CN-121791440-B - Autonomous inspection method and system for substation inspection robot

CN121791440BCN 121791440 BCN121791440 BCN 121791440BCN-121791440-B

Abstract

The invention relates to the technical field of substation inspection robots and discloses an autonomous inspection method and system for a substation inspection robot. The method comprises the steps of collecting current environmental parameters of a transformer substation and real-time state information of a patrol robot to construct a patrol environment model, analyzing historical patrol records, classifying historical patrol strategies into a plurality of experience levels to construct a multi-layer experience pool, calculating intelligent patrol strategies required for achieving the target patrol states according to target patrol states preset in the patrol environment model, carrying out real-time correction on the intelligent patrol strategies and the experience patrol strategies by adopting a two-way coordination mechanism to respond to dynamic changes of the transformer substation environment to generate final patrol strategies, executing control instructions based on the final patrol strategies to complete patrol tasks and record actual patrol states, comparing deviation reasons of the actual patrol states and the target patrol states, adjusting experience contents in the multi-layer experience pool according to the deviation reasons, and optimizing experience data.

Inventors

LI ZHENGYU
WANG YU
YAN LITING
GUO LIN
LIU XUFEI
ZHAO YONGLI

Assignees

国网山西省电力有限公司超高压变电分公司

Dates

Publication Date: 20260508
Application Date: 20260305

Claims (10)

1. An autonomous inspection method for an inspection robot of a transformer substation is characterized by comprising the following steps: collecting current environmental parameters of a transformer substation and real-time state information of a patrol robot, and constructing a patrol environment model; analyzing the historical inspection records, classifying the historical inspection strategies into a plurality of experience levels according to the confidence level of the historical inspection strategies and the difference between the historical inspection results and the expected targets, and constructing a multi-layer experience pool; according to a target inspection state preset in the inspection environment model, calculating an intelligent inspection strategy required for achieving the target inspection state, and simultaneously extracting an experience inspection strategy from a multi-layer experience pool based on strategy confidence and state matching degree; The intelligent inspection strategy and the experience inspection strategy are corrected in real time by adopting a double-channel coordination mechanism so as to respond to the dynamic change of the transformer substation environment and generate a final inspection strategy; executing a control instruction based on a final inspection strategy, completing an inspection task and recording an actual inspection state; And comparing the deviation reasons of the actual inspection state and the target inspection state, adjusting experience contents in the multi-layer experience pool according to the deviation reasons, and optimizing experience data.
2. The autonomous inspection method of the substation inspection robot according to claim 1, wherein the step of collecting current environmental parameters of the substation and real-time state information of the inspection robot and constructing an inspection environment model comprises: acquiring environmental parameter data by using a multi-source sensor array deployed in a transformer substation, wherein the multi-source sensor array comprises a temperature sensor, a humidity sensor, a vibration sensor and an electromagnetic sensor; Capturing an image sequence of a patrol area through a visual perception system of the patrol robot, and analyzing the equipment state and the robot pose by adopting an image recognition technology; And carrying out space-time synchronous fusion on the multisource sensor data and the image sequence information, and constructing a dynamically updated digital twin model serving as a patrol environment model.
3. The autonomous inspection method of a substation inspection robot according to claim 2, wherein the step of analyzing the historical inspection records and classifying the historical inspection policies into a plurality of experience levels according to the confidence level of the historical inspection policies and the difference between the historical inspection results and the expected targets, and constructing a multi-layer experience pool comprises: Extracting an execution record of the inspection strategy from a historical database, wherein the execution record comprises strategy parameters, inspection results and time stamps; calculating a success rate index and a deviation index of each historical inspection strategy by using a statistical analysis method; and automatically dividing the historical inspection strategy into a core experience layer, a standard experience layer and a basic experience layer according to the threshold ranges of the success rate index and the deviation index to form a layering experience pool.
4. The autonomous inspection method of a substation inspection robot according to claim 3, wherein the step of calculating an intelligent inspection policy required for achieving a target inspection state according to a target inspection state preset in the inspection environment model, and extracting an experience inspection policy from a multi-layer experience pool based on a policy confidence and a state matching degree comprises: Inputting the current environmental parameters and the robot state into an intelligent decision system, and generating an optimized inspection strategy by taking the target inspection state as a constraint condition; Executing pattern matching operation in the multi-layer experience pool, and screening out a historical strategy case with high similarity with the current target inspection state; And combining the optimized inspection strategy with the historical strategy case through a strategy synthesis technology to generate a preliminary strategy set.
5. The autonomous inspection method of a substation inspection robot according to claim 4, wherein the step of generating a final inspection strategy by correcting the intelligent inspection strategy and the empirical inspection strategy in real time by using a two-way coordination mechanism in response to a dynamic change of a substation environment comprises: continuously monitoring environmental parameters of a transformer substation and operation indexes of a robot, and setting an abnormality detection threshold; When the monitoring data exceeds an abnormal detection threshold, automatically switching to an experience inspection strategy channel, and selecting an adaptation strategy from a core experience layer; and in the normal monitoring range, keeping the channel of the intelligent patrol strategy dominant, and periodically evaluating the applicability of the experience strategy to realize dynamic strategy adjustment.
6. The autonomous inspection method of a substation inspection robot according to claim 5, wherein the step of executing a control command based on a final inspection strategy, completing an inspection task and recording an actual inspection state comprises: analyzing a final inspection strategy into a specific action instruction sequence, wherein the specific action instruction sequence comprises a moving path, a detection point and a data acquisition command; the robot is controlled to execute the inspection task according to the instruction sequence, and equipment state acquisition and image recording are completed at each detection point; And integrating the sensor data and visual feedback of the robot to generate an actual inspection status report, wherein the actual inspection status report comprises abnormal equipment readings and environmental changes.
7. The autonomous inspection method of a substation inspection robot according to claim 6, wherein the comparing the deviation reasons of the actual inspection state and the target inspection state, adjusting the experience content in the multi-layer experience pool according to the deviation reasons, and optimizing the experience data comprises: calculating the difference value between the actual inspection time and the target inspection time and the difference value of the task completion degree; Analyzing the generation root of the difference value and identifying key influence factors; The strategy content in the multi-layer experience pool is revised according to the key influencing factors, and the weights of the experience layers and the case data are adjusted.
8. The autonomous inspection method of a substation inspection robot according to claim 4, wherein in the step of calculating the intelligent inspection strategy required to achieve the target inspection state, the generation of the intelligent inspection strategy focuses on resource optimization, including minimizing robot energy consumption, optimizing inspection path length, and reducing sensor usage frequency; In the step of extracting the experience inspection strategy from the multi-layer experience pool, the selection of the experience inspection strategy is based on case base reasoning, similar scenes are searched from historical successful cases, and the final strategy is determined by combining real-time matching degree evaluation.
9. The autonomous inspection method of a substation inspection robot according to claim 5, wherein in the step of correcting the intelligent inspection strategy and the empirical inspection strategy in real time by adopting a dual-path coordination mechanism, the dual-path coordination mechanism has a fault tolerance function, and when a main strategy channel fails, a standby strategy channel is immediately started to ensure continuity of the inspection process.
10. An autonomous inspection system for an inspection robot of a transformer substation, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor, when executing the computer program, implements the steps of the autonomous inspection method for an inspection robot of a transformer substation according to any one of the preceding claims 1 to 9.

Description

Autonomous inspection method and system for substation inspection robot Technical Field The invention relates to the technical field of substation inspection robots, in particular to an autonomous inspection method and system for a substation inspection robot. Background In a power system, a transformer substation is a key link for guaranteeing stable power transmission and distribution, and stable operation of equipment is critical to the safety of the whole power grid. Traditionally, inspection of a transformer substation is mainly performed manually, but the method has a plurality of difficulties which are difficult to overcome. The manual inspection requires operation and maintenance personnel to rush to the transformer substation site regularly, which not only consumes a great deal of time, but also involves traffic cost. If the transformer substation is located in a remote area, the traffic is inconvenient, and the travel difficulty and cost of operation and maintenance personnel can be further increased. During the trip to the scene, traffic jams, bad weather and other conditions can be faced, and the timeliness of inspection is affected. Moreover, the manual inspection has personal safety risks, high voltage, strong current and other dangerous factors exist in the transformer substation, and once equipment faults or anomalies occur, operation and maintenance personnel can be directly exposed in the danger, so that life safety is threatened. The traditional manual inspection means is single and mainly depends on the sense organs of operation and maintenance personnel and simple tools. The mode is greatly influenced by factors such as weather, environment and the like, and under severe weather conditions such as heavy rain, heavy snow, thick fog and the like, the observation and detection work of operation and maintenance personnel can be seriously hindered, and the equipment state is difficult to accurately judge. The accuracy and the efficiency of manual inspection rely on personnel quality and experience to a great extent, and the professional level and the working state of different operation and maintenance personnel have differences, so that the deviation of inspection results is easy to cause, equipment defects or anomalies cannot be found in time, serious faults are further developed, and the stable operation of a power grid is affected. With the rapid development of economy, the power grid scale is continuously enlarged, the number of substations is increased, and the complexity of equipment is increased. The inspection workload is greatly increased, and the number of professional operation and maintenance personnel is relatively slow to increase, so that the contradiction between personnel deficiency and the increase of the inspection workload is increasingly prominent. The manual inspection efficiency is low, and the requirement of the rapid development of the power grid is difficult to meet. In order to solve the problem of traditional manual inspection, a transformer substation inspection robot is generated, but the existing robot inspection technology still has a plurality of limitations. The robot partially adopting the magnetic navigation inspection method needs to lay a magnetic wire on the ground and set a necessary electronic tag on the inspection path so as to send an instruction to the inspection robot to coordinate the inspection robot to execute the action. The method is high in early-stage paving cost, complex in construction and troublesome in later-stage maintenance, and once a magnet wire or an electronic tag breaks down, the normal operation of the robot can be influenced. Moreover, the detection form of the robot is single, and only simple tasks can be completed, so that the detection requirement of all-around and multiple parameters of substation equipment is difficult to meet. In the aspect of obstacle avoidance capability, the robots are poor in performance, and when sudden obstacles or environmental changes are encountered, the robots cannot flexibly cope with the obstacles or the environmental changes, and the inspection interruption and even the damage of the robots can be caused. The working mode of part of single inspection robots also has obvious defects. In a large-area medium-high voltage transformer substation, a plurality of areas need to be inspected, the inspection efficiency of a single robot is low, and the comprehensive inspection task is difficult to complete within a specified time. In addition, once the robot fails in the inspection process, the remaining inspection tasks cannot be completed continuously, so that the inspection work is interrupted, and manual intervention is needed. Most of the existing inspection robots cannot effectively inspect complex magnetic field environments of transformer substations. The presence of various electrical devices within a substation can create complex magnetic fields, and magnetic field anomalies can be indicativ