Unmanned aerial vehicle-based spray pressure adaptive adjustment method for water washing of insulators

Abstract

The embodiment of the application provides a method for adaptive adjustment of spray pressure of water washing of insulators based on a UAV, which comprises the following steps: synchronously collecting surface state, environmental working condition and relative pose data of insulators by a multi-source sensing unit carried by the UAV; determining a contamination level according to the surface state data and matching a reference spray pressure; calculating corresponding correction coefficients in combination with the environmental working condition and the relative pose data, and then calling an insulator material adaptation coefficient; substituting all kinds of coefficients and the reference spray pressure into a multi-factor coupling formula to calculate an initial target pressure, determining a target spray pressure after boundary constraint, generating a control instruction to regulate the spray pressure of a high-pressure water washing device, and adjusting the actual spray pressure through closed-loop regulation. The technical scheme provided by the application can realize full-process adaptive dynamic adjustment of washing pressure, and can take into account cleaning effect, operation efficiency and working condition adaptability, so that automatic, low-delay and high-precision insulator washing operation is achieved.

Description

Technical Field

[0001] This invention belongs to the field of live-line working technology for power transmission lines, and particularly relates to an adaptive adjustment method for jet pressure of water flushing of insulators based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Insulators, as core insulating components of transmission lines, are exposed to the outdoor environment for extended periods. Their surfaces easily accumulate dust, salt, oil, and other contaminants, making them highly susceptible to flashover accidents in humid weather, a major cause of line tripping and power outages. Live-line water flushing, with its advantages of uninterrupted power supply and high cleaning efficiency, has become the mainstream technology for insulator contamination control. In recent years, industrial drone technology has rapidly evolved. Drone-borne insulator water flushing operations, with their core advantages of eliminating the need to climb towers, keeping personnel away from live parts, operational flexibility, and adaptability to complex terrain, are gradually replacing traditional manual tower-climbing and insulator bucket truck flushing methods, becoming a core piece of equipment for intelligent power grid operation and maintenance. However, the existing drone-based water flushing technology has a rigid architecture and generally suffers from the following core bottlenecks: traditional solutions only support fixed spray pressure or manual remote gear switching by ground personnel, lacking multi-dimensional adaptive pressure adjustment capabilities; they only provide scattered coverage of single-dimensional data such as dirt identification and distance monitoring, without simultaneously considering the coupled effects of environmental conditions, drone position and attitude, and the inherent properties of insulators, resulting in extremely poor adaptability to complex field conditions; they rely primarily on open-loop control, lacking a closed-loop pressure control mechanism adapted to the dynamic operating characteristics of drones, leading to lag in response and low control accuracy; and they have not constructed a graded safety protection system that complies with the specifications for live-line work on power grids, setting only a single pressure upper limit, which can easily cause safety accidents such as insulator glaze damage, skirt breakage, and live-line flashover, severely restricting the large-scale promotion and application of drone-based live-line water flushing technology.

[0003] Existing technologies, such as CN202511006136.8, a live-line cleaning machine system for high-voltage electrical equipment based on drones, are equipped with a visible light camera and a dirt detection sensor. This system enables basic identification of the dirt type of insulators, allows switching of basic flushing levels according to the dirt type, and optimizes the nozzle layout and flushing trajectory. However, existing technologies suffer from the following unavoidable technical defects due to the lack of a full-link adaptive pressure regulation architecture, the absence of a unified time-series synchronization and fusion processing mechanism for heterogeneous multi-source data, the lack of a high-frequency closed-loop pressure control model adapted to the dynamic operation characteristics of drones, the lack of a hierarchical safety protection system compliant with power grid live-line operation specifications, and the failure to achieve fully automated closed-loop operation: low pressure regulation accuracy and lag response, failing to match the dynamic requirements of insulator contamination levels, resulting in incomplete cleaning or overvoltage damage to insulators; extremely poor adaptability to complex operating conditions, unable to cope with the effects of wind speed, temperature and humidity, and drone attitude fluctuations, leading to problems such as water flow deviation, impact force attenuation, and increased flashover risk; lack of a closed-loop safety protection mechanism, unable to identify risks such as insulator defects and partial discharge, easily causing safety accidents such as insulator breakage and live-line flashover, failing to meet the compliance requirements for power grid live-line operation; and reliance on manual operation, resulting in low operating efficiency and large human error, failing to meet the needs of large-scale intelligent operation and maintenance of transmission lines. Summary of the Invention

[0004] The method for adaptive adjustment of jet pressure for insulator water flushing based on UAV provided in this application embodiment can realize adaptive dynamic adjustment of flushing pressure throughout the entire process, taking into account cleaning effect, operation efficiency and working condition adaptability, and achieving automated, low-latency and high-precision insulator flushing operation.

[0005] In a first aspect, embodiments of this application provide an adaptive adjustment method for jet pressure of insulator water flushing based on unmanned aerial vehicles (UAVs), including:

[0006] S1. Multi-source heterogeneous data is collected synchronously by a multi-source sensing unit mounted on the UAV, and the multi-source heterogeneous data is preprocessed to obtain preprocessed multi-source heterogeneous data. The multi-source heterogeneous data includes surface state data of the target insulator, environmental condition data of the current working environment, and relative pose data between the UAV and the target insulator.

[0007] S2. Based on the pre-processed surface condition data, determine the pollution level of the target insulator and obtain the benchmark injection pressure according to the pollution level;

[0008] S3. Based on the preprocessed environmental condition data, calculate the comprehensive correction coefficient for environmental impact; based on the preprocessed relative pose data, calculate the comprehensive correction coefficient for relative pose; and obtain the material compatibility coefficient corresponding to the target insulator material.

[0009] S4. Substitute the reference injection pressure, the comprehensive correction coefficient for environmental impact, the comprehensive correction coefficient for relative posture, and the material adaptation coefficient into the multi-factor coupled pressure decision formula to calculate the initial target pressure. Then, perform boundary constraint processing on the initial target pressure to obtain the final target injection pressure.

[0010] S5. Generate a pressure control command based on the target injection pressure and send it to the high-pressure water flushing execution unit to adjust the injection pressure of the high-pressure water flushing execution unit;

[0011] S6. The actual injection pressure is collected in real time through the pressure feedback acquisition unit, and the high-pressure water flushing execution unit is closed-loop feedback control is performed based on the deviation between the actual injection pressure and the target injection pressure so that the actual injection pressure approaches the target injection pressure.

[0012] In an optional implementation, in S1, the surface condition data includes a visible light image, an infrared thermal imaging image, and dirt adhesion status data of the target insulator.

[0013] Environmental operating data include wind speed, wind direction, ambient temperature, and relative humidity;

[0014] The relative pose data includes the relative distance between the UAV and the target insulator, the angle between the flushing direction and the normal to the surface of the target insulator, and the flight attitude data of the UAV.

[0015] In an optional implementation, in S2, determining the pollution level of the target insulator based on the pre-processed surface condition data includes:

[0016] Preprocessing of visible light and infrared thermal imaging images to extract image features and temperature rise distribution features of the target insulator surface;

[0017] Image features, temperature rise distribution features, and dirt adhesion status data are input into a pre-trained dirt level assessment model to obtain the dirt level.

[0018] In one optional implementation, in S2, obtaining the baseline injection pressure based on the level of contamination includes:

[0019] Query the preset contamination level-baseline pressure mapping database to obtain the baseline injection pressure corresponding to the contamination level.

[0020] In an optional implementation, in step S3, the comprehensive environmental impact correction coefficient is calculated based on the preprocessed environmental condition data, including:

[0021] Calculate the wind speed compensation coefficient based on wind speed and wind direction;

[0022] Calculate the environmental adaptability coefficient based on ambient temperature and relative humidity;

[0023] The comprehensive environmental impact correction coefficient is calculated by multiplying the wind speed compensation coefficient and the environmental adaptability coefficient.

[0024] In an optional implementation, in step S3, the relative pose comprehensive correction coefficient is calculated based on the preprocessed relative pose data, including:

[0025] Calculate the distance compensation coefficient based on the relative distance;

[0026] Based on the included angle and flight attitude data, the attitude compensation coefficient is calculated;

[0027] The relative pose comprehensive correction coefficient is calculated by multiplying the distance compensation coefficient and the attitude compensation coefficient.

[0028] In one optional implementation, in step S3, obtaining the material compatibility coefficient corresponding to the target insulator material includes:

[0029] The material information of the target insulator is queried from the preset insulator material matching parameter library, and the corresponding material matching coefficient is obtained.

[0030] In an optional implementation, in S4, the multi-factor coupled pressure decision formula is:

[0031] ;

[0032] in, Indicates the initial target pressure. Indicates the reference injection pressure. This represents the comprehensive environmental impact correction factor. This represents the relative pose comprehensive correction coefficient. This indicates the material compatibility factor.

[0033] In an optional implementation, in step S4, boundary constraint processing is applied to the initial target pressure, including:

[0034] The initial target pressure is compared with the preset upper and lower pressure thresholds.

[0035] If the initial target pressure is greater than the upper pressure threshold, then the upper pressure threshold is determined as the target injection pressure;

[0036] If the initial target pressure is less than the lower pressure threshold, then the lower pressure threshold is determined as the target injection pressure;

[0037] If the initial target pressure is greater than or equal to the lower pressure threshold and less than or equal to the upper pressure threshold, then the initial target pressure is determined as the target injection pressure.

[0038] Secondly, embodiments of this application provide an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the method provided in embodiments of this application.

[0039] Thirdly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method provided in embodiments of this application.

[0040] The technical solution provided in this application firstly establishes a quantitative decision-making model based on a baseline pressure plus multi-factor dynamic correction. This model synchronously collects multi-dimensional data such as pollution, environment, position, and material, and performs comprehensive calculations by substituting them into a coupling formula. It quantifies various influencing factors into correction coefficients to dynamically compensate for the baseline pressure. This allows the system to automatically adapt to changes in field variables such as wind speed, distance, and angle, fundamentally solving the problems of water flow deviation, impact force attenuation, or incomplete cleaning caused by variable working conditions in traditional methods. Secondly, it matches the insulator's impact resistance capability with a material coefficient and uses upper and lower threshold values ​​to adjust the calculated pressure. By imposing strict constraints, the system effectively avoids the risks of excessive pressure leading to insulator glaze damage, skirt breakage, or flashover, achieving an optimal balance between cleaning efficiency and equipment and insulation safety. Furthermore, by integrating high-frequency closed-loop pressure control with effect-based parameter self-optimization, and using real-time pressure feedback for PID adjustment, the system ensures that the actual output quickly and accurately tracks the target value. Simultaneously, based on visual feedback of the rinsing effect, the system dynamically optimizes the decision model parameters. This not only reduces reliance on human experience and improves the response speed and accuracy of pressure control, but also enables the system to continuously learn and evolve, enhancing the reliability for large-scale applications. Attached Figure Description

[0041] Figure 1 This is a schematic flowchart of the adaptive adjustment method for jet pressure of insulator water flushing based on UAV provided in the embodiments of this application;

[0042] Figure 2 This is a logic diagram of the adaptive adjustment method for jet pressure of insulator water flushing based on UAV provided in the embodiments of this application;

[0043] Figure 3 This is a schematic diagram of the adaptive adjustment system for jet pressure of water flushing of unmanned aerial vehicle insulators provided in the embodiments of this application. Detailed Implementation

[0044] The present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Figure 1This is a schematic flowchart of the adaptive adjustment method for jet pressure of insulator water flushing based on unmanned aerial vehicles (UAVs) provided in the embodiments of this application. Figure 2 This is a logic diagram of an adaptive adjustment method for jet pressure of water flushing of insulators based on a drone, provided in an embodiment of this application. The method can be executed by an adaptive adjustment system for jet pressure of water flushing of insulators carried by a drone. The system can be implemented by software and / or hardware and can be configured in electronic devices such as computers.

[0046] like Figure 1 and Figure 2 As shown, the technical solution provided in this application includes the following steps:

[0047] S1. Multi-source heterogeneous data is collected synchronously by a multi-source sensing unit mounted on the UAV, and the multi-source heterogeneous data is preprocessed to obtain preprocessed multi-source heterogeneous data. The multi-source heterogeneous data includes surface state data of the target insulator, environmental condition data of the current working environment, and relative pose data between the UAV and the target insulator.

[0048] In some embodiments, system power-on initialization and full-link hardware self-test are performed first. After the device is powered on and started via the UAV's onboard power supply, the core processing unit first performs a full-module hardware self-test and communication link verification to complete the initialization of the system and the core decision formula's operating environment. The specific implementation process is as follows:

[0049] Power-on status checks were performed on all modules of the core control layer, multi-source perception layer, execution control layer, interface interaction layer, and storage and configuration management layer in sequence to confirm that there were no hardware abnormalities in each module and that the communication links were connected normally.

[0050] The high-pressure water flushing actuator and the pressure feedback acquisition unit were tested for connectivity and zero-point calibration to confirm that the pipeline was in normal condition and the pressure acquisition data was valid.

[0051] Complete the initialization of system operating parameters and core decision formulas, configure multi-source data time-series synchronization rules, pollution level assessment model operating parameters, calibration coefficients of multi-factor coupled pressure decision formulas, closed-loop control parameters, and work status triggering rules, and retrieve the locally stored pollution level-benchmark pressure mapping database and insulator material adaptation parameter library;

[0052] When the self-test result is abnormal, the system issues a prompt through the status indicator component and records the abnormal information in the local storage module; after all self-tests are normal, the system enters a low-power standby mode.

[0053] In some embodiments, the system first uses its pose sensing unit to confirm whether the relative position of the UAV and the target insulator is within a preset optimal operating range, and simultaneously confirms whether the angle between the flushing water flow direction and the normal direction of the insulator surface is within a preset optimal angle range. If any condition is not met, the system sends a pose adjustment command to the UAV's flight control system, driving the UAV to adjust to a suitable operating position and attitude to ensure the effectiveness of subsequent data acquisition and the safety and effectiveness of the flushing operation.

[0054] Once the drone has hovered stably in a suitable position, the core control unit controls the various functional units of the multi-source sensing layer to synchronously collect the following three types of multi-source heterogeneous data according to the preset synchronous acquisition frequency:

[0055] Surface condition data of the target insulator: collected through the insulator condition sensing unit, specifically including:

[0056] Visible light images: used for visual analysis of the distribution and type of contamination on insulator surfaces;

[0057] Infrared thermal imaging images: used to obtain temperature distribution data on the surface of insulators to help determine the pollution status and partial discharge hotspots;

[0058] Dirt adhesion status data: Quantified dirt information acquired by a dedicated dirt detection sensor.

[0059] Current working environment environmental condition data: collected through the environmental sensing unit, specifically including:

[0060] Wind speed: Affects the trajectory of the water jet in the air and the attenuation of its impact force;

[0061] Wind direction: The angle between the wind direction and the water flow direction determines the effective headwind component;

[0062] Ambient temperature: affects the evaporation rate of water droplets.

[0063] Relative humidity: affects the insulation strength of air and the properties of water droplets.

[0064] Relative pose data between the UAV and the target insulator: collected through a pose sensing unit, specifically including:

[0065] Relative distance: The straight-line distance between the drone nozzle and the surface of the target insulator;

[0066] The angle between the flushing direction and the normal to the surface of the target insulator: that is, the water inflow angle, which affects the effective impact area;

[0067] The flight attitude data of the drone, such as pitch, roll, yaw angles and their fluctuation amplitude, reflects the stability of the drone while hovering.

[0068] Furthermore, to address the issues of noise interference, inconsistent formats, and misaligned timing in the aforementioned multi-source heterogeneous data, preprocessing of the collected multi-source heterogeneous data is also required, including:

[0069] Denoising, distortion correction, and target region segmentation are performed on the insulator surface image data to extract the image data of the effective shed area of ​​the insulator and filter out background interference; non-uniformity correction and noise filtering are performed on the infrared temperature distribution data to extract the temperature rise distribution characteristics of the insulator surface.

[0070] Smoothing filtering is performed on the environmental condition data of the current working environment and the relative pose data between the UAV and the target insulator, which are time-series sensor data, to remove outliers caused by sudden interference, and the data is normalized to unify the data format.

[0071] Based on timestamps, spatial registration and feature fusion are performed on all preprocessed multi-source data to form a unified operational condition feature dataset, which is then synchronously transmitted to the pollution level assessment unit and the multi-factor stress calculation unit.

[0072] The merged data is validated to remove invalid data. If core data is missing, the operator is prompted to adjust the drone's attitude or check the status of the sensor components.

[0073] Simultaneously, the system synchronously retrieves the inherent attribute parameters of the target insulator from the local database of the storage and configuration management layer. These parameters are pre-stored known information, including:

[0074] Insulator material (such as ceramic, composite silicone rubber), skirt structure, rated voltage level, upper and lower safety pressure thresholds, etc.

[0075] S2. Based on the pre-processed surface condition data, determine the pollution level of the target insulator and obtain the reference injection pressure according to the pollution level.

[0076] The image features, temperature rise distribution features, and pollution adhesion status data obtained after the above preprocessing are input into a pre-trained deep learning pollution level assessment model. This model analyzes information such as image texture, pollution color distribution, area ratio, and abnormal local temperature rise to comprehensively judge and output the pollution type (such as dust, salt density, etc.) and pollution level (for example, quantifying the pollution degree into levels I, II, III or specific equivalent salt density values) of the target insulator.

[0077] After obtaining the pollution level, the system queries the pre-built pollution level-baseline jet pressure mapping database in the local storage and configuration management layer. This database is pre-built based on industry standards for transmission line insulator operation and maintenance and a large amount of field flushing test data, establishing a mapping relationship between the optimal cleaning pressure ranges corresponding to different pollution types and pollution levels. Among them, the baseline jet pressure refers to the basic cleaning pressure that matches the current insulator pollution level and type under standard operating conditions, and is the core benchmark value for achieving the cleaning effect.

[0078] Based on the type and level of contamination output by the evaluation model, the system searches and matches the corresponding baseline injection pressure value from the aforementioned mapping database.

[0079] Furthermore, while determining the reference spray pressure, the system also extracts the characteristics of the contamination distribution on the insulator surface, marking areas where contamination is concentrated or particularly severe, providing a basis for subsequent dynamic fine-tuning of the flushing trajectory and pressure.

[0080] S3. Based on the preprocessed environmental condition data, calculate the comprehensive correction coefficient for environmental impact; based on the preprocessed relative pose data, calculate the comprehensive correction coefficient for relative pose; and obtain the material compatibility coefficient corresponding to the target insulator material.

[0081] In some embodiments, the comprehensive environmental impact correction factor is used to compensate for the combined effects of environmental factors such as wind speed, ambient temperature, and relative humidity on the attenuation of water flow impact force, water droplet evaporation, and the insulation safety of live-line work. The calculation formula is as follows:

[0082] ;

[0083] in, This represents the comprehensive environmental impact correction factor. This represents the wind speed compensation coefficient. This represents the environmental adaptability coefficient.

[0084] The calculation of the wind speed compensation coefficient includes: calculating the effective headwind speed (i.e., the wind speed component opposite to the water flow direction) based on real-time wind speed and direction collected by the environmental sensing unit, and assigning a value according to the calibration rules determined by the aerodynamic characteristics of the water jet and field test data: when the effective headwind speed is in the light wind range, =1 (no compensation required); after the effective headwind speed exceeds the light wind range. The value increases with wind speed to compensate for the deviation in water flow trajectory and the attenuation of kinetic energy caused by the wind. The logic for determining a "light wind zone" is that the influence of wind on the water jet trajectory and kinetic energy at that wind speed is negligible. By default, an effective headwind speed ≤ 2 m / s can be defined as a light wind zone. If the equipment is large or the jet pressure is high, the light wind zone can be adaptively increased according to the actual situation. When the effective headwind speed reaches the upper limit of the allowable operation... The preset maximum value is locked. The preset maximum value of the wind speed compensation coefficient can be determined jointly based on the rated working pressure of the high-pressure water pump, water pipeline and nozzle, and the safety impact pressure of the insulator. It can be adjusted according to different voltage levels, pressure conditions and other actual conditions.

[0085] The calculation of the environmental adaptability coefficient includes: based on real-time ambient temperature and relative humidity collected by the environmental sensing unit, and using standard operating conditions as a benchmark, assigning values ​​according to calibration rules determined by the water droplet evaporation characteristics and the insulation safety requirements for live-line work: In high-temperature and low-humidity environments, water droplets evaporate quickly, requiring an appropriate increase in pressure to ensure the water column reaches the insulator surface. >1; In low temperature and high humidity environments, no compensation may be required or the coefficient may be slightly less than 1.

[0086] In some embodiments, the relative pose comprehensive correction coefficient is used to compensate for the relative distance and angle between the UAV and the target insulator, as well as the changes in water flow energy dispersion and effective impact area caused by the UAV's own attitude fluctuations. The calculation formula is as follows:

[0087] ;

[0088] in, This represents the relative pose comprehensive correction coefficient. Indicates the distance compensation coefficient. This represents the attitude compensation coefficient.

[0089] The calculation of the distance compensation coefficient includes: based on the real-time relative distance between the UAV and the insulator obtained by the pose sensing unit, and using a preset standard operating distance as a benchmark, assigning a value according to the attenuation law of the water jet in the air and the calibration rules determined by the field test data: when the relative distance is less than or equal to the standard operating distance, =1 (no compensation required). In this embodiment, the standard operating distance is set to 2m according to the technical specifications, but this can be adjusted according to actual conditions. If the distance exceeds this, The value increases with distance to compensate for the kinetic energy loss of the water column due to air resistance; when the maximum allowable operating distance is reached, The preset maximum value is locked. The preset maximum value of the distance compensation coefficient is determined by the maximum allowable working distance, the rated working pressure of the equipment, and the safety impact pressure of the insulator. It can be calibrated on-site according to the system equipment parameters.

[0090] The calculation of the attitude compensation coefficient includes: based on the angle between the flushing direction and the normal to the insulator surface obtained by the pose sensing unit, and the flight attitude fluctuation data of the UAV (such as the fluctuation amplitude of pitch and roll angles), the coefficient is assigned a value according to the calibration rules determined by the effective impact area of ​​the water flow and the energy dispersion law: the larger the angle (deviation from vertical flushing), the greater the amplitude of the UAV attitude fluctuation; the smaller the effective impact area of ​​the water flow, the more dispersed the energy; and the higher the required coefficient. The higher the value, the more necessary it is to prevent excessive compensation from causing damage to the insulator due to excessive pressure. The upper limit value is set based on the effective impact area of ​​the water flow, the amplitude of the drone's flight attitude fluctuation, and the impact resistance of the insulator. It can be adjusted according to the actual situation, such as the type of insulator at the actual work site.

[0091] In some embodiments, the insulator material compatibility coefficient is used to match the physical impact resistance and surface contamination adhesion characteristics of insulators made of different materials. Its core purpose is to ensure cleaning effectiveness while preventing damage to insulators (such as the glaze of porcelain insulators and the skirts of composite insulators) due to excessive pressure. Based on the inherent property parameters of the target insulator (especially the insulator material, such as ceramic or composite silicone rubber) retrieved from local storage, the corresponding insulator material compatibility coefficient is obtained by directly querying the pre-set insulator material compatibility parameter library in the storage and configuration management layer. The value. For example, for porcelain insulators with high mechanical strength and impact resistance, The value can be set to 1.0 or higher; for composite insulators with weaker impact resistance, The value will be set to less than 1.0 to actively reduce impact pressure and provide safety protection.

[0092] S4. Substitute the reference injection pressure, the comprehensive correction coefficient for environmental impact, the comprehensive correction coefficient for relative posture, and the material adaptation coefficient into the multi-factor coupled pressure decision formula to calculate the initial target pressure. Then, perform boundary constraint processing on the initial target pressure to obtain the final target injection pressure.

[0093] In some embodiments, the multi-factor coupled pressure decision formula is as follows:

[0094] ;

[0095] in, Indicates the initial target pressure. Indicates the reference injection pressure. This represents the comprehensive environmental impact correction factor. This represents the relative pose comprehensive correction coefficient. This indicates the material compatibility factor.

[0096] Furthermore, to ensure operational safety and prevent pressure from exceeding the equipment or insulator's tolerance range due to calculation errors or extreme operating conditions, safety constraints must be imposed on the initial target pressure. The boundary constraint formula is as follows:

[0097] ;

[0098] in, Indicates the target injection pressure. This indicates the upper pressure threshold, derived from the smaller of the retrieved insulator's inherent property parameter (safe pressure upper limit) or the equipment's rated operating pressure. This indicates the lower pressure threshold, which is the minimum pressure value required to ensure basic cleaning effect or stable system operation.

[0099] S5. Generate a pressure control command based on the target injection pressure and send it to the high-pressure water flushing execution unit to adjust the injection pressure of the high-pressure water flushing execution unit.

[0100] In some embodiments, based on the finally determined The core control layer breaks it down and generates specific hardware execution instructions, which typically include:

[0101] Pump operation control commands: used to control the pump speed or power, and to achieve coarse pressure adjustment;

[0102] Pressure regulating valve opening control command: used to finely adjust the valve opening to achieve pressure fine-tuning.

[0103] Before sending a command, the system verifies the validity of the generated target pressure command. If an abnormal command is detected (such as exceeding the limit or a sharp change), the system will maintain the current stable pressure output and issue an alarm to avoid operation interruption or danger caused by erroneous commands.

[0104] The two generated control commands are sent in real time from the core control layer to the corresponding functional units of the execution control layer through the system's highly reliable internal bus. The water pump drive control unit receives the water pump operation control command, and the high-pressure water flushing execution unit (including the pressure regulating valve actuator) receives the pressure regulating valve opening control command.

[0105] Upon receiving an instruction, each unit in the execution control layer immediately drives the hardware device to perform the action:

[0106] The water pump drive control unit adjusts the motor drive signal of the high-pressure water pump connected to it in real time according to the instructions, thereby changing the output power and flow rate of the water pump.

[0107] The pressure regulating valve in the high-pressure water flushing actuator adjusts the position of its internal valve core according to the command, by a motor or electromagnetic drive mechanism, thereby changing the cross-sectional area of ​​the water passage and achieving precise adjustment of the pipeline resistance.

[0108] The coordinated action of the water pump (as a pressure source) and the pressure regulating valve (as a resistance regulator) alters the flow and resistance characteristics of the high-pressure water pipeline system, thereby shifting the actual injection pressure of the system from its current value towards... The direction was initially adjusted.

[0109] S6. The actual injection pressure is collected in real time through the pressure feedback acquisition unit, and the high-pressure water flushing execution unit is closed-loop feedback control is performed based on the deviation between the actual injection pressure and the target injection pressure so that the actual injection pressure approaches the target injection pressure.

[0110] In some embodiments, after the execution control unit receives the target pressure command, it adjusts the water pump operation status by driving the water pump control unit and adjusts the valve opening by adjusting the pressure regulating valve to complete the initial pressure adjustment.

[0111] The pressure feedback acquisition unit collects the actual injection pressure data in the pipeline in real time according to the preset sampling frequency and feeds it back to the PID closed-loop control module of the core control layer.

[0112] The closed-loop control module calculates the deviation between the actual injection pressure and the target injection pressure output by the formula according to the preset control cycle. Based on the calculated pressure deviation, and usually combined with the historical changes (integral) and future trends (derivative) of the deviation, the closed-loop control module performs calculations according to the preset PID (proportional-integral-derivative) control algorithm, and outputs incremental control signals to the water pump drive control unit and the pressure regulating valve to achieve fine dynamic adjustment of the pressure and ensure that the deviation between the actual output pressure and the target pressure is controlled within the preset accuracy range.

[0113] The system monitors the pressure control status in real time. If there are situations such as continuous pressure deviation or abnormal fluctuations in pipeline pressure, it automatically performs parameter self-tuning and status calibration to ensure stable pressure output.

[0114] In some embodiments, during the rinsing operation, the rinsing effect is verified in real time and the parameters of the core decision formula are dynamically optimized to form a closed-loop optimization throughout the entire process, continuously improving the formula's scenario adaptability and decision accuracy. The specific implementation process is as follows:

[0115] The system continuously collects insulator surface condition data during the flushing process through the insulator condition sensing unit, and verifies the dirt cleaning effect in real time.

[0116] If it is detected that dirt in a local area has not been effectively removed, the target spray pressure at the corresponding rinsing point will be automatically increased, and the rinsing trajectory will be adjusted simultaneously to complete the supplementary spraying operation; if it is detected that the dirt has been completely removed, the pressure will be automatically reduced to the maintenance range to complete the final rinsing.

[0117] Based on the feedback on the flushing effect of this operation, the system automatically optimizes the dirt level-baseline pressure mapping parameters and the calibration coefficients in the multi-factor correction formula. The optimized parameters are then updated to the local database to continuously improve the accuracy of the pressure decision formula and its adaptability to complex working conditions.

[0118] In some embodiments, during the operation, system operation status optimization and full archiving of job data are performed simultaneously. The specific implementation process is as follows:

[0119] Dynamic energy consumption control: After the system completes the flushing operation of a single set of insulators and enters the standby state, the core control unit dynamically controls the power supply of each module, shuts down the power supply of unnecessary computing modules and sensing components, and reduces the overall energy consumption of the machine. In the long-term standby state, only the power supply of the core control unit and the status monitoring unit is retained, and the other modules enter the power-off sleep state to achieve deep low-power standby.

[0120] Full data archiving: After each pressure adjustment and flushing operation is completed, the local storage module automatically archives all data of this operation, including operation time, target insulator information, pollution level, full parameters of formula input and output, target pressure, actual pressure change curve, environmental conditions, and operation effect parameters, which facilitates subsequent operation review and formula optimization.

[0121] In some embodiments, after the system completes all insulator flushing operations and receives a return-to-base shutdown command, it executes an orderly shutdown process to ensure equipment stability and prevent loss of optimized parameters. The specific implementation process is as follows:

[0122] Upon receiving the task completion instruction, the system first controls the high-pressure water pump to stop running, adjusts the pressure regulating valve to the fully open state to complete the pipeline depressurization, and at the same time stops the data acquisition operation of all sensing components;

[0123] The system configuration parameters, job logs, historical running data, and formula calibration and optimization parameters are completely saved to the local storage module to ensure that no data is lost.

[0124] The power supply circuits of each module in the execution control layer, interface interaction layer, and multi-source perception layer are shut down in sequence, leaving only the basic power supply to the core control unit and storage management unit.

[0125] After the drone returns to base and lands, it receives a shutdown command. The system executes a complete power-off process, shutting down all power supply circuits and completing the shutdown.

[0126] Figure 3 This is a schematic diagram of the adaptive adjustment system for jet pressure of water flushing of insulators on a UAV provided in an embodiment of this application. This embodiment also provides an adaptive adjustment system for jet pressure of water flushing of insulators on a UAV, which includes: a core control layer, a multi-source sensing layer, an execution control layer, an interface interaction layer, and a storage and configuration management layer. Each layer communicates with each other through an internal bus, as detailed below:

[0127] The core control layer, including the programmable logic core processing unit, is the system's core computation and scheduling hub, and also the hardware unit carrying the multi-factor coupled pressure decision calculation formula. It employs programmable logic devices with parallel processing capabilities, responsible for the system's timing synchronization control, multi-source data fusion processing, pressure decision formula calculation, scheduling logic execution, closed-loop control calculation, system status monitoring, and operation scheduling. The unit incorporates pipelined parallel processing logic, a multi-source data timing synchronization module, an insulator pollution level assessment model, a multi-factor pressure correction calculation unit, a switching control state machine, and a PID closed-loop control module, enabling parallel processing of multiple heterogeneous signals to ensure the real-time performance of pressure decision calculations and the synchronization and accuracy of the pressure regulation process.

[0128] The multi-source sensing layer provides full real-time input data for the stress decision-making formula. It contains three core functional units, and all collected data is timestamped to ensure time-series synchronization.

[0129] Insulator condition sensing unit: Composed of visible light imaging component, infrared thermal imaging component, and dirt detection component, it is coaxially installed with flushing actuator to collect surface images, temperature distribution data, and dirt adhesion status data of target insulator in real time, providing data support for the benchmark injection pressure, the core input item of the decision formula.

[0130] Environmental sensing unit: Composed of wind speed and direction detection components and temperature and humidity detection components, it is installed in an unobstructed position on the drone body to collect real-time data on wind speed, wind direction, ambient temperature, and relative humidity at the operation site, providing a basis for calculating the comprehensive environmental impact correction coefficient of the decision formula.

[0131] The pose perception unit consists of a distance detection component, a UAV attitude sensing component, and a high-precision positioning component. It collects the relative distance, relative angle, flight attitude fluctuation data, and spatial positioning data between the UAV and the target insulator in real time, providing a basis for calculating the relative pose comprehensive correction coefficient of the decision formula.

[0132] The execution control layer is the execution terminal of the decision formula output. It is responsible for receiving the target pressure command output from the core control layer, accurately adjusting and stabilizing the flushing pressure, and collecting actual operating data to achieve closed-loop feedback. It includes four core functional units:

[0133] High-pressure water flushing execution unit: It consists of a high-pressure water pump, pressure regulating valve, flushing nozzle, and water supply pipeline. It is responsible for regulating and stabilizing the flushing water pressure according to control commands to achieve flushing operations at different pressure levels.

[0134] Water pump drive control unit: It is connected to the high-pressure water pump, receives control signals from the core control unit, and adjusts the output power and operating status of the water pump in real time to achieve coarse control of the flushing pressure.

[0135] Pressure feedback acquisition unit: Composed of multiple pressure detection components, which are installed at the water pump outlet and the front end of the nozzle pipeline respectively, to collect the actual injection pressure data in the pipeline in real time and feed it back to the PID closed-loop control module of the core control layer to realize closed-loop feedback regulation of pressure.

[0136] Flight control collaborative interaction unit: It communicates with the UAV flight control system in real time through a standardized bus, synchronously acquires the UAV's flight status and attitude data, and sends operation status and flight adjustment commands to the flight control system to achieve coordinated linkage between flushing control and flight control.

[0137] The interface interaction layer serves as the system's internal and external communication channel, responsible for receiving job instructions, transmitting status feedback, and expanding peripherals. It comprises three functional units:

[0138] Host communication interface unit: It is equipped with multiple standardized communication interfaces to achieve bidirectional communication with the UAV flight control system and the ground control station, and to complete the reception of operation instructions, the feedback of operation status, and the real-time transmission of sensing data and video streams.

[0139] Switching control and status indication unit: Supports two control triggering methods: remote software commands from the ground station and onboard mechanical buttons. It is also equipped with multiple status indication components to provide real-time feedback on system power supply, operating status, pressure status and other operating information.

[0140] Peripheral expansion interface unit: Reserves multiple standardized peripheral interfaces, which can be compatible with the connection of additional detection and sensing components, auxiliary operation components and other peripherals, thereby improving the system's scenario scalability and functional compatibility.

[0141] The storage and configuration management layer provides basic parameter support and data archiving capabilities for core decision-making formulas. It is responsible for system configuration management, basic database storage, and job data archiving, and comprises two functional units:

[0142] Parameter and Database Management Unit: Responsible for storing system configuration parameters, pollution level-baseline pressure mapping database, insulator material adaptation parameter library, equipment rated parameter library, and decision formula calibration coefficient library. It supports online configuration and updates of parameters and provides basic data support for multi-factor pressure decision formulas.

[0143] Local storage and log management unit: It adopts non-volatile storage chip and is responsible for storing job logs, running status records, historical job data, and formula calculation process data. All data is stored locally on the device and supports local log export, which facilitates job review, formula coefficient calibration and equipment operation and maintenance management.

[0144] The execution process of the system part of this application embodiment is the same as that of the method part of the embodiment described above, and will not be repeated here.

[0145] This application also provides an electronic device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform any of the following: an adaptive adjustment method for jet pressure of water flushing of insulators based on unmanned aerial vehicles.

[0146] This application also proposes a computer storage medium storing a computer program, which, when executed by a processor, implements any one of the following methods for adaptive adjustment of jet pressure for water flushing of insulators based on unmanned aerial vehicles (UAVs).

[0147] Computer storage media may be simply referred to as media. Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Dual Data SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM). The various embodiments described in this specification are presented in a progressive manner, with reference allowed to each other for similar or identical parts. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatuses, devices, and non-volatile computer storage media are described simply because they are substantially similar to the method embodiments; relevant details can be found in the descriptions of the method embodiments.

[0148] The above embodiments are merely illustrative examples and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this application.

Claims

1. A method for adaptive adjustment of jet pressure for water flushing of insulators based on unmanned aerial vehicles (UAVs), characterized in that, include: S1. Multi-source heterogeneous data is collected synchronously by a multi-source sensing unit mounted on the UAV, and the multi-source heterogeneous data is preprocessed to obtain preprocessed multi-source heterogeneous data; the multi-source heterogeneous data includes surface state data of the target insulator, environmental condition data of the current working environment, and relative pose data between the UAV and the target insulator. S2. Based on the pre-processed surface condition data, determine the pollution level of the target insulator, and obtain a reference injection pressure according to the pollution level; S3. Based on the preprocessed environmental condition data, calculate the comprehensive environmental impact correction coefficient; Based on the preprocessed relative pose data, calculate the relative pose comprehensive correction coefficient; and obtain the material adaptation coefficient corresponding to the target insulator material. S4. Substitute the reference injection pressure, the comprehensive correction coefficient for environmental impact, the comprehensive correction coefficient for relative pose, and the material adaptation coefficient into the multi-factor coupled pressure decision formula to calculate the initial target pressure, and perform boundary constraint processing on the initial target pressure to obtain the final target injection pressure. S5. Generate a pressure control command based on the target injection pressure and send it to the high-pressure water flushing execution unit to adjust the injection pressure of the high-pressure water flushing execution unit; S6. The actual injection pressure is collected in real time by the pressure feedback acquisition unit, and the high-pressure water flushing execution unit is closed-loop feedback control is performed based on the deviation between the actual injection pressure and the target injection pressure, so that the actual injection pressure approaches the target injection pressure.

2. The method according to claim 1, characterized in that, In step S1, the surface condition data includes visible light images, infrared thermal imaging images, and dirt adhesion status data of the target insulator. The environmental operating data includes wind speed, wind direction, ambient temperature, and relative humidity; The relative pose data includes the relative distance between the UAV and the target insulator, the angle between the flushing direction and the normal to the surface of the target insulator, and the flight attitude data of the UAV.

3. The method according to claim 2, characterized in that, In step S2, determining the pollution level of the target insulator based on the preprocessed surface state data includes: The visible light image and the infrared thermal imaging image are preprocessed to extract the image features and temperature rise distribution features of the target insulator surface; The image features, the temperature rise distribution features, and the dirt adhesion status data are input into a pre-trained dirt level assessment model to obtain the dirt level.

4. The method according to claim 1, characterized in that, In step S2, the baseline injection pressure is obtained based on the level of contamination, including: Query the preset contamination level-reference pressure mapping database to obtain the reference injection pressure corresponding to the contamination level.

5. The method according to claim 2, characterized in that, In step S3, the comprehensive environmental impact correction coefficient is calculated based on the preprocessed environmental condition data, including: Calculate the wind speed compensation coefficient based on the wind speed and wind direction; Calculate the environmental adaptability coefficient based on the ambient temperature and relative humidity; The comprehensive environmental impact correction coefficient is calculated by multiplying the wind speed compensation coefficient and the environmental adaptability coefficient.

6. The method according to claim 2, characterized in that, In step S3, the relative pose comprehensive correction coefficient is calculated based on the preprocessed relative pose data, including: Based on the relative distance, calculate the distance compensation coefficient; Based on the included angle and the flight attitude data, calculate the attitude compensation coefficient; The relative pose comprehensive correction coefficient is calculated based on the product of the distance compensation coefficient and the attitude compensation coefficient.

7. The method according to claim 1, characterized in that, In step S3, obtaining the material compatibility coefficient corresponding to the target insulator material includes: The material information of the target insulator is queried from the preset insulator material adaptation parameter library, and the corresponding material adaptation coefficient is obtained.

8. The method according to claim 1, characterized in that, In S4, the multi-factor coupled pressure decision-making formula is: ； in, Indicates the initial target pressure. This indicates the reference injection pressure. This represents the comprehensive environmental impact correction factor. This represents the relative pose comprehensive correction coefficient. This indicates the material compatibility coefficient.

9. The method according to claim 8, characterized in that, In step S4, the initial target pressure is subjected to boundary constraint processing, including: The initial target pressure is compared with preset upper pressure threshold and lower pressure threshold; If the initial target pressure is greater than the upper pressure threshold, then the upper pressure threshold is determined as the target injection pressure; If the initial target pressure is less than the lower pressure threshold, then the lower pressure threshold is determined as the target injection pressure; If the initial target pressure is greater than or equal to the lower pressure threshold and less than or equal to the upper pressure threshold, then the initial target pressure is determined as the target injection pressure.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed in a computer, causes the computer to perform any one of the methods described in claims 1 to 9.

Images