Flexible tactile sensor array and force control system for power equipment operation and maintenance robot

Abstract

The application discloses a kind of power equipment operation and maintenance robot flexible touch sensor array and force control system, it is related to industrial automatic control technical field, including flexible touch sensor array module, using piezoresistance or optical principle, flexible substrate adapts to the irregular surface of equipment, realizes 0.1N~50N measurement;Signal conditioning and data acquisition module, multi-channel anti-interference design, synchronous acquisition transmission data;Force control algorithm processing module, fusion impedance and PID control, force control error ≤5%;Robot system integration module, double-link communication adapts mechanical arm;Environment adaptation module, humidity compensation and anti-electromagnetic interference;Application verification module, preset process and form operation and maintenance specification.The application improves tactile perception accuracy and force control stability, has fault tolerance and data traceability ability, forms standard operation and maintenance specification, solves the problem of low accuracy, weak adaptation, no fault tolerance of existing system, meets the demand of power equipment automation operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of industrial automatic control technology, and in particular to a flexible tactile sensor array and force control system for power equipment maintenance robots. Background Technology

[0002] As a core component of the power system, the quality of operation and maintenance of power equipment directly determines the safe and stable operation of the power grid. This is especially true for critical equipment such as 10kV circuit breakers and switchgear, which require regular maintenance including opening and closing operations, bolt tightening, and terminal block insertion and removal to prevent power outages caused by equipment failures. Traditional operation and maintenance methods rely primarily on manual operation. Maintenance personnel must enter high-voltage environments such as substation distribution rooms, facing safety risks such as electric shock and equipment misoperation. Furthermore, the precision of manual operation depends on experience, which can easily lead to damage to equipment components due to excessive force or poor contact due to insufficient force. In addition, manual operation and maintenance is inefficient and cannot meet the needs of large-scale power equipment maintenance.

[0003] With the application and promotion of robotics technology in the power sector, power inspection robots are gradually replacing manual labor in completing some maintenance tasks. However, existing robot maintenance systems have significant limitations in tactile perception and force control. In terms of tactile perception, most robots are equipped with tactile sensors that use a rigid substrate design, which cannot adapt to the irregular surfaces of power equipment such as circuit breaker operating handles and switchgear bolts, resulting in incomplete contact force acquisition. The sensor's measurement range and accuracy are difficult to balance; either they cannot cover the 0.1N~50N operating force range of power equipment, or their resolution is insufficient to capture subtle force changes. Furthermore, most sensors lack high and low temperature resistance and electromagnetic interference resistance design, making them prone to signal drift or failure in substations with temperature fluctuations of -20℃ to 60℃, humidity variations of 5% to 95%, and high-frequency electromagnetic environments, affecting the accuracy of perception.

[0004] At the force control system level, existing technologies mostly employ a single PID control algorithm, resulting in poor dynamic response performance. They cannot automatically adjust parameters based on the force-displacement characteristics of different operational tasks of power equipment, leading to large errors in force control. Some systems fail to consider the impact of environmental factors on force control effectiveness; changes in temperature and humidity alter the friction coefficient of equipment components, causing the actual operating force to deviate from the target value. Furthermore, the lack of effective sensor calibration and fault tolerance mechanisms means that sensor accuracy declines after long-term use, or signal loss or data transmission errors occur. The system cannot promptly correct or switch to backup solutions, easily causing maintenance task interruptions or even damage to power equipment. In addition, existing systems lack standardized maintenance operation specifications; maintenance parameters for different models of power equipment need to be manually reset, resulting in a lack of uniformity in operation procedures. This hinders technology replication and large-scale application, restricting the deep implementation of robotic maintenance in the power sector. Summary of the Invention

[0005] The present invention proposes a flexible tactile sensor array and force control system for power equipment operation and maintenance robots to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a flexible tactile sensor array and force control system for power equipment operation and maintenance robots, comprising the following modules: Flexible tactile sensor array module: The flexible tactile sensor array is designed using the piezoresistive effect or optical principle to adapt to the irregular surface of power equipment. It achieves environmental resistance through epoxy resin potting and polytetrafluoroethylene coating encapsulation and collects physical signals during operation in real time. Signal conditioning and data acquisition module: Develop a 16-channel sensor signal conditioning circuit, a method for generating image description text based on a large model, suppress electromagnetic interference in the power environment, synchronously acquire tactile data through a high-speed data acquisition card, convert it into a digital signal and transmit it to the force control algorithm processing module via CAN bus; Force control algorithm processing module: Based on impedance control theory, a force feedback algorithm is developed, which integrates incremental PID and two-dimensional fuzzy control to optimize dynamic response, constructs an online recognition model of operating force-displacement curve, extracts force change and displacement features, automatically adapts to different operating tasks, calculates force deviation and outputs PWM control commands; Robot System Integration Module: Integrates a flexible tactile sensor array, signal conditioning module, and force control algorithm into the power inspection robot platform. It achieves inter-module communication through dual links, develops motion-force control collaborative linkage program, and matches the robot arm's movement speed with the signal acquisition frequency. Environmental adaptation module: Built-in temperature and humidity compensation and adjustment unit, dynamically corrects sensor parameters. The sensor is treated with anti-fouling insulating coating and has the ability to withstand pressure and resist electromagnetic interference, meeting the safety specifications of power operation and maintenance. Application verification module: For typical power equipment, standardized operation and maintenance processes are preset, and the success rate of operation is ensured through scenario testing, forming a tactile operation and maintenance standard for power equipment.

[0007] Furthermore, it also includes a sensor dynamic calibration module, which is equipped with a piezoelectric force calibration device. By periodically collecting four standard force signals of 0.1N, 10N, 25N, and 50N and comparing them with the sensor output signal, a calibration error model based on the least squares method is established, and a calibration coefficient table including temperature and humidity correction terms is generated. The calibration process is automatically started every time the robot starts or the ambient temperature and humidity change exceeds ±5℃, correcting the sensor measurement deviation.

[0008] Furthermore, it also includes a multimodal data fusion module, which collects equipment contour data from robot vision sensors, motion position data from laser displacement sensors, and contact force data from tactile sensors. It achieves synchronization of multi-source data frames through a timestamp alignment algorithm and integrates feature information using a weighted fusion algorithm to construct a three-dimensional state model of power equipment operation.

[0009] Furthermore, the force control algorithm processing module also integrates an impedance parameter dynamic optimization unit to construct an adaptive adjustment mechanism for impedance control parameters. ,in For the desired impedance, Based on the basic stiffness coefficient, Based on the basic damping coefficient, Based on the fundamental inertia coefficient, For the Laplace operator, For real-time ambient temperature, For standard reference temperature, For real-time operation, For target operational force, This is the stiffness correction factor. For real-time relative humidity, For standard reference humidity, To maximize humidity tolerance, Minimum humidity tolerance, This is the damping correction factor.

[0010] Furthermore, the robot system integration module also includes an operation task identification unit, which pre-stores force-displacement feature templates for typical tasks. By collecting force and displacement signals during the operation process in real time, it calculates the cosine similarity with each template. The template with the highest similarity exceeding 0.9 is determined as the current task type, and the corresponding force control parameter combination is called to complete the adaptive switching of different operation and maintenance tasks.

[0011] Furthermore, it also includes a fault diagnosis and fault tolerance module, which monitors the signal amplitude integrity of the flexible tactile sensor, the transmission frame loss rate of the data acquisition module, and the running error of the force control algorithm in real time. When a sensor signal loss, data transmission error, or force control error exceeds 5% and lasts for 3 consecutive sampling cycles, the fault tolerance mechanism is automatically activated, switching to the backup sensor channel or enabling the simplified force control algorithm, while recording the fault information and issuing an early warning via Ethernet.

[0012] Furthermore, the signal conditioning and data acquisition module also includes a multi-channel synchronous calibration unit to construct a tactile data acquisition synchronization error correction system. ,in For the data synchronization timestamps of each channel, The timestamp for the first channel is collected. This represents the total number of data acquisition channels. For the first Channel signal transmission delay, For the first Channel signal amplitude, The amplitude of the first channel signal. The maximum signal amplitude, For the minimum signal amplitude, This is the amplitude-dependent delay correction amount. For the first The physical distance between the passage and the first passage. Maximum channel spacing This is the distance-related delay correction amount.

[0013] Furthermore, the environmental adaptation module also includes a vibration and electromagnetic shielding unit. The sensor array is packaged with a multi-layer buffer structure of "silicone shock-absorbing pad + metal frame" to absorb vibration interference in power equipment operation and maintenance scenarios. The signal conditioning circuit and data acquisition module are packaged with a 1.5mm thick cold-rolled steel plate shielding box to effectively suppress high-frequency electromagnetic interference in the substation.

[0014] Furthermore, it also includes a data storage and traceability module, which adopts a combination of local edge storage and cloud backup. Key information during the operation process is stored locally, and data can be retrieved by device type, operation date, and task type. The cloud synchronization storage cycle is ≤1 hour, and historical data is retained for at least 1 year.

[0015] Furthermore, the application verification module also includes a standardized operation specification output unit, which generates a standardized operation process guide based on operation and maintenance test data. The guide includes an equipment model adaptation table, which clarifies the force control parameter range corresponding to different equipment specifications. It also includes built-in operation step guidance and safety warning logic. When the robot's operation parameters exceed the specified range, it automatically pauses and prompts the robot to adjust its direction through the human-machine interface.

[0016] Compared with existing technologies, the beneficial effects of this invention are: At the tactile sensing level, the flexible tactile sensor array uses a composite polymer material substrate and a special encapsulation process. This allows it to closely adhere to the irregular surfaces of power equipment, enabling comprehensive acquisition of contact forces, while also withstanding extreme temperatures, humidity levels, and electromagnetic interference, ensuring the stability of sensing data in complex power environments. The multi-channel signal conditioning and synchronous acquisition design improves the accuracy and response speed of force measurement, accurately capturing subtle force changes during operation and maintenance, providing reliable data support for subsequent force control decisions. The dynamic calibration module periodically corrects sensor deviations, avoiding accuracy degradation caused by long-term use and ensuring continuous stability of sensing performance. This fundamentally solves the problems of poor adaptability, weak environmental tolerance, and insufficient accuracy of existing sensors.

[0017] At the force control system level, the force feedback design based on impedance control and multi-algorithm fusion overcomes the limitations of single PID control. It can automatically adjust parameters according to the force-displacement characteristics of different operation and maintenance tasks, significantly improving force control accuracy. The impedance parameter dynamic optimization unit combines ambient temperature and humidity with real-time operating force data to dynamically adjust control parameters, offsetting the impact of environmental factors on force control effects and ensuring the stability of operating force under different working conditions. The operation task identification unit realizes automatic identification and parameter calling of typical operation and maintenance tasks, and can complete task switching without manual intervention, significantly improving the degree of operation automation and effectively solving the problems of poor adaptability and cumbersome parameter adjustment of existing force control algorithms.

[0018] At the system integration and operation and maintenance level, the multimodal data fusion module integrates visual, displacement, and tactile data to provide more comprehensive information support for force control decisions and reduce operational risks caused by deviations in data from a single sensor. The fault diagnosis and fault tolerance mechanism can monitor the system's operating status in real time and automatically switch to backup solutions when sensor failures or data transmission anomalies occur, ensuring continuous operation and maintenance tasks and avoiding equipment damage or task interruption. The vibration-resistant and electromagnetic shielding design further enhances the system's stability in complex power environments, ensuring reliable data transmission and command execution.

[0019] In terms of application standards and efficiency, the standardized operation specification output unit, based on typical power equipment test data, forms a unified specification containing operation parameters and process guidelines, which can be directly replicated and applied to the operation and maintenance of similar equipment, reducing the difficulty of technology promotion; the data storage and traceability module retains complete operation and maintenance data, providing support for subsequent effect analysis and parameter optimization; the overall system realizes the automation, precision and standardization of power equipment operation and maintenance, which not only avoids the safety risks of manual operation and maintenance, but also improves operation and maintenance efficiency and quality, providing strong technical support for the large-scale application of robot operation and maintenance of power equipment. Attached Figure Description

[0020] Figure 1 This is a schematic block diagram of the flexible tactile sensor array and force control system for the power equipment operation and maintenance robot proposed in this invention; Figure 2 Line graph showing the variation of measurement error of flexible tactile sensors under different ambient temperatures; Figure 3 A bar chart comparing torque control accuracy for different bolt specifications; Figure 4 A bar chart comparing system fault recovery times under different fault types. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0023] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0024] Reference Figures 1 to 4 A flexible tactile sensor array and force control system for power equipment maintenance robots, comprising the following modules: Flexible tactile sensor array module: The flexible tactile sensor array is designed using the piezoresistive effect or optical principle. The flexible substrate is made of silicone rubber and polyimide composite polymer material that is resistant to high and low temperatures and anti-aging. It is adapted to the irregular surfaces of power equipment, such as the non-flat contours of components like circuit breaker operating handles and switch cabinet bolts. The array has a scale of 32×32 sensing units, and the measurement range of a single unit covers 0.1N~50N. The force measurement resolution reaches ±0.1N, and the sensor response time is ≤10ms. It is encapsulated with epoxy resin and polytetrafluoroethylene coating to achieve environmental temperature resistance of -20℃~60℃ and non-condensing operation at 5%~95% relative humidity. It can collect physical signals such as the magnitude, uniformity of distribution and dynamic change trend of contact force during the operation of power equipment in real time. Signal Conditioning and Data Acquisition Module: Develops a 16-channel sensor signal conditioning circuit, integrating a low-noise instrumentation amplifier, an 8th-order Butterworth filter circuit, and a 12-bit level conversion unit to suppress electromagnetic interference in the power environment above 30dB, achieving noise reduction and amplitude linear calibration of tactile signals. Equipped with a high-speed USB 3.0 data acquisition card, it supports simultaneous acquisition of 16 channels of tactile data, with a sampling frequency ≥1kHz and a data transmission rate ≥10Mbps. After converting analog signals into 16-bit digital signals, they are transmitted to the force control algorithm processing module via the CAN bus. Force control algorithm processing module: Based on impedance control theory, a force feedback algorithm is developed, which integrates incremental PID control and two-dimensional fuzzy control to optimize dynamic response performance. An online recognition model of operation force-displacement curve is constructed. By extracting the slope of force change and displacement inflection point features during the operation process, the module realizes automatic parameter adjustment for different operation tasks such as insertion, rotation, and pressing. The module calculates the deviation between the operation force and the target force in real time and outputs PWM control commands, so that the robot operation force control error is ≤5%. It supports adaptive adjustment of contact force on different material surfaces such as metal, ceramic, and plastic, with an adjustment step size of ≤0.05N. Robot System Integration Module: Integrates a flexible tactile sensor array, signal conditioning module, and force control algorithm into the existing power inspection robot platform. Communication between modules is achieved through dual links of CAN bus and Ethernet, with a communication success rate of ≥99%. It is adapted to the robot's 6-DOF robotic arm motion mechanism and develops a motion-force control collaborative linkage program to match the robotic arm's motion speed with the tactile force signal acquisition frequency. It supports typical maintenance operations such as 10kV circuit breaker opening and closing, switchgear bolt tightening (torque range 5-15N・m), and terminal block insertion and removal. Environmental Adaptation Module: It has a built-in platinum resistance temperature compensation circuit and a capacitive humidity adaptive adjustment unit. It dynamically corrects the sensor measurement parameters by monitoring the ambient temperature and humidity data in real time. The correction coefficient is linearly adjusted at 5% RH per ℃. The sensor surface is treated with nano anti-fouling and epoxy insulation coating. The withstand voltage rating is ≥10kV. The electromagnetic interference resistance meets the level 3 test requirements in GB / T17626.3-2016 standard and meets the safety specifications for power equipment operation and maintenance. Application Verification Module: For typical power equipment such as 10kV circuit breakers (e.g., VS1 type) and switchgear (e.g., KYN28 type), a standardized operation and maintenance process is preset. The process includes three stages: equipment status detection before operation, force parameter monitoring during operation, and effect verification after operation. Through simulated high temperature and high humidity environment of substation and real power distribution room scenario testing, the operation success rate is ≥98%. It forms a replicable power equipment tactile operation and maintenance specification that includes operation steps, force control parameters, and safety precautions, and supports real-time recording and timestamp traceability of operation data.

[0025] This invention also includes a sensor dynamic calibration module, which is equipped with a high-precision piezoelectric force calibration device with a calibration range of 0.1N to 50N and a calibration accuracy of ±0.01N. By periodically collecting four standard force signals of 0.1N, 10N, 25N, and 50N and comparing them with the sensor output signal, a calibration error model based on the least squares method is established, and a calibration coefficient table including temperature and humidity correction terms is generated. The calibration process is automatically started every time the robot starts or the ambient temperature and humidity change exceeds ±5℃, correcting the sensor measurement deviation and maintaining a stable output of force measurement accuracy during long-term operation. After calibration, the sensor measurement deviation is controlled within ±0.05N.

[0026] This invention also includes a multimodal data fusion module. This module collects equipment contour data from a robot vision sensor (resolution ≥1920×1080), motion position data from a laser displacement sensor (measurement range 0-500mm, accuracy ±0.02mm), and contact force data from a tactile sensor. It achieves synchronization of multi-source data frames through a timestamp alignment algorithm and integrates feature information using a weighted fusion algorithm. The weights are allocated as follows: visual data 0.3, displacement data 0.2, and tactile data 0.5. This constructs a three-dimensional state model of power equipment operation, providing a more comprehensive decision-making basis for the force control algorithm, reducing operational deviations caused by single sensor data biases, and improving the accuracy of force control command execution to over 98% in complex operation and maintenance scenarios.

[0027] In this invention, the force control algorithm processing module also integrates an impedance parameter dynamic optimization unit to construct an adaptive adjustment mechanism for impedance control parameters. ,in For the desired impedance, It is the basic stiffness coefficient (unit: N / m). Based on the basic damping coefficient, Based on the fundamental inertia coefficient, For the Laplace operator, For real-time ambient temperature, For standard reference temperature, For real-time operation, For target operational force, This is the stiffness correction factor. Real-time relative humidity (, For standard reference humidity, To maximize humidity tolerance, Minimum humidity tolerance, As a damping correction factor, this optimization unit reads the temperature and humidity data fed back by the environmental monitoring module and the operating force data collected by the force sensor in real time, substitutes them into the formula to calculate the desired impedance value, and drives the force control algorithm to dynamically adjust the output parameters, so that the robot can maintain a compliant and precise force control effect under different environments and operating conditions.

[0028] In this invention, the robot system integration module also includes an operation task identification unit, which pre-stores force-displacement feature templates for typical tasks such as circuit breaker opening and closing, bolt tightening, and terminal block insertion and removal. The templates include parameters such as the force change range, displacement stroke interval, and number of feature inflection points for each task. By collecting force and displacement signals during the operation in real time, the cosine similarity with each template is calculated. The template with the highest similarity exceeding 0.9 is determined as the current task type, and the corresponding force control parameter combination is called. The adaptive switching of different operation and maintenance tasks can be completed without manual intervention. The task identification response time is ≤200ms, which improves the automation and efficiency of the system operation.

[0029] This invention also includes a fault diagnosis and fault tolerance module, which monitors in real time the signal amplitude integrity of the flexible tactile sensor (normal range 0.5V-5V), the transmission frame loss rate of the data acquisition module, and the operating error of the force control algorithm. When sensor signal loss, data transmission error, or force control error exceeds 5% for three consecutive sampling cycles, the fault tolerance mechanism is automatically activated, switching to a backup sensor channel or enabling a simplified force control algorithm. Simultaneously, information such as the fault occurrence time, fault type, and current operation steps is recorded, and an early warning is issued via Ethernet to prevent power equipment maintenance delays caused by task interruptions. The completion rate of basic maintenance operations remains above 95%. In this invention, the signal conditioning and data acquisition module further includes a multi-channel synchronous calibration unit to construct a tactile data acquisition synchronization error correction system. ,in For the data synchronization timestamps of each channel, The timestamp for the first channel is collected. This represents the total number of data acquisition channels. For the first Channel signal transmission delay, For the first Channel signal amplitude, The amplitude of the first channel signal. The maximum signal amplitude, For the minimum signal amplitude, This is the amplitude-dependent delay correction amount. For the first The physical distance between the passage and the first passage. Maximum channel spacing As a distance-related delay correction, this unit calculates the synchronization timestamp by measuring the physical distance and signal amplitude of each channel in real time and combining it with preset delay correction parameters, ensuring that the data acquisition time difference of the 16 channels is ≤1ms, thereby improving the force control algorithm's ability to accurately judge the distribution of contact force.

[0030] In this invention, the environmental adaptation module further includes a vibration-resistant and electromagnetic shielding unit. The sensor array is packaged using a multi-layer buffer structure of "silicone damping pad + metal frame". The silicone damping pad is 5mm thick and has a hardness of 50 Shore A, absorbing vibration interference in power equipment operation and maintenance scenarios. The signal conditioning circuit and data acquisition module are packaged in a 1.5mm thick cold-rolled steel plate shielding box, with the shielding layer passing through a 6mm... 2 The grounding wire is grounded with a grounding resistance of ≤1Ω. The cable is a twisted pair shielded cable with an aluminum foil shielding layer and a shielding coverage of ≥90%, which effectively suppresses high-frequency electromagnetic interference in the substation and makes the sensor data transmission packet loss rate ≤0.1% and the force control command response delay ≤50ms in complex power environments.

[0031] This invention also includes a data storage and traceability module, which combines local edge storage with cloud backup. The local storage stores key information such as tactile data, force control parameters, and operation results during the operation process. It supports data retrieval by device type, operation date, and task type. The cloud synchronization storage period is ≤1 hour, retaining at least 1 year of historical data. The data retrieval response time is ≤10 seconds, providing continuous and complete data support for power equipment operation and maintenance effect analysis, system parameter optimization, and fault tracing.

[0032] In this invention, the application verification module also includes a standardized operation specification output unit. Based on the operation and maintenance test data of equipment such as 10kV circuit breakers and switchgear, it generates a standardized operation procedure guide containing parameters such as operating force threshold, movement speed limit, and contact time control. The guide includes an equipment model adaptation table, which clarifies the force control parameter range corresponding to different specifications of equipment. It has built-in operation step guidance and safety warning logic. When the robot's operation parameters exceed the standard range, it automatically pauses and prompts the robot to adjust its direction through the human-machine interface. This ensures that the standard compliance rate of operation and maintenance reaches 100% and the accident rate is reduced to 0. At the same time, the specification can be exported as a PDF through a USB interface, providing a replicable technical reference for the robot operation and maintenance of similar power equipment.

[0033] The following two examples further illustrate specific embodiments of the present invention: Example 1: Application of 10kV VS1 type circuit breaker opening and closing operation and maintenance scenario This embodiment focuses on the operation and maintenance of a 10kV VS1 vacuum circuit breaker. It integrates a flexible tactile sensor array and a force control system into a power inspection robot to achieve precise control and process monitoring of the opening and closing operation force. The system configuration includes a 32×32 flexible tactile sensor array, a 16-channel signal conditioning circuit, a force control algorithm processing unit, and an environmental adaptation module. It focuses on verifying the accuracy of tactile perception, force control error control, and environmental adaptability, and fully covers all technical solutions.

[0034] 1. Deployment of flexible tactile sensor array modules The flexible tactile sensor array employs a piezoresistive effect design. The flexible substrate is a composite material of silicone rubber and polyimide, 0.8 mm thick, with a Shore A hardness of 45. It can closely conform to the curved surface of the circuit breaker operating handle, with a radius of curvature of 15 mm. The array is divided into 32×32 sensing units, each with an effective sensing area of ​​1 mm × 1 mm. A carbon nanotube piezoresistive layer is fabricated using a screen printing process. The measurement range is 0.1 N to 50 N, with a resolution of ±0.1 N. The sensor is encapsulated in epoxy resin with a thickness of 0.5 mm, and covered with a 0.1 mm thick polytetrafluoroethylene coating. After high and low temperature chamber testing, the output signal drift is ≤0.02 V within the range of -20℃ to 60℃, and there is no condensation phenomenon in an environment with 5% to 95% relative humidity. The response time, measured by an oscilloscope, is 8 ms, meeting the real-time sensing requirements for circuit breaker opening and closing operations. The sensor array is attached to the end effector of the robot arm using high-temperature resistant double-sided adhesive. The actuator is adapted to the size of the circuit breaker operating handle to ensure full contact between the sensor and the handle during operation.

[0035] 2. Signal conditioning and data acquisition module operation The signal conditioning circuit integrates a 16-channel low-noise instrumentation amplifier with a gain set to 100x, paired with an 8th-order Butterworth low-pass filter circuit with a cutoff frequency of 100Hz. It also includes a 12-bit level conversion unit. The circuit's input impedance is ≥10MΩ, and its output impedance is ≤100Ω, effectively suppressing 35dB of high-frequency electromagnetic interference within the substation, covering an interference frequency range of 10kHz to 1MHz. The high-speed data acquisition card uses a USB 3.0 interface, with a sampling frequency set to 1kHz. It simultaneously acquires 16 channels, each with a 16-bit AD conversion accuracy. The measured data transmission rate is 12Mbps. Digital signals are transmitted to the force control algorithm processing module via a CAN bus with a baud rate of 250kbps.

[0036] The multi-channel synchronous calibration unit initiates synchronization error correction. Substituting into the formula: in =16, =100ms is the timestamp for the first channel acquisition. The value range of 0.1ms to 0.3ms represents the transmission delay for channels 2-16. =2.5V is the amplitude of the first channel signal, which corresponds to a force of 25N. The signal amplitudes for channels 2-16 are set between 2.3V and 2.7V. Corresponding to a force of 50N, =0.5V corresponds to 0.1N force. =0.05ms, =0mm is the reference position for the first channel. =5mm~30mm is the physical distance between channels 2-16 and channel 1. =30mm, =0.02ms. Taking channel 5 as an example, The calculation yields: The synchronization time difference of the 16 channels was ultimately controlled at 0.8ms to ensure the time consistency of the contact force distribution data.

[0037] 3. Force control algorithm processing module integrated with robot The force control algorithm processing module uses an STM32H743 embedded controller, based on impedance control theory and integrating incremental PID and two-dimensional fuzzy control. The incremental PID has a proportional coefficient of 1.2, an integral coefficient of 0.05, and a derivative coefficient of 0.1. The two-dimensional fuzzy control input variables are force deviation and the rate of change of deviation, and the output is the control quantity correction value. The target operating force for circuit breaker opening and closing is set to 10N. The impedance parameter dynamic optimization unit reads environmental data in real time, where T_env = 30℃ and RH_env = 65%. Substituting these values ​​into the formula: in =500N / m, =20 N·s / m =0.5kg, =25℃ is the standard reference temperature =9.8N is the real-time operating force. =10N is the target operating force. =50N / m is the stiffness correction factor. =60% is the standard reference humidity. =95% is the maximum tolerance humidity. =5% is the minimum tolerance humidity. =2N・s / m is the damping correction factor, calculated as follows: The algorithm is based on The system outputs PWM control commands to drive the robot's 6-DOF robotic arm. The end effector speed is set to 30 mm / s, enabling precise control of the opening and closing force. The robot system integration module communicates via a dual-link CAN bus and Ethernet, with a measured communication success rate of 99.5%. The robotic arm's motion and tactile signal acquisition frequencies are matched at 1 kHz, and there is no command delay during operation.

[0038] 4. Environmental Adaptability and Fault Tolerance Testing The environmental adaptation module incorporates a PT100 platinum resistance temperature sensor and an SHT30 humidity sensor, monitoring ambient temperature and humidity in real time and dynamically correcting sensor parameters. When the temperature rises from 25℃ to 35℃, the correction coefficient automatically adjusts to 1.02, and the sensor output signal deviation is controlled within ±0.03V. The vibration resistance unit, tested on a vibration table, exhibits sensor signal fluctuations ≤0.05V under vibration conditions within a frequency range of 5Hz to 500Hz and an acceleration of 2g. The electromagnetic shielding unit demonstrates a data transmission packet loss rate of 0.08% under 1MHz high-frequency interference, meeting the Level 3 requirements of GB / T17626.3-2016 standard.

[0039] The fault diagnosis and fault tolerance module simulates sensor signal loss. In the simulated scenario, the amplitude of the signal on channel 8 drops to 0.05V. The system detects the fault within three sampling cycles (3ms each), automatically switches to the backup channel 9, and simultaneously sends an early warning message via Ethernet. Maintenance tasks are not interrupted; the opening and closing operations continue, and the operating force control error remains ≤5%. The sensor dynamic calibration module performs calibration monthly using a high-precision piezoelectric force calibration device with an accuracy of ±0.01N. It collects standard force signals of 0.1N, 10N, 25N, and 50N to establish a calibration error model. After calibration, the sensor measurement deviation is controlled within ±0.04N.

[0040] 5. Application Validation and Effect Data The application verification module pre-sets the opening and closing operation and maintenance procedures for the 10kV VS1 circuit breaker: Before operation, the position of the circuit breaker handle is detected by a visual sensor, with a positioning error ≤0.5mm; during operation, the operating force is monitored in real time and controlled between 9.5N and 10.5N; after operation, the opening and closing are confirmed to be in place by a displacement sensor, with a stroke error ≤1mm. 100 opening and closing tests were completed in a real substation environment, with a success rate of 98%, forming a standardized specification including operation steps and force control parameters.

[0041] Table 1 is a comparison table of the opening and closing operation and maintenance performance of the 10kV VS1 type circuit breaker: Table 1 shows that existing robot systems suffer from poor adaptability of tactile sensors, making it impossible to comprehensively collect the contact force of the circuit breaker handle. This results in large operational force control errors, signal failures under fluctuating temperature and humidity or electromagnetic interference, and the lack of fault tolerance mechanisms, leading to low operation success rates. The system of this invention achieves comprehensive perception through flexible sensors, and its force control algorithm is dynamically optimized based on environmental parameters. This results in smaller operational force control errors, stronger environmental adaptability and fault tolerance, and a significantly improved operation success rate. It fully meets the high precision and high stability requirements of 10kV circuit breaker opening and closing maintenance, effectively replacing manual operation and reducing safety risks.

[0042] Example 2: Application of bolt tightening maintenance in 10kV KYN28 switchgear This embodiment focuses on the bolt tightening maintenance task of a 10kV KYN28 type switchgear, with a bolt tightening torque range of 5-15 N·m. It integrates a flexible tactile sensor array and force control system into a power maintenance robot, and focuses on verifying the bolt tightening torque control accuracy, multi-task adaptive switching and standardized output capability, fully covering all technical solutions.

[0043] 1. Configuration of Flexible Tactile Sensor Array and Signal Conditioning Module The flexible tactile sensor array is designed using optical principles. The flexible substrate is made of transparent polyimide with a thickness of 0.6mm. It incorporates 32×32 grating sensing units, each capable of detecting pressure from 0.1N to 50N. This pressure is converted into a light intensity signal through optical deformation, with a resolution of ±0.1N. The sensor is fixed to the inner wall of a bolt sleeve at the end of the robot arm using laser welding. This bolt sleeve is compatible with M8-M12 bolt sizes, ensuring full contact between the sensor and the bolt head during tightening to collect pressure distribution data. The encapsulation uses a 0.05mm thick polyimide film, resistant to temperatures from -20℃ to 60℃ and humidity from 5% to 95%, with a tested response time of 7ms.

[0044] The signal conditioning circuit integrates a 16-channel photoelectric conversion unit with a conversion efficiency of ≥98%, paired with a low-noise amplifier with an 80x gain. It also includes an 8th-order Butterworth filter circuit with a cutoff frequency of 80Hz to suppress electromagnetic interference in substations, achieving an interference suppression capability of over 30dB. The high-speed data acquisition card has a sampling frequency of 1kHz, simultaneously acquiring data from 16 channels at a data transmission rate of 11Mbps, transmitting the data to the force control algorithm processing module via a CAN bus. The multi-channel synchronous calibration unit uses formulas to calculate and control the synchronization time difference of the 16 channels to within 0.9ms, ensuring the synchronization of pressure distribution data.

[0045] 2. Force control algorithm and operation task identification and operation The force control algorithm processing module is based on impedance control fused with PID and fuzzy control. The target torque for bolt tightening is set to 10 N·m, which corresponds to an operating force of 12.5 N. The impedance parameter dynamic optimization unit reads environmental data, including... =28℃ =70%, substituting into the formula: in =600N / m, =25 N·s / m, =0.6kg, =25℃ is the standard reference temperature =12.3N is the real-time operating force. =12.5N is the target operating force. =60N / m is the stiffness correction factor. =60% is the standard reference humidity. =95% is the maximum tolerance humidity. =5% is the minimum tolerance humidity. =3N・s / m is the damping correction factor, calculated as follows: The algorithm outputs control commands to drive the robot arm. The robot arm's rotation speed is set to 10 r / min to complete the bolt tightening. The operating force control error is ≤5%.

[0046] The task recognition unit pre-stores force-displacement feature templates for bolt tightening and terminal block insertion / removal. The bolt tightening template has a force range of 8N~15N and a displacement range of 5mm~10mm, while the terminal block insertion / removal template has a force range of 2N~8N and a displacement range of 2mm~5mm. When the system switches to the bolt tightening task, it collects force and displacement signals in real time. The force signal is 12.3N, and the displacement signal is 8mm. The calculated cosine similarity with the bolt tightening template is 0.93, which is higher than the 0.9 threshold, thus identifying it as a bolt tightening task. The system automatically calls upon the stress control parameters, with PID coefficients of 1.3, 0.06, and 0.12. The task recognition response time is 180ms, requiring no manual intervention.

[0047] 3. Multimodal fusion and environment-adaptive operation The multimodal data fusion module acquires bolt position data from a visual sensor (2048×1536 resolution, positioning error ≤0.3mm) and simultaneously acquires robotic arm displacement data from a laser displacement sensor (0-500mm measurement range, ±0.02mm accuracy). This data is then aligned with pressure data from a tactile sensor using timestamps, with a synchronization error ≤0.5ms. The data is then weighted and fused according to a weighting of 0.3 for visual, 0.2 for displacement, and 0.5 for tactile, to construct a three-dimensional bolt tightening state model. When the visual sensor detects a bolt position offset of 0.2mm, the fusion module adjusts the force control command to maintain stable operating force and prevent bolt stripping.

[0048] The temperature compensation circuit of the environmental adaptation module has a correction coefficient of 1.03 at -10℃, with a sensor output deviation of ≤0.03V. The vibration resistance unit exhibits signal fluctuations of ≤0.06V under vibration conditions of 500Hz frequency and 3g acceleration. The electromagnetic shielding unit achieves a data transmission packet loss rate of 0.07% under 10kHz interference. The data storage and traceability module uses a 128GB industrial-grade SD card for local storage with a one-year storage period. Data is synchronized to the cloud every 30 minutes via the MQTT protocol, supporting data retrieval by bolt specification (M8 / M10 / M12) and operation date, with a retrieval response time of 8 seconds.

[0049] 4. Fault testing and standardized specification output The fault diagnosis and fault tolerance module simulates data transmission errors. In the simulated scenario, three consecutive frames of data are lost. The system detects the fault within 3ms, switches to Ethernet link for data transmission, and the bolt tightening task continues. The operating force control error remains ≤5%. The sensor dynamic calibration module starts calibration every two months, using a high-precision force calibration device to collect standard force signals. After calibration, the measurement deviation is ≤0.05N.

[0050] The application verification module generates standardized operating procedures based on test data of M8, M10, and M12 bolts from a 10kV KYN28 switchgear: M8 bolt tightening force 8N~10N, speed 8r / min; M10 bolt 10N~12N, speed 10r / min; M12 bolt 12N~15N, speed 12r / min. The procedures include an equipment model compatibility table and operating step guidance. When the operating force exceeds the range, for example, an M10 bolt force of 13N, the system automatically pauses and prompts "Force value exceeds limit, it is recommended to reduce the speed." The operating procedure compliance rate is 100%.

[0051] Table 2 is a comparison table of the bolt tightening and maintenance performance of the 10kV KYN28 type switchgear: Table 2 shows that existing robot systems have low accuracy in controlling bolt tightening torque, easily resulting in excessive looseness or tightness. They cannot automatically switch operating parameters for different bolt specifications, lack data traceability and standardized guidance, leading to poor operational standardization and greater torque deviation when the environment fluctuates. The system of this invention achieves high torque control accuracy through precise force control and multimodal fusion. The task recognition unit enables automatic parameter switching, the data storage module supports traceability, standardized specifications ensure operational compliance, and environmentally adaptable design guarantees stability. It fully meets the high-precision, standardized maintenance requirements for switchgear bolt tightening, improving maintenance quality and efficiency, and reducing equipment failure risks.

[0052] Reference Figure 2 The line graph clearly illustrates the impact of temperature on sensor measurement accuracy and the environmental adaptability advantages of this invention. Existing sensors exhibit significantly increased errors at low temperatures (-20℃) and high temperatures (60℃), reaching 0.35N and 0.32N respectively, due to the lack of temperature compensation design. This invention, through the platinum resistance temperature compensation circuit in the environmental adaptability module, controls the error to 0.03N~0.08N across the entire temperature range, with the lowest error at 25℃ (room temperature) being only 0.03N, maintaining high accuracy even under extreme temperatures. The data verifies the effectiveness of the special packaging and temperature correction mechanism of the flexible tactile sensor array module, ensuring the perception accuracy of power equipment operation and maintenance in scenarios with seasonal temperature differences and local temperature variations in substations, and avoiding misjudgments of operating force due to temperature.

[0053] Reference Figure 3 The grouped bar chart highlights the torque control accuracy advantage of this invention under different bolt specifications. Existing systems, due to their use of single PID control, fail to consider bolt specification differences and environmental factors, resulting in torque errors generally exceeding 7%, with M16 bolts exhibiting an error of 11.5%, easily leading to bolts being either too loose or too tight. This invention, through dynamic optimization of impedance parameters and multimodal data fusion in the force control algorithm processing module, combined with the operation task identification unit automatically matching the force control parameters for different bolts, controls errors to within 3%, with the lowest error for M10 bolts at only 1.8%. Data demonstrates that this invention can adapt to the bolt tightening requirements of various specifications of power equipment, ensuring the reliability of bolt connections in switchgear and other equipment, and reducing the risk of poor contact or component damage.

[0054] Reference Figure 4The bar chart visually demonstrates the rapid response capability of the fault-tolerant module of this invention. Existing systems lack fault detection and backup switching mechanisms, making them completely unrecoverable in the face of faults such as sensor signal loss and electromagnetic interference, requiring manual intervention and causing maintenance interruptions. This invention rapidly identifies faults through a real-time monitoring module. For example, under high-frequency electromagnetic interference, it can initiate the electromagnetic shielding enhancement program in just 2ms; when sensor signals are lost, it switches to the backup channel in 3ms; and when the force control algorithm malfunctions, it activates the simplified control core in 8ms. All fault recovery times are controlled within 10ms, without affecting the continuity of maintenance operations. The data verifies the design effectiveness of the fault diagnosis and fault-tolerant module, ensuring the stability of power equipment maintenance under complex fault scenarios.

[0055] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A flexible tactile sensor array and force control system for a power equipment maintenance robot, characterized in that, Includes the following modules: Flexible tactile sensor array module: The flexible tactile sensor array is designed using the piezoresistive effect or optical principle to adapt to the irregular surface of power equipment. It achieves environmental resistance through epoxy resin potting and polytetrafluoroethylene coating encapsulation and collects physical signals during operation in real time. Signal conditioning and data acquisition module: Develop a multi-channel sensor signal conditioning circuit to suppress electromagnetic interference in the power environment. Synchronously acquire tactile data through a high-speed data acquisition card, convert it into a digital signal, and then transmit it to the force control algorithm processing module via the CAN bus. Force control algorithm processing module: Based on impedance control theory, a force feedback algorithm is developed, which integrates incremental PID and two-dimensional fuzzy control to optimize dynamic response, constructs an online recognition model of operating force-displacement curve, extracts force change and displacement features, automatically adapts to different operating tasks, calculates force deviation and outputs PWM control commands; Robot System Integration Module: Integrates a flexible tactile sensor array, signal conditioning module, and force control algorithm into the power inspection robot platform. It achieves inter-module communication through dual links, develops motion-force control collaborative linkage program, and matches the robot arm's movement speed with the signal acquisition frequency. Environmental adaptation module: Built-in temperature and humidity compensation and adjustment unit, dynamically corrects sensor parameters. The sensor is treated with anti-fouling insulating coating and has the ability to withstand pressure and resist electromagnetic interference, meeting the safety specifications of power operation and maintenance. Application verification module: For typical power equipment, standardized operation and maintenance processes are preset, and the success rate of operation is ensured through scenario testing, forming a tactile operation and maintenance standard for power equipment.

2. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, It also includes a sensor dynamic calibration module, which is equipped with a piezoelectric force calibration device. By periodically collecting four standard force signals of 0.1N, 10N, 25N and 50N and comparing them with the sensor output signal, a calibration error model based on the least squares method is established, and a calibration coefficient table including temperature and humidity correction terms is generated. The calibration process is automatically started every time the robot starts or the ambient temperature and humidity change exceeds ±5℃ to correct the sensor measurement deviation.

3. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, It also includes a multimodal data fusion module, which collects equipment contour data from robot vision sensors, motion position data from laser displacement sensors, and contact force data from tactile sensors. It achieves synchronization of multi-source data frames through a timestamp alignment algorithm and integrates feature information using a weighted fusion algorithm to construct a three-dimensional state model of power equipment operation.

4. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, The force control algorithm processing module also integrates an impedance parameter dynamic optimization unit to construct an adaptive adjustment mechanism for impedance control parameters. ,in For the desired impedance, Based on the basic stiffness coefficient, Based on the basic damping coefficient, Based on the fundamental inertia coefficient, For the Laplace operator, For real-time ambient temperature, For standard reference temperature, For real-time operation, For target operational force, This is the stiffness correction factor. For real-time relative humidity, For standard reference humidity, To maximize humidity tolerance, Minimum humidity tolerance, This is the damping correction factor.

5. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, The robot system integration module also includes an operation task identification unit, which pre-stores force-displacement feature templates for typical tasks. By collecting force and displacement signals during the operation process in real time, it calculates the cosine similarity with each template. The template with the highest similarity and exceeding 0.9 is determined as the current task type, and the corresponding force control parameter combination is called to complete the adaptive switching of different operation and maintenance tasks.

6. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, It also includes a fault diagnosis and fault tolerance module, which monitors the signal amplitude integrity of the flexible tactile sensor, the transmission frame loss rate of the data acquisition module, and the running error of the force control algorithm in real time. When a sensor signal loss, data transmission error, or force control error exceeds 5% and lasts for 3 consecutive sampling cycles, the fault tolerance mechanism is automatically activated, switching to the backup sensor channel or enabling the simplified force control algorithm. At the same time, the fault information is recorded and an early warning is issued via Ethernet.

7. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, The signal conditioning and data acquisition module also includes a multi-channel synchronous calibration unit to construct a synchronous error correction system for tactile data acquisition. ,in For the data synchronization timestamps of each channel, The timestamp for the first channel is collected. This represents the total number of data acquisition channels. For the first Channel signal transmission delay, For the first Channel signal amplitude, The amplitude of the first channel signal. The maximum signal amplitude, For the minimum signal amplitude, This is the amplitude-dependent delay correction amount. For the first The physical distance between the passage and the first passage. Maximum channel spacing This is the distance-related delay correction amount.

8. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, The environmental adaptation module also includes a vibration and electromagnetic shielding unit. The sensor array is packaged with a multi-layer buffer structure of "silicone shock-absorbing pad + metal frame" to absorb vibration interference in power equipment operation and maintenance scenarios. The signal conditioning circuit and data acquisition module are packaged with a 1.5mm thick cold-rolled steel plate shielding box to effectively suppress high-frequency electromagnetic interference in the substation.

9. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, It also includes a data storage and traceability module, which combines local edge storage with cloud backup. It stores key information during the operation process locally, supports data retrieval by device type, operation date, and task type, and the cloud synchronization storage cycle is ≤1 hour, retaining at least 1 year of historical data.

10. The flexible tactile sensor array and force control system for power equipment operation and maintenance robots according to claim 1, characterized in that, The application verification module also includes a standardized operation specification output unit, which generates a standardized operation process guide based on operation and maintenance test data. The guide includes an equipment model adaptation table, which clarifies the force control parameter range corresponding to different specifications of equipment. It also includes built-in operation step guidance and safety warning logic. When the robot's operation parameters exceed the specified range, it will automatically pause and prompt the robot to adjust its direction through the human-machine interface.

Images