positioning technology according to claim 1 is characterized in that: The inspection execution module includes an autonomous navigation unit that uses a reinforcement learning (DQN) algorithm to generate a dynamic inspection path; a multimodal data acquisition unit that integrates an infrared thermal imager, a high-definition camera, a partial discharge sensor, and a voiceprint sensor.

4. The substation intelligent inspection system based on spatial vision precise positioning technology according to claim 1 is characterized in that: The data analysis module includes an improved YOLOv7 model, which is compressed using knowledge distillation technology and deployed on embedded devices to detect equipment defects in real time; and an LSTM neural network to predict equipment life and generate maintenance recommendations.

5. A substation digital twin system, characterized in that: Based on the spatial positioning module and data analysis module described in claim 1, it includes a real-time three-dimensional modeling unit, which builds a dynamic twin of the substation through lidar and visual SLAM; an AR interaction unit, which supports operation and maintenance personnel to view the internal parameters of the equipment through AR glasses and remotely mark fault points.

6. A multi-robot collaborative inspection path planning method, characterized in that: The system applied to claim 1 includes a quantum annealing algorithm for solving the NP-hard problem of multi-device task allocation and collision avoidance; and an energy consumption optimization module for dynamically adjusting the inspection path according to the device battery status and task priority.

7. The inspection path planning method according to claim 6, characterized in that: In scenarios with more than 20 inspection devices, the quantum annealing algorithm takes ≤5 seconds to plan paths, reducing energy consumption by 50%.

8. A method for storing inspection data, characterized in that: The system as claimed in claim 1 includes a data uploading unit that generates a unique hash value from the device defect image and sensor data and writes the hash value into the blockchain; The multi-party collaborative verification unit supports the simultaneous data verification of power grid companies, insurance institutions and regulatory authorities.

9. A method for deploying an inspection system in an extreme environment, characterized in that: This includes installing pulse heating devices on drone rotors to prevent icing in rainy and snowy weather; and using ATEX-certified explosion-proof inspection robots and intrinsically safe sensors in explosion-proof areas of chemical plants.

Images