Booster station inspection machine and remote operation robot system

Abstract

The invention provides a booster station inspection machine and a remote operation robot system. The inspection machine comprises a mobile platform, a multi-source sensor module, a data processing module, a communication module, an autonomous navigation module and a mechanical interface. The robot system is carried on the inspection machine through a mechanical interface and comprises a remote operation end, a mechanical arm, a force feedback control terminal, a three-dimensional vision module and a safety control module. According to the invention, the mobile platform of the all-terrain self-adaptive chassis is adopted, the multi-source sensor module collects multi-dimensional data, the data is analyzed by using a fault diagnosis and dynamic early warning model based on multi-source data fusion in the data processing module, and the autonomous navigation module is combined to accurately position and plan a path. The remote operation robot system carrying the high-precision mechanical arm realizes remote refined operation; finally, automatic inspection, remote operation and real-time monitoring of booster station equipment are realized, and the inspection efficiency, the operation precision and the safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment inspection, and in particular to a booster station inspection machine and a remote operation robot system. Background Art

[0002] As electricity demand continues to grow, the scale of power systems is becoming increasingly large. As a key link in power transmission and distribution, booster stations are crucial for their safe and stable operation. In the early days, booster stations relied primarily on manual inspections, with staff regularly visiting the site to check each device individually and record its operating parameters and status. However, manual inspections are inefficient, unable to monitor equipment frequently, and are easily affected by staff expertise, work status, and environmental factors, leading to the risk of missing potential faults. In recent years, with the rapid development of robotics, sensor, communication, and artificial intelligence technologies, power equipment inspection has gradually moved towards intelligence and automation. Intelligent inspection robots have begun to be used in booster stations. They can move autonomously along preset routes and use various sensors on board to collect data from equipment. This has improved inspection efficiency and accuracy to a certain extent, and can also reduce the impact of harsh environments on inspection personnel. However, although the existing inspection system is equipped with a variety of sensors, the data fusion effect of different types of sensors is poor, making it difficult to fully utilize the comprehensive analysis advantages of multi-source data; in terms of fault diagnosis, it often relies on fixed parameter models and lacks the ability to adapt to the dynamic changes in the equipment's operating status, which can easily lead to misjudgment or omission of faults and the inability to detect potential equipment problems in a timely and accurate manner; at the same time, for equipment that requires remote operation, the existing remote operation robot system has inaccurate feedback, and operators cannot perceive the operation results in time, affecting the accuracy and efficiency of the operation; in terms of safety control, there is a lack of a complete multiple protection mechanism. When the operation trajectory exceeds the safety range, the braking response is not fast enough, making it difficult to effectively ensure the safety of equipment and personnel, and posing a large safety risk; Therefore, there is an urgent need in the art for a booster station inspection machine and a remotely operated robot system to solve the above problems. Summary of the Invention

[0003] The present invention provides a booster station inspection machine and a remote-operated robot system, aiming to solve the problems of low inspection efficiency, high labor costs, and great safety risks in traditional booster stations, as well as the single function and insufficient operation accuracy of existing intelligent inspection technologies; by adopting a mobile platform with an all-terrain adaptive chassis and a multi-source sensor module to collect multi-dimensional data, utilizing the fault diagnosis and dynamic early warning model based on multi-source data fusion in the data processing module to analyze the data, combining the precise positioning and path planning of the autonomous navigation module, and the remote-operated robot system equipped with a high-precision mechanical arm to realize remote refined operation; ultimately, realizing automated inspection, remote operation and real-time monitoring of booster station equipment, thereby improving inspection efficiency, operation accuracy and safety.

[0004] In one aspect, the present invention provides a booster station inspection machine, comprising: The mobile platform adopts an all-terrain adaptive chassis to adapt to the complex terrain environment of the booster station; A multi-source sensor module, which is mounted on the mobile platform and is used to synchronously collect multi-dimensional data from the device; A data processing module, which is mounted on the mobile platform and connected to the multi-source sensor module, has a built-in fault diagnosis model and a dynamic early warning model based on multi-source data fusion, determines equipment abnormalities through multi-dimensional data entropy fusion, generates graded early warning signals based on historical diagnosis results, and generates diagnosis results and early warning results; A communication module, which is mounted on the mobile platform and supports real-time data transmission with a remote operation terminal via 5G / 4G, enabling rapid and stable transmission of diagnostic and warning results; An autonomous navigation module, which is mounted on the mobile platform and integrates visual SLAM and path planning algorithms to achieve positioning and real-time path planning; The mechanical interface is provided on the mobile platform and is used to carry the remote-operated robot, providing a carrying platform for the remote-operated robot.

[0005] According to a booster station inspection machine provided by the present invention, the fault diagnosis model is:

[0006] in, For fault information; is the number of sensors; is the real-time measurement value of the i-th sensor at time t; and During normal operation The mean and standard deviation of historical data; is the entropy weight coefficient, which is used to measure the weight of sensor i in the fault diagnosis model. ,in is the data information entropy of sensor i, which is used to reflect the uncertainty or confusion of the data collected by sensor i; Collect data for sensor i Probability of occurrence; is the maximum value of information entropy; It is a standardized activation function used to normalize the processed sensor data; is the time attenuation factor, which is used to describe the gradual weakening of the impact of data on the current fault diagnosis structure over time. The thermal time constant of the device is used to reflect the speed of temperature change during the operation of the device; when When , the output diagnosis result is fault, where for The historical standard deviation.

[0007] According to a booster station inspection machine provided by the present invention, the dynamic early warning model is:

[0008] in, is the adaptive threshold, which represents the warning threshold at time t and is used to determine whether to trigger the warning; The 99th percentile of the diagnostic value within the window period is used to measure the high end of the data distribution, reflecting the high level of diagnostic values in most cases, and avoiding false triggering of warnings due to occasional large diagnostic values; is the sliding window size, and is the sampling period. The sliding window method takes into account the changes in recent data, so that the warning threshold can adapt to the dynamic changes in the equipment operation status. Representative Moment The fault diagnosis value, that is, the equipment status evaluation value at different times calculated by the fault diagnosis algorithm, is used to analyze the changing trend of the equipment status over time; Related adjustment items are used to fine-tune the warning threshold; When the output diagnosis result is a fault, a warning judgment is made, that is, when The warning result triggered when it lasts for three cycles is a level one warning, and the warning result triggered when it lasts for five cycles is an emergency shutdown command.

[0009] According to a booster station inspection machine provided by the present invention, the multi-source sensor module includes: Infrared thermal imager, used to detect the temperature of booster station equipment; UV corona detector, used to detect corona conditions in booster station equipment; An ultrasonic sensor, used to measure the distance between the mobile platform and obstacles; Sulfur hexafluoride gas sensor, used to detect the concentration of sulfur hexafluoride gas in the environment; The data of the above sensors are collected synchronously.

[0010] According to a booster station inspection machine provided by the present invention, the autonomous navigation module includes: Visual SLAM algorithm unit, used to achieve environmental perception and positioning; Dynamic path planning unit, which performs obstacle avoidance and path planning based on preset paths; The Beidou / GPS dual-mode positioning unit is connected to the visual SLAM algorithm unit and is used to provide location information.

[0011] According to the booster station inspection machine provided by the present invention, the communication module is used to send the diagnosis result and the warning result to the remote operation terminal.

[0012] In another aspect, the present invention provides a remote-controlled robot system, which is mounted on a booster station inspection machine and connected to the mechanical interface. The remote-controlled robot system includes: The remote operation terminal is connected to the communication module and is used to receive the diagnosis result and the warning result, and when the output diagnosis result is a fault, generate a processing plan based on the type of the warning result; A robotic arm is mounted on the mobile platform and is connected to the remote operation terminal by signal. The administrator issues control instructions through the remote operation terminal to drive the robotic arm to perform the corresponding action; A force feedback control terminal, connected to the robotic arm and the remote operation terminal, is used to provide real-time feedback on the operating force of the robotic arm, so that management personnel can adjust control instructions based on the feedback of the operating force; A 3D vision module, mounted on the mobile platform and connected to the remote operation terminal, acquires 3D information of the operating environment through binocular cameras and a lidar, and sends this information to the remote operation terminal to construct a 3D model. Managers can adjust control instructions based on the 3D model. The safety control module is used to calculate the deviation between the operation trajectory and the preset safety area in real time. When the operation deviation exceeds the threshold, automatic braking is performed, and the management personnel readjust the control instructions. If the number of automatic braking times exceeds the preset number, the on-site personnel are notified to handle the situation.

[0013] According to a remote-controlled robot system provided by the present invention, the feedback operation force of the force feedback control terminal The calculation formula is:

[0014] in, Represents the position deviation of the end of the robot arm, that is, the actual position With target location the distance between them; The rate of change of position deviation is used to reflect the speed of the position change of the end of the robot arm; is the integral of the position deviation, which is used to consider the cumulative effect of the position deviation over a period of time; It is a pre-set stiffness matrix based on the material and connection relationship of the robot arm, which is used to adjust the magnitude of the feedback force.

[0015] According to a remote-operated robot system provided by the present invention, in the safety control module, the automatic braking is a three-level braking, namely deceleration, pause, and retraction, and each level of braking corresponds to a different threshold.

[0016] According to a remote operation robot system provided by the present invention, the three-dimensional model constructed by the three-dimensional vision module is displayed on the remote operation terminal through a display device.

[0017] Compared with the prior art, the present invention has the following advantages: 1. The multi-source sensor module of this application comprehensively collects data. The fault diagnosis model uses entropy weight coefficients, standardized activation functions and time decay factors to accurately judge equipment abnormalities. The dynamic early warning model generates adaptive thresholds based on historical diagnosis. The graded early warning is more timely and accurate, reducing misjudgment and missed faults.

[0018] 2. The communication module of this application supports 5G / 4G, and transmits data at high speed and low latency, ensuring that the diagnosis and warning results are delivered to the remote operation end in a timely manner; the autonomous navigation module integrates multiple technologies, with accurate positioning and intelligent path planning, enabling the inspection machine to operate stably in the complex substation environment.

[0019] 3. The force feedback control terminal of the remote-operated robot system of this application provides real-time feedback on force to improve operation accuracy; the three-dimensional vision module constructs a three-dimensional model of the operating environment to assist operation; the safety control module conducts real-time monitoring and automatically brakes when the threshold is exceeded to ensure the safety of equipment and personnel and reduce accident risks.

[0020] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0021] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a structural diagram of a booster station inspection machine provided by an embodiment of the present invention; Figure 2 It is a structural diagram of a remote-controlled robot system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0024] Example 1: The embodiment of the present invention provides a booster station inspection machine, please refer to Figure 1 ,include: The mobile platform adopts an all-terrain adaptive chassis to adapt to the complex terrain environment of the booster station; Multi-source sensor module, which is mounted on a mobile platform and is used to synchronously collect multi-dimensional data from devices; The data processing module is mounted on the mobile platform and connected to the multi-source sensor module. It has a built-in fault diagnosis model and dynamic early warning model based on multi-source data fusion. It determines equipment abnormalities through multi-dimensional data entropy fusion, generates graded early warning signals based on historical diagnosis results, and generates diagnosis results and early warning results; The communication module is mounted on a mobile platform and supports real-time data transmission between 5G / 4G and the remote operation terminal, enabling fast and stable transmission of diagnostic and early warning results. The autonomous navigation module is mounted on a mobile platform and integrates visual SLAM and path planning algorithms to achieve positioning and real-time path planning; The mechanical interface is set on the pre-mobile platform and is used to carry the remote-operated robot, providing a carrying platform for the remote-operated robot.

[0025] The principle and beneficial effects of this embodiment are as follows: the multi-source sensor module is mounted on a mobile platform, which can synchronously collect multi-dimensional data of the booster station equipment, obtain the operating information of the equipment from different angles, and provide a data basis for subsequent analysis; the data processing module is connected to the multi-source sensor module to receive the collected data; its built-in fault diagnosis model based on multi-source data fusion uses the multi-dimensional data entropy fusion method to determine whether the equipment is abnormal, organically combines different sensor data, and comprehensively analyzes the equipment status; the dynamic early warning model generates graded early warning signals based on historical diagnosis results, and predicts future equipment failures by learning and analyzing historical data; the communication module supports 5G / 4G, and the data processing The diagnosis results and warning results generated by the management module are transmitted to the remote operation end in real time, realizing fast and stable data transmission, ensuring that the remote operation end can obtain equipment status information in a timely manner. At the same time, the communication module is also bidirectional, sending control instructions to the following robotic arm drive equipment; the autonomous navigation module integrates visual SLAM and path planning algorithms, realizes environmental perception and positioning through visual SLAM algorithms, and then combines the path planning algorithm to plan a suitable operation path for the mobile platform, so that the inspection machine can move autonomously in the complex terrain of the booster station and complete the inspection task; the mobile platform is equipped with a mechanical interface for carrying the remote operation robot, providing it with a stable carrying platform for the subsequent realization of remote operation functions.

[0026] In order to further optimize the above embodiment, the fault diagnosis model is:

[0027] in, For fault information; is the number of sensors; is the real-time measurement value of the i-th sensor at time t; and During normal operation The mean and standard deviation of historical data; is the entropy weight coefficient, which is used to measure the weight of sensor i in the fault diagnosis model. ,in is the data information entropy of sensor i, which is used to reflect the uncertainty or confusion of the data collected by sensor i; Collect data for sensor i Probability of occurrence; is the maximum value of information entropy; It is a standardized activation function used to normalize the processed sensor data; is the time attenuation factor, which is used to describe the gradual weakening of the impact of data on the current fault diagnosis structure over time. The thermal time constant of the device is used to reflect the speed of temperature change during the operation of the device; when When , the output diagnosis result is fault, where for The historical standard deviation.

[0028] It should be noted that the principle of this fault diagnosis model formula is to integrate multi-source sensor data and comprehensively evaluate the equipment operating status to determine whether there is a fault. First, D(t) in the formula represents the fault information, which is the final basis for determining the equipment fault. n represents the number of sensors. Different types of sensors collect equipment data from multiple dimensions. For each type of sensor, is the real-time measurement value of the i-th sensor at time t, and These are the historical mean and standard deviation of the sensor's data during normal operation. By comparing the real-time measurement value with the historical mean and standard deviation, the degree to which the current measurement value deviates from the normal range can be determined. The historical mean and standard deviation are determined based on at least 1,000 historical operating data. The entropy weight coefficient is determined by the maximum value of the data information entropy and the information entropy. The greater the data information entropy of sensor i, the higher the uncertainty of the data. The entropy weight coefficient automatically assigns weights based on the uncertainty of the sensor data. The more stable the data and the lower the uncertainty, the higher the weight, and vice versa. The normalized activation function is used to normalize processed sensor data, mapping it to a specific range for unified analysis. The time decay factor is based on the device's thermal time constant. It takes into account the timeliness of data, gradually reducing the impact of earlier data on current fault diagnosis results and placing more emphasis on more recent data. Finally, the sensor data is summed after the above calculations and multiplied by the time decay factor to obtain D(t). When D(t) is greater than the historical standard deviation, the equipment is judged to have failed. In this way, accurate diagnosis of equipment failures can be achieved.

[0029] In order to further optimize the above embodiment, the dynamic early warning model is:

[0030] in, is the adaptive threshold, which represents the warning threshold at time t and is used to determine whether to trigger the warning; The 99th percentile of the diagnostic value within the window period is used to measure the high end of the data distribution, reflecting the high level of diagnostic values in most cases, and avoiding false triggering of warnings due to occasional large diagnostic values; is the sliding window size, and is the sampling period. The sliding window method takes into account the changes in recent data, so that the warning threshold can adapt to the dynamic changes in the equipment operation status. Representative Moment The fault diagnosis value, that is, the equipment status evaluation value at different times calculated by the fault diagnosis algorithm, is used to analyze the changing trend of the equipment status over time; Related adjustment items are used to fine-tune the warning threshold; When the output diagnosis result is a fault, a warning judgment is made, that is, when The warning result triggered when it lasts for three cycles is a level one warning, and the warning result triggered when it lasts for five cycles is an emergency shutdown command.

[0031] It should be noted that the principle of this dynamic early warning model is to achieve accurate early warning based on the dynamic analysis of equipment fault diagnosis values; the model first sets an adaptive threshold This is a key indicator for determining whether an early warning is triggered; by calculating the 99th percentile of the diagnostic value within the window period, the high end of the data distribution can be measured. This can effectively avoid false alarms caused by occasional large diagnostic values and ensure the reliability of the early warning. The sliding window size is one of the model's core parameters, determining the data range used for analysis. As time passes, the sliding window continuously shifts, incorporating new fault diagnosis values into the calculation. This takes into account recent data changes, allowing the warning threshold to adjust dynamically with changes in the equipment's operating status. When calculating the adaptive threshold, a weighted average of the fault diagnosis values within the window period is taken, and an adjustment term related to the sliding window size W is added to fine-tune the warning threshold, further improving its accuracy. When the fault diagnosis model outputs a diagnosis result of a fault, an early warning judgment begins. If the situation of D(t)>Th(t) persists for three cycles, it indicates that there is a high possibility of equipment abnormality, and a level 1 warning is triggered at this time, reminding operation and maintenance personnel to pay attention to the equipment status. If this situation persists for five cycles, it indicates that the risk of equipment failure is extremely high, and an emergency shutdown command is directly triggered to avoid possible serious accidents and ensure the safety of equipment and personnel.

[0032] To further optimize the above embodiment, the multi-source sensor module includes: Infrared thermal imager, used to detect the temperature of booster station equipment; UV corona detector, used to detect corona conditions in booster station equipment; Ultrasonic sensor, used to measure the distance between the mobile platform and obstacles; Sulfur hexafluoride gas sensor, used to detect the concentration of sulfur hexafluoride gas in the environment; The data of the above sensors are collected synchronously.

[0033] It should be noted that in actual implementation, high-resolution, high-precision infrared thermal imagers can be selected and installed on an adjustable bracket on a mobile platform. Through pan-tilt control, the infrared thermal imager can be flexibly rotated to conduct all-round temperature detection of equipment at different locations in the booster station. The instrument is equipped with an autofocus function, which can automatically adjust the focal length according to the distance between the equipment and the inspection machine to ensure the acquisition of clear and accurate thermal imaging images. The data acquisition frequency is set to 5 times per second to capture subtle changes in equipment temperature in a timely manner. The UV corona detector is installed at the front end of the mobile platform to ensure a wide field of view without obstructions that could affect the detection results. The detector is connected to the data acquisition unit via optical fiber to reduce electromagnetic interference and ensure the accuracy of the detection data. Its sensitivity is set to 5pC, which can effectively detect weak corona signals. When corona phenomena are detected, the detector records not only the corona intensity but also the time and location of occurrence. Ultrasonic sensors are arranged in an array around the mobile platform. Each sensor is spaced 10 cm apart, enabling comprehensive detection of the distance between the mobile platform and surrounding obstacles. The sensor operating frequency is set at 40 kHz, with a ranging accuracy of ±1 mm. A multi-sensor fusion algorithm comprehensively processes the data from each sensor to improve the accuracy and reliability of distance detection. When the distance to an obstacle is detected to be less than the safety threshold, a signal is immediately sent to the autonomous navigation module to adjust the inspection machine's path in a timely manner. The SF6 gas sensor is installed near the ground on the mobile platform because SF6 gas is denser than air and tends to accumulate in low areas. The sensor uses a high-precision electrochemical sensor with a detection limit of 0.1 ppm. To prevent interference from other gases, a gas filter is installed at the front end of the sensor to ensure that only SF6 gas concentration is detected. The sensor and data acquisition unit are connected via a wired connection to ensure stable data transmission. Data is collected every 10 seconds and uploaded to the data processing module in real time. Sulfur hexafluoride (SF6) is widely used in booster station equipment, such as high-voltage circuit breakers and gas-insulated metal-enclosed switchgear (GIS), due to its excellent insulation and arc-extinguishing properties. However, during long-term operation, SF6 gas may leak due to poor sealing or internal faults. Once the gas leaks, the insulation performance of the equipment will degrade, and its arc-extinguishing ability will also be affected. In severe cases, it may cause equipment short circuits, explosions, and other faults, threatening the safe and stable operation of the booster station. By monitoring the concentration of SF6 gas in the environment, leaks can be detected in a timely manner, allowing operation and maintenance personnel to take measures to avoid equipment failures. SF6 gas itself is non-toxic, but under the influence of high temperature, high pressure, and corona, it will decompose to produce some toxic and corrosive gases, such as sulfur fluoride and sulfur dioxide. These gases leaking into the air can harm the health of on-site workers, such as irritating the respiratory tract and causing poisoning. Continuous monitoring of SF6 gas concentration can sound an alarm at the early stage of gas leaks, reminding workers to evacuate the site in time to ensure their safety. The data collection of each sensor is controlled by a unified time synchronization module. This module uses a high-precision clock as a benchmark to send synchronization pulse signals to each sensor to ensure the synchronous collection of data from each sensor, making subsequent data fusion and analysis more accurate and reliable.

[0034] In order to further optimize the above embodiment, the autonomous navigation module includes: Visual SLAM algorithm unit, used to achieve environmental perception and positioning; Dynamic path planning unit, which performs obstacle avoidance and path planning based on preset paths; The Beidou / GPS dual-mode positioning unit is connected to the visual SLAM algorithm unit to provide location information.

[0035] It should be noted that, in actual implementation, the visual SLAM algorithm unit can use an advanced deep learning-based visual SLAM framework, such as the ORB-SLAM series of algorithms (existing technical means); the inspection machine is equipped with a high-resolution, wide-angle camera, which is installed on the top of the mobile platform to ensure that a large range of environmental images can be obtained; the camera collects image data at a rate of 30 frames per second and transmits it to the visual SLAM algorithm unit in real time; after receiving the image, the algorithm unit constructs an environmental map through feature extraction, matching and other operations to realize the identification and positioning of equipment, obstacles, etc. in the booster station; at the same time, the back-end optimization algorithm is used to continuously optimize the map and its own pose estimation to improve positioning accuracy; The dynamic path planning unit is optimized based on classic path planning algorithms such as the A* algorithm and the Dijkstra algorithm (existing technology). First, a patrol route is preset based on the layout of the substation and the distribution of equipment, and stored in the system. When the patrol machine begins working, the dynamic path planning unit combines the environmental information provided by the visual SLAM algorithm unit and the current location information obtained by the Beidou / GPS dual-mode positioning unit to monitor the obstacles around the mobile platform in real time. Once an obstacle is detected, the path planning unit will re-plan a new path to avoid the obstacle based on the preset path. During the planning process, the optimal path is selected by comprehensively considering factors such as path length, safety, and difficulty of passage. The path information is then sent to the mobile platform's drive system to control the patrol machine to move along the new path. The Beidou / GPS dual-mode positioning unit uses a high-precision positioning module and is installed above the mobile platform to avoid obstruction as much as possible. The module receives Beidou and GPS satellite signals at the same time and improves positioning accuracy through a fusion algorithm. The positioning module sends location information, including longitude and latitude, altitude, etc., to the visual SLAM algorithm unit and dynamic path planning unit at a frequency of 1Hz. The visual SLAM algorithm unit uses the location information provided by the Beidou / GPS dual-mode positioning unit as the initial positioning reference, and combines visual information for more precise positioning, further improving the reliability and accuracy of positioning. Indoors or in areas with weak satellite signals, the visual SLAM algorithm unit mainly relies on visual information for positioning to ensure that the inspection machine can work continuously and stably.

[0036] In order to further optimize the above embodiment, the communication module is used to send the diagnosis results and warning results to the remote operation terminal.

[0037] It should be noted that the communication module uses an industrial-grade communication module with integrated 5G / 4G communication functions, such as a 5G module that supports SA / NSA dual-mode. This module is connected to the data processing module of the inspection machine via a high-speed PCIe interface to ensure stable and high-speed data transmission. In terms of hardware, to ensure the quality of signal reception and transmission, high-gain, omnidirectional 5G / 4G antennas are installed on mobile platforms. By optimizing the antenna layout and signal amplifiers, signal strength is enhanced and signal interference is reduced. At the software level, the communication module integrates a network protocol stack that supports multiple network protocols such as TCP / IP and UDP. During data transmission, the appropriate protocol is selected based on the data type and real-time requirements. For example, for early warning results with high real-time requirements, the UDP protocol is used to ensure fast data transmission; for diagnostic results with high accuracy requirements, the TCP / IP protocol is used to ensure reliable data transmission. Before data transmission, the communication module will encapsulate the diagnosis and warning results, add device identification, timestamp and other information to facilitate remote operation end parsing and management; to ensure data security, the AES-256 encryption algorithm is used to encrypt the transmitted data to prevent the data from being stolen or tampered with during transmission; in addition, a retransmission mechanism is also set up. If the remote operation end does not receive the data correctly, the communication module will automatically retransmit it to ensure data integrity.

[0038] Example 2: The present invention provides a remote control robot system. Figure 2 , mounted on the booster station inspection machine and connected to the mechanical interface, the remote operation robot system includes: A remote operation terminal connected to the communication module is used to receive diagnostic results and warning results. When the output diagnostic result is a fault, a processing plan is generated based on the type of warning result. The robotic arm is mounted on a mobile platform and connected to the remote operation terminal. The manager issues control commands through the remote operation terminal to drive the robotic arm to perform the corresponding actions. The force feedback control terminal is connected to the robot arm and the remote operation terminal to provide real-time feedback on the robot arm's operating force. Managers can adjust control instructions based on the feedback of the operating force. The 3D vision module is mounted on a mobile platform and connected to the remote control terminal. It uses binocular cameras and lidar to obtain 3D information about the operating environment and sends it to the remote control terminal to build a 3D model. Managers can then adjust control instructions based on the 3D model. The safety control module is used to calculate the deviation between the operation trajectory and the preset safety area in real time. When the operation deviation exceeds the threshold, automatic braking is performed, and the management personnel readjust the control instructions. If the number of automatic braking times exceeds the preset number, the on-site personnel are notified to handle the situation.

[0039] The principle and beneficial effects of this embodiment are as follows: the remote operation terminal receives the diagnosis results and warning results of the booster station inspection machine through the communication module; when the diagnosis result is a fault, based on the type of warning result, such as a level 1 warning or an emergency shutdown instruction, a preset processing solution library is called to generate a corresponding processing strategy; The manager issues control commands on the remote operation terminal, which are transmitted to the robotic arm mounted on the mobile platform via a communication link. After receiving the commands, the robotic arm drives the motor to operate, causing the mechanical joints to move, thereby performing the corresponding actions. The force feedback control terminal monitors the force applied to the robotic arm during operation in real time. When the robotic arm contacts an object or is subjected to an external force, the force feedback terminal converts information such as the magnitude and direction of the force into an electrical signal and feeds it back to the remote operation terminal via a communication link. Based on the feedback of the operating force, the management personnel can adjust the control instructions to achieve more precise operation. The binocular camera and lidar in the 3D vision module work simultaneously. The binocular camera uses the principle of parallax to obtain 2D image information of the environment, while the lidar scans the surrounding environment and obtains distance information of objects. After fusing and processing this information, 3D information of the operating environment is generated and transmitted to the remote operation terminal. The remote operation terminal constructs a 3D model based on this 3D information. The manager can observe the operating environment from different angles, make more intuitive judgments, and adjust control instructions accordingly. The safety control module continuously obtains the operating trajectory information of the robotic arm, compares it with the preset safety area, and calculates the deviation between the operating trajectory and the preset safety area in real time; once the deviation exceeds the set threshold, the safety control module immediately activates the automatic braking program to stop the current action of the robotic arm; if the number of automatic braking times exceeds the preset number, it indicates that there is a greater risk in the operation of the robotic arm, and at this time, on-site personnel are notified to deal with it.

[0040] In order to further optimize the above embodiment, the feedback operation force of the force feedback control terminal The calculation formula is:

[0041] in, Represents the position deviation of the end of the robot arm, that is, the actual position With target location the distance between them; The rate of change of position deviation is used to reflect the speed of the position change of the end of the robot arm; is the integral of the position deviation, which is used to consider the cumulative effect of the position deviation over a period of time; It is a pre-set stiffness matrix based on the material and connection relationship of the robot arm, which is used to adjust the magnitude of the feedback force.

[0042] It should be noted that in the formula Represents the position deviation of the robot end. It measures the difference between the current position of the robot and the desired position by calculating the distance between the actual position of the robot end and the target position. This deviation is the basis for generating feedback force. The larger the deviation, the further the robot deviates from the target, and the greater the feedback force required to correct it. is the rate of change of position deviation, which reflects how fast the position of the end of the robot arm changes; if the position of the robot arm changes very quickly, that is, A large value indicates that the robot arm's motion is unstable and may quickly deviate from the target position. In this case, a large feedback force is required to suppress this rapid change, stabilize the motion, and avoid excessive overshoot or loss of control. The integral of the position deviation is used to account for the cumulative effect of position deviation over a period of time. Even if the rate of change of the position deviation is small at a certain moment, if the position deviation persists for a long time, the accumulated deviation will increase through the integral calculation. The feedback force generated by this integral result will prompt the robot arm to return to the target position, just like the elastic force of a spring, constantly correcting the deviation of the robot arm. The stiffness matrix K is pre-set based on the arm's material and connection relationship. It is used to adjust the magnitude of the feedback force. Different arm materials and connection methods will affect their mechanical properties, and the K matrix can be adjusted based on these characteristics. For example, a manipulator with greater stiffness will generate greater feedback force under the same position deviation and change rate, and can correct the manipulator's position deviation more quickly. On the other hand, a manipulator with less stiffness will have a relatively small feedback force and a softer motion response. By adjusting K, the feedback force of the force feedback control terminal can be matched to the actual characteristics of the manipulator, providing a more realistic and effective operating experience. When the manager feels the feedback force, he or she can first determine the movement trend of the robot arm based on the magnitude and direction of the feedback force. If the feedback force is large and points toward the target position, it means that the robot arm's current position deviation is large and it is moving toward the target position, but the speed may be too fast or the position deviation has accumulated a lot. In this case, the manager can appropriately reduce the speed parameter in the control command to make the robot arm move more smoothly and avoid overshoot. If the feedback force is small, the manager can appropriately increase the amplitude of the control command to speed up the robot arm's movement to the target position and improve operational efficiency. For example, when performing fine operations, if the feedback force is close to zero, it means that the robot arm is close to the target position, but there may still be a slight deviation. At this time, the manager can fine-tune the control command to allow the robot arm to reach the target position accurately. When the feedback force changes abnormally, such as a sudden increase or a sudden change in direction, the manager should immediately stop the current operation and check whether the robotic arm has encountered an obstacle or malfunctioned; re-plan the operation path or adjust the control instructions according to the actual situation to ensure safe operation.

[0043] In order to further optimize the above embodiment, in the safety control module, the automatic braking is a three-level braking, namely deceleration, pause, and retraction, and each level of braking corresponds to a different threshold.

[0044] It should be noted that the thresholds corresponding to each level of braking are set according to the working characteristics of the robotic arm and the safety requirements of the operating environment. In the horizontal direction, the deceleration threshold is set to a deviation of 1mm, the pause threshold is set to 1.5mm, and the retraction threshold is set to 2mm. In the vertical direction, the deceleration threshold is set to 0.5mm, the pause threshold is set to 0.8mm, and the retraction threshold is set to 1mm. In terms of torque deviation, the deceleration threshold is 3N·m, the pause threshold is 4N·m, and the retraction threshold is 5N·m. These thresholds can be flexibly adjusted based on actual testing and risk assessment. The safety control module continuously obtains the robot's operating trajectory data and calculates its deviation from the preset safety zone in real time through an algorithm. The calculated deviation value is compared with the braking thresholds at each level to determine whether to trigger the corresponding braking action. When the horizontal or vertical deviation reaches the deceleration threshold, or the torque deviation reaches 3 N·m, the safety control module sends a deceleration command to the robot arm's drive system to reduce the robot arm's movement speed. If the deviation increases further and reaches the pause threshold, the drive system receives the command and stops the robot arm's current movement, keeping it stationary. Once the deviation exceeds the retraction threshold, the drive system controls the robot arm to slowly retract according to the preset program and return to the safe area. Each time the brakes are triggered, the safety control module not only performs the braking action, but also sends an alarm message to the remote operation terminal to inform the management personnel of the braking situation; at the same time, it records the time, type, operation trajectory and other information of the braking, which facilitates subsequent analysis of the cause of the fault and optimization of the operation process.

[0045] In order to further optimize the above embodiment, the three-dimensional model constructed by the three-dimensional vision module is displayed on the remote operation terminal through a display device.

[0046] It should be noted that the binocular camera in the 3D vision module continuously collects images of the operating environment at a high frame rate (such as 60fps), and the laser radar scans synchronously to obtain accurate distance information; the collected data is first pre-processed locally, including noise removal, data calibration, etc., to improve data quality; advanced point cloud reconstruction algorithms, such as the octree-based algorithm (existing technology), are used to fuse the image data of the binocular camera and the distance data of the laser radar to construct a point cloud model of the operating environment; this algorithm can efficiently process large-scale data and accurately restore the shape and position of objects; using 5G communication technology to Point cloud model data is quickly transmitted to the remote operation end. To reduce the amount of data transmission, the data is compressed and encoded before transmission, such as using a voxel compression algorithm to reduce data redundancy. The remote operation end is equipped with a high-resolution display screen (such as 4K resolution). After receiving the data, the point cloud model is rendered into an intuitive three-dimensional model for display through a graphics rendering engine (such as Unity or UnrealEngine). At the same time, interactive functions are provided, and managers can use the mouse, keyboard or handle to zoom, rotate, translate and other operations on the three-dimensional model, observe the operating environment from different angles, and facilitate the adjustment of control instructions.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A booster station inspection machine, characterized in that: include: The mobile platform adopts an all-terrain adaptive chassis to adapt to the complex terrain environment of the booster station; A multi-source sensor module, which is mounted on the mobile platform and is used to synchronously collect multi-dimensional data from the device; A data processing module, which is mounted on the mobile platform and connected to the multi-source sensor module, has a built-in fault diagnosis model and a dynamic early warning model based on multi-source data fusion, determines equipment abnormalities through multi-dimensional data entropy fusion, generates graded early warning signals based on historical diagnosis results, and generates diagnosis results and early warning results; A communication module, which is mounted on the mobile platform and supports real-time data transmission with a remote operation terminal via 5G / 4G, enabling rapid and stable transmission of diagnostic and warning results; An autonomous navigation module, which is mounted on the mobile platform and integrates visual SLAM and path planning algorithms to achieve positioning and real-time path planning; The mechanical interface is provided on the mobile platform and is used to carry the remote-operated robot, providing a carrying platform for the remote-operated robot.

2. A booster station inspection machine according to claim 1, characterized in that: The fault diagnosis model is: in, For fault information; is the number of sensors; is the real-time measurement value of the i-th sensor at time t; and During normal operation The mean and standard deviation of historical data; is the entropy weight coefficient, which is used to measure the weight of sensor i in the fault diagnosis model. ,in is the data information entropy of sensor i, which is used to reflect the uncertainty or confusion of the data collected by sensor i; Collect data for sensor i Probability of occurrence; is the maximum value of information entropy; It is a standardized activation function used to normalize the processed sensor data; is the time attenuation factor, which is used to describe the gradual weakening of the impact of data on the current fault diagnosis structure over time. The thermal time constant of the device is used to reflect the speed of temperature change during the operation of the device; when When , the output diagnosis result is fault, where for The historical standard deviation.

3. The booster station inspection machine according to claim 1, characterized in that: The dynamic early warning model is: in, is the adaptive threshold, which represents the warning threshold at time t and is used to determine whether to trigger the warning; The 99th percentile of the diagnostic value within the window period is used to measure the high end of the data distribution, reflecting the high level of diagnostic values in most cases, and avoiding false triggering of warnings due to occasional large diagnostic values; is the sliding window size, and is the sampling period. The sliding window method takes into account the changes in recent data, so that the warning threshold can adapt to the dynamic changes in the equipment operation status. Representative Moment The fault diagnosis value, that is, the equipment status evaluation value at different times calculated by the fault diagnosis algorithm, is used to analyze the changing trend of the equipment status over time; Related adjustment items are used to fine-tune the warning threshold; When the output diagnosis result is a fault, a warning judgment is made, that is, when The warning result triggered when it lasts for three cycles is a level one warning, and the warning result triggered when it lasts for five cycles is an emergency shutdown command.

4. The booster station inspection machine according to claim 1, characterized in that: The multi-source sensor module comprises: Infrared thermal imager, used to detect the temperature of booster station equipment; UV corona detector, used to detect corona conditions in booster station equipment; An ultrasonic sensor, used to measure the distance between the mobile platform and obstacles; Sulfur hexafluoride gas sensor, used to detect the concentration of sulfur hexafluoride gas in the environment; The data of the above sensors are collected synchronously.

5. The booster station inspection machine according to claim 1, characterized in that: The autonomous navigation module includes: Visual SLAM algorithm unit, used to achieve environmental perception and positioning; Dynamic path planning unit, which performs obstacle avoidance and path planning based on preset paths; The Beidou / GPS dual-mode positioning unit is connected to the visual SLAM algorithm unit and is used to provide location information.

6. The booster station inspection machine according to claim 1, characterized in that: The communication module is used to send the diagnosis result and the warning result to the remote operation terminal.

7. A remotely operated robot system, characterized in that: Mounted on the booster station inspection machine according to any one of claims 1 to 6 and connected to the mechanical interface, the remotely operated robot system includes: The remote operation terminal is connected to the communication module and is used to receive the diagnosis result and the warning result, and when the output diagnosis result is a fault, generate a processing plan based on the type of the warning result; A robotic arm is mounted on the mobile platform and is connected to the remote operation terminal by signal. The administrator issues control instructions through the remote operation terminal to drive the robotic arm to perform the corresponding action; A force feedback control terminal, connected to the robotic arm and the remote operation terminal, is used to provide real-time feedback on the operating force of the robotic arm, so that management personnel can adjust control instructions based on the feedback of the operating force; A 3D vision module, mounted on the mobile platform and connected to the remote operation terminal, acquires 3D information of the operating environment through binocular cameras and a lidar, and sends this information to the remote operation terminal to construct a 3D model. Managers can adjust control instructions based on the 3D model. The safety control module is used to calculate the deviation between the operation trajectory and the preset safety area in real time. When the operation deviation exceeds the threshold, automatic braking is performed, and the management personnel readjust the control instructions. If the number of automatic braking times exceeds the preset number, the on-site personnel are notified to handle the situation.

8. A remote-controlled robot system according to claim 7, characterized in that: The feedback operation force of the force feedback control terminal The calculation formula is: in, Represents the position deviation of the end of the robot arm, that is, the actual position With target location the distance between them; The rate of change of position deviation is used to reflect the speed of the position change of the end of the robot arm; is the integral of the position deviation, which is used to consider the cumulative effect of the position deviation over a period of time; It is a pre-set stiffness matrix based on the material and connection relationship of the robot arm, which is used to adjust the magnitude of the feedback force.

9. A remote-controlled robot system according to claim 7, characterized in that: In the safety control module, the automatic braking is a three-level braking, which is deceleration, pause, and retraction, and each level of braking corresponds to a different threshold.

10. A remote-controlled robot system according to claim 7, characterized in that: The three-dimensional model constructed by the three-dimensional vision module is displayed on the remote operation terminal through a display device.

Images