Inspection robot control method and system based on deep learning

Abstract

The invention relates to the technical field of intelligent control, in particular to an inspection robot control method and system based on deep learning, and aims to improve the reliability and safety of a power transmission system and reduce the risk of interruption and failure of the power system. The method comprises the following steps: carrying out unit division processing on a cable to be inspected to obtain a plurality of cable units; controlling an inspection robot to inspect the cable according to the inspection plan time, and obtaining state information of each cable unit by using the inspection robot; the state information comprises the number of cable strand breakage times in the cable unit, the average temperature of the cable, the average gray value of the cable image and the maximum sag distance; arranging the state information of the plurality of cable units collected by the inspection robot according to the connection sequence, and aligning the cable strand breaking times, the average temperature of the cable, the average gray value of the cable image and the maximum sag distance to obtain a cable state characteristic matrix;

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and in particular to a patrol robot control method and system based on deep learning. Background Art

[0002] Inspection of high-voltage overhead cables is a key step to ensure safe operation and reliable power supply of power transmission systems. Regular inspection and maintenance can reduce the risk of power system interruptions and failures. Using drone inspection robots to inspect high-voltage overhead cables is an advanced and efficient method that can improve safety, reduce costs and reduce human risks. It brings significant improvements to the reliability and safety of power systems. The high-definition camera carried by the drone inspection robot can clearly observe the operating status and fault type of high-voltage transmission lines.

[0003] However, when existing inspection robots inspect high-voltage overhead cables, they mostly perform inspection operations according to pre-planned plans. It is difficult for the inspection robots to independently set the next inspection time node based on the actual inspection results of the cables. When the cable condition deteriorates during the window period of the inspection plan, it is easy to miss the best time for cable repair. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a patrol robot control method based on deep learning, which improves the reliability and safety of the power transmission system and reduces the risk of power system interruption and failure.

[0005] In a first aspect, the present invention provides a control method for an inspection robot based on deep learning, the method comprising: The cables to be inspected are divided into units to obtain multiple cable units; According to the inspection plan time, the inspection robot is controlled to inspect the cables, and the status information of each cable unit is obtained by the inspection robot; the status information includes the number of cable strand breaks in the cable unit, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; The status information of several cable units collected by the inspection robot is arranged in the order of connection, and the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image and the maximum sag distance are aligned to obtain the cable status feature matrix; Obtaining the voltage level of the cable, and according to the voltage level of the cable, calling up a cable state analysis model corresponding to the voltage level; The cable status feature matrix is ​​input into the cable status analysis model corresponding to the voltage level, and the cable status analysis model outputs the time node for the next cable inspection; Compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the cable in the inspection plan: If the inspection time node output by the cable status analysis model is earlier than the time node of the next inspection of the cable in the inspection plan, the next inspection time will be the inspection time node output by the cable status analysis model, and the inspection plan will be updated accordingly; If the inspection time node output by the cable status analysis model is later than the time node of the next inspection of the cable in the inspection plan, the next inspection time will still be carried out according to the time node in the inspection plan.

[0006] Furthermore, the cable state characteristic matrix is: Among them, B n Indicates the number of times the wire strands are broken in the nth cable unit; the T n represents the average temperature of the nth cable unit; the C n represents the average grayscale value of the image of the nth cable unit; n represents the maximum sag distance of the nth cable unit; the elements of each row in the cable state characteristic matrix represent different state quantified characteristic values ​​of the cable segment between two adjacent cable towers.

[0007] Furthermore, the method for dividing the cables to be inspected into units includes: Divide the cables between two adjacent cable towers into one cable unit; Each cable unit is marked with a unique identifier for inspection robots to identify and track; The geographical location information of each cable unit is associated with the corresponding cable unit and saved for navigation and positioning by the inspection robot.

[0008] Furthermore, the method of controlling the inspection robot to inspect the cable includes: According to the pre-established inspection plan, determine the time to inspect each cable unit; According to the geographical location information of the cable unit, the inspection robot is positioned to the cable unit that needs to be inspected; Control the inspection robot to collect status information of the cable unit.

[0009] Furthermore, the method for controlling the inspection robot to collect status information of the cable unit includes: Use image sensors to obtain images of the cable unit surface, use image processing technology to compare the cable line in the image with the line in normal state to perform difference analysis to detect broken strands in the image, and use algorithms to mark and count the number of broken strands to obtain the number of broken strands in the cable unit; Use a temperature sensor or a thermal imager to measure the surface temperature of multiple locations on the cable unit, and obtain the average temperature of the cable unit by statistically averaging the measured values; An image sensor is used to obtain a cable unit surface image, pixel values ​​of the cable unit surface image are converted into gray values, and an average gray value of all pixel gray values ​​is calculated to obtain an average gray value of the cable unit image; Use laser ranging technology to determine the lowest point of the cable unit, and calculate the continuous vertical distances between the lowest point of the cable unit and the two end points of the cable unit to obtain the maximum sag distance of the cable unit.

[0010] Furthermore, the method for constructing the cable status analysis model includes: Collect high-voltage overhead cable inspection sample data, including cable status information under different states, wherein the cable status information includes the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; Preprocess the collected sample data, including data cleaning, standardization, and normalization; and convert the sample data into the cable status feature matrix format; Based on the evaluation of domain experts, the cable status information is associated with the inspection time to determine when each cable status information item needs to be inspected when it reaches the set threshold; Based on deep learning technology, a model architecture is built to handle cable status analysis problems; Use the preprocessed sample data to train and optimize the selected deep learning model; The cable status analysis model is deployed in an actual inspection system for testing to verify the model's predictive and generalization capabilities.

[0011] Furthermore, the model architecture of the cable status analysis model selects a convolutional neural network, a recurrent neural network or a long short-term memory network.

[0012] On the other hand, the present application also provides a patrol robot control system based on deep learning, the system comprising: A division processing module is used to divide the cables that need to be inspected into units to obtain multiple cable units; The inspection control module is used to control the inspection robot to inspect the cables according to the inspection plan time, and use the inspection robot to obtain the status information of each cable unit; the status information includes the number of cable strand breaks in the cable unit, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; The data processing module is used to arrange the status information of several cable units collected by the inspection robot in the order of connection, and align the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance, respectively, to obtain a cable status feature matrix; The model calling module is used to obtain the voltage level of the cable and, according to the voltage level of the cable, call the cable state analysis model corresponding to the voltage level; A state analysis module is used to input the cable state characteristic matrix into a cable state analysis model corresponding to the voltage level, and the cable state analysis model outputs the time node for the next cable inspection; The inspection plan update module is used to compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the cable in the inspection plan: If the inspection time node output by the cable status analysis model is earlier than the time node of the next inspection of the cable in the inspection plan, the next inspection time will be the inspection time node output by the cable status analysis model, and the inspection plan will be updated accordingly; If the inspection time node output by the cable status analysis model is later than the time node of the next inspection of the cable in the inspection plan, the next inspection time will still be carried out according to the time node in the inspection plan.

[0013] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: through unit division processing, the inspection work of the overhead bare wire cable is refined to each cable unit, which improves the refinement of the inspection and helps to more accurately reflect the actual status of the cable; the status information of each cable unit is collected, including the number of broken strands, average temperature, average gray value and maximum sag distance, etc., which provides comprehensive data support for subsequent status analysis and prediction.

[0016] By using a deep learning model to analyze the cable status feature matrix, the health status of the cable can be automatically determined, and the next inspection time node can be output, reducing the tedious work and errors of manually formulating inspection plans. When comparing the inspection time nodes, flexible adjustments can be made according to the actual situation to ensure the timeliness of the inspection and maintain the stability of the inspection plan. The method is highly versatile and scalable, and can be widely used in overhead bare wire cable inspection operations of various voltage levels. Through automated inspections and intelligent analysis, the inspection robot can autonomously adjust the inspection plan according to the actual status of the cable without having to rely entirely on the predetermined plan. It can detect cable problems more promptly, reduce the possibility of missing maintenance opportunities, improve the reliability and safety of the power transmission system, and reduce the risk of power system interruptions and failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of the present invention; Figure 2 It is a flow chart for controlling the inspection robot to inspect the cables; Figure 3 It is a flow chart for constructing a cable status analysis model; Figure 4 This is the structural diagram of the inspection robot control system based on deep learning. DETAILED DESCRIPTION

[0018] In the description of this application, those skilled in the art should know that this application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, this application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, this application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.

[0019] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0020] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.

[0021] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.

[0022] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.

[0023] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0024] The present application is described below in conjunction with the drawings in the present application. Example

[0025] like Figures 1 to 3 As shown, the inspection robot control method based on deep learning of the present invention specifically includes the following steps: S1. Divide the cables to be inspected into units to obtain multiple cable units; When inspecting high-voltage overhead cables, the cable lines usually span multiple cable towers. In order to effectively monitor and manage the status of the cables, the cable segments need to be divided into units. The division of the units is critical because it will help the robot monitor different parts of the cables more finely, thereby more accurately determining the inspection frequency and timing. Step S1 specifically includes the following: S11. Cable line segmentation: The cable line needs to be divided into a series of cable units; this can be achieved by the distance between the cable towers, with each cable unit corresponding to a pair of adjacent cable towers; this division is based on geographical location, dividing the cable into specific sections; S12. Cable unit identification: Each cable unit needs to be uniquely identified, usually using a naming convention or numbering system so that the robot can easily identify and track each unit; S13. Location information recording: In order to better navigate the inspection robot, the location information of each cable unit (such as GPS coordinates or other geographic reference points) needs to be recorded so that the robot can accurately locate and move to each unit; S14. Cable unit information collection: On each cable unit, some basic information needs to be collected, such as the starting and ending points of the cable, and other environmental factors related to the unit, such as vegetation conditions, weather conditions, etc.; this helps the robot better understand the inspection environment.

[0026] In addition, considering the dangerous nature of high-voltage cables, the safety of the inspection robot must be ensured; therefore, appropriate safety measures, such as safety areas, are set on each cable unit to protect the robot operator and surrounding personnel; in summary, the goal of step S1 is to divide the cable line into easily manageable units, each with a clearly defined location and related information; this division method will help subsequent steps, such as data collection and analysis, to better manage and maintain the high-voltage overhead cable system.

[0027] S2. According to the inspection plan time, control the inspection robot to inspect the cables, and use the inspection robot to obtain the status information of each cable unit; the status information includes the number of cable strand breaks in the cable unit, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; Step S2 is the core step of the deep learning inspection robot control method, which involves obtaining and processing the actual status information of the cable. According to the inspection plan time, the inspection robot is controlled to inspect the cable, and the status information of each cable unit is obtained by the inspection robot. Specifically, this includes the following steps: S21. Determine the inspection plan time: According to the pre-established inspection plan, determine the time for inspection of each cable unit; this time can be within a fixed interval or can be adjusted according to the actual situation of the cable; S22, positioning and navigation: After determining the inspection time, the inspection robot needs to be accurately positioned to the cable unit that needs to be inspected; this can be achieved by using GPS, laser radar or other navigation technologies; S23, collecting status information: after arriving at the designated location, the inspection robot uses the high-definition camera on board to capture images of the cable unit and collect other relevant information; such information includes but is not limited to the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance, etc.; S24, storage and transmission: The collected status information is stored in the memory of the inspection robot, and can be transmitted to the background server or data center in real time through the communication module for further processing and analysis.

[0028] How to collect the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance in step S23 includes the following: a. Detection of cable strand breakage: Cable strand breakage refers to a partial break in the insulation or conductor of a cable. To detect the number of cable strand breakages, a high-resolution camera or image sensor is usually used to obtain an image of the cable surface. Image processing techniques, such as computer vision and image analysis, are used to detect strand breakages in the image. This can be achieved by comparing the cable line in the image with the line in a normal state for difference analysis. An algorithm is then used to mark and count the number of strand breakages, thereby obtaining the number of cable strand breakages. b. Detection of the average temperature of the cable: The temperature of the cable is a key parameter that can indicate whether the cable has potential overload or other problems. Temperature sensors or thermal imagers are usually used to measure the surface temperature of the cable. Temperature sensors directly measure the temperature of the cable, while thermal imagers generate temperature images through infrared radiation. The average temperature of the cable can be obtained by statistically averaging the measured values, which will provide a more comprehensive understanding of the thermal status of the cable. c. Detection of the average gray value of the cable image: The average gray value of the cable image is used to detect surface unevenness, reflecting damage or corrosion; the average gray value refers to the average value of the brightness level of the pixels in the image; the average gray value of the cable image can be calculated through image processing algorithms; this usually includes converting the pixel values ​​of the image into gray values ​​and then calculating the average value of all pixels; d. Detection of maximum sag distance: sag is the maximum droop distance of the cable between two supporting points; its measurement requires the use of a sag measuring instrument; the sag measuring instrument can determine the sag distance of the cable through laser or other distance measurement technology; sag distance measurement is essential to understanding the tension of the cable and the status of the supporting structure, because excessive sag may indicate that there is a problem with the cable and it needs to be repaired.

[0029] In this step, by controlling the inspection robot to conduct inspections, automatic detection and intelligent analysis of cables can be achieved, greatly improving the efficiency and accuracy of inspections; this step can simultaneously collect multiple status information of the cable, including the number of cable breakages, average temperature, average grayscale value, and maximum sag distance, so as to have a more comprehensive understanding of the cable status; the inspection robot can collect and transmit data in real time, so that the cable status information can be obtained and analyzed in a timely manner, which is conducive to timely discovery and resolution of potential problems; the collected data can be stored in the inspection robot's memory, or it can be transmitted to the background server or data center in real time, facilitating subsequent data analysis and processing; in general, the advantage of step S2 is that it can mainly meet the needs of automation, intelligence, real-time monitoring, high security, and data storage and transmission of cables.

[0030] S3, arranging the status information of several cable units collected by the inspection robot in the order of connection, and aligning the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance, respectively, to obtain a cable status feature matrix; The goal of step S3 is to organize the cable status information collected from the inspection robot into a standardized feature matrix so that the subsequent cable status analysis model can process and analyze this data; the following is a detailed introduction to step S3: S31, data alignment and arrangement: the inspection robot collects the status information of the cable unit, and these data need to be arranged in the connection order and aligned; the purpose of alignment is to ensure that the data of different cable units can be directly compared; data alignment includes the following steps: S311, ensuring that the data of each cable unit has the same data items, such as the number of broken strands, temperature, gray value and sag distance; S312. Ensure that the units of the data are consistent, for example, the temperature can be expressed in degrees Celsius, and the sag distance can be expressed in meters; S313, arranging the data according to the connection sequence of the cable units for subsequent processing; S32. Generation of cable status characteristic matrix: After data alignment and arrangement are completed, these data can be organized into a cable status characteristic matrix; this matrix is ​​a table, in which each row represents a cable unit, and each column represents a status information item, such as the number of broken strands, temperature, gray value and sag distance; the elements in the matrix are the status information of the corresponding cable unit; S33. Use cable status feature matrix: The generated cable status feature matrix will be used for cable status analysis in subsequent steps; the purpose of this matrix is ​​to provide a simplified and standardized way to represent the status of different cable units for analysis and prediction by deep learning models.

[0031] More specifically, the cable status characteristic matrix is: Among them, B n Indicates the number of times the wire strands are broken in the nth cable unit; the T n represents the average temperature of the nth cable unit; the C n represents the average grayscale value of the image of the nth cable unit; n represents the maximum sag distance of the nth cable unit; the elements of each row in the cable state characteristic matrix represent different state quantified characteristic values ​​of the cable segment between two adjacent cable towers.

[0032] In this step, the status data of different cable units are ensured to have the same data items, such as the number of broken strands, temperature, gray value and sag distance, by aligning and arranging the data. This ensures the consistency of the data, making it directly comparable and analyzable. Since the status information of different cable units may use different units, such as Celsius and Fahrenheit, or meters and feet, the data in the cable status feature matrix is ​​easier to understand and compare by standardizing the data units. The cable status feature matrix integrates the status information of different cable units into a table, with each row representing a cable unit and each column representing a status information item; this integration makes the data easier to process and analyze, while reducing the input complexity of the deep learning model; the cable status feature matrix provides a simplified and standardized way to represent the status information of the cable unit; this helps reduce the complexity of the data and makes it easier for the deep learning model to understand and process this information; most importantly, the goal of the S3 step is to provide data for the deep learning model to enable it to perform cable status analysis and prediction; by standardizing and organizing the data, S3 provides clearer and more consistent data input for subsequent analysis models, which helps improve the performance and accuracy of the model; In general, step S3 helps to improve the efficiency and reliability of cable status analysis and makes the data collected from the inspection robot easier to process and apply to deep learning models; these advantages help to improve the maintainability and safety of the cable system and reduce the risk of cable failure.

[0033] S4. Obtaining the voltage level of the cable, and according to the voltage level of the cable, retrieving a cable state analysis model corresponding to the voltage level; Step S4 involves obtaining the voltage level of the cable and calling the corresponding cable status analysis model according to the voltage level of the cable; the specific execution process includes the following steps: S41. Obtain the voltage level of the cable: The voltage level of the cable refers to the voltage level used for power transmission, usually in kilovolts (kV). Cables of different voltage levels have different technical specifications, standards and working requirements. This information can usually be found in the documents of the power system, or can be obtained through the system database or identification. S42, determining a cable status analysis model: In this step, according to the voltage level of the cable, it is necessary to call a previously established cable status analysis model; these models are deep learning or machine learning models that have been trained to analyze the status of the cable; cables of different voltage levels generally require different models because they may be affected by different environments, current loads, and thermal characteristics; S43. Retrieve the corresponding cable status analysis model: In this step, the system will automatically select or call the corresponding cable status analysis model according to the voltage level of the cable; this can be achieved through a model selection algorithm or a model library; the model selection should be real-time, because the cable status may change over time, and it is necessary to ensure that the selected model matches the current status and characteristics of the cable.

[0034] Among them, the cable status analysis model is built based on deep learning technology. The specific construction process is as follows: S421. Collect sample data: First, a large amount of high-voltage overhead cable inspection sample data needs to be collected, including the status information of each cable unit, such as the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance. These data are recorded in the form of time series to understand the evolution of the cable status over time. S422, data preprocessing: preprocessing the collected sample data, including data cleaning, standardization, normalization and other operations, to eliminate noise and outliers in the data, and convert the data into a format suitable for deep learning model processing, such as the cable state feature matrix format in step S3; S423, Feature Engineering: Feature engineering is required for cable status information; based on the evaluation of domain experts, the cable status information is associated with the inspection time; this can be achieved by analyzing historical data and expert experience, such as analyzing the correlation between the number of cable strand breaks and the inspection time, so as to determine that an inspection is required when the number of cable strand breaks reaches a certain threshold; S424. Build model architecture: Based on deep learning technology, build a model architecture suitable for processing cable status analysis problems; you can choose common deep learning models such as convolutional neural network, recurrent neural network or long short-term memory network; S425, model training and optimization: Use the feature-engineered data set to train the selected deep learning model; during the training process, the model will learn the relationship between the cable status and the actual inspection time; the training loss function is usually the prediction error of the inspection time; during the training process, it is necessary to continuously evaluate the performance and effect of the model, and optimize the model by adjusting the model's hyperparameters, increasing or decreasing the number of layers, changing the activation function, etc.; S426, deployment and testing: After training is completed, the model is deployed to the actual inspection system for testing to verify the model's prediction and generalization capabilities; S427. Iterative Update: The cable status analysis model needs to be updated regularly to adapt to the evolution of the cable status and changes in the model performance; this can be achieved by regularly retraining the model and using new data.

[0035] It should be noted that when constructing a cable status analysis model, the characteristics and inspection requirements of high-voltage overhead cables should be fully considered, and appropriate deep learning algorithms and model structures should be selected to improve the prediction accuracy and generalization ability of the model; at the same time, for different voltage levels and different types of cables, the model also needs to be adjusted and optimized according to actual conditions; in short, the construction of a cable status analysis model involves multiple steps such as data collection, preprocessing, feature engineering, model selection, training, evaluation, hyperparameter tuning, deployment and regular updates; the goal of this model is to predict the state evolution of cables through deep learning technology to improve the efficiency of inspection robots and the reliability of power systems.

[0036] S5, inputting the cable status feature matrix into the cable status analysis model corresponding to the voltage level, and the cable status analysis model outputs the time node for the next cable inspection; Step S5 is a key step to use the deep learning model to analyze the cable status and predict the next inspection time node; the specific process is as follows: S51, obtaining a cable status feature matrix: In step S3, the status information of several cable units collected by the inspection robot is arranged in the order of connection, and the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance are aligned respectively to form a cable status feature matrix; this matrix contains important information about the cable status and can be used as an input for a deep learning model; S52, retrieve the corresponding cable status analysis model: In step S4, the cable status analysis model corresponding to the voltage level is retrieved according to the voltage level of the cable; this model is a pre-trained deep learning model, which has the ability to analyze and predict the cable status; S53, inputting the cable state characteristic matrix into the model: inputting the cable state characteristic matrix obtained in step S3 into the cable state analysis model retrieved in step S4; S54. Model predicts the next inspection time node: The cable status analysis model makes inferences based on the input cable status feature matrix and outputs the time node for the next inspection; this time node is predicted based on the current status of the cable and historical data, and can serve as an important reference for the inspection plan.

[0037] In this step, by using the deep learning model, the cable status can be monitored in real time, and the next inspection time node can be predicted based on the model, which provides an important basis for formulating a reasonable inspection plan; the inspection robot control method based on deep learning can automatically formulate the next inspection time node according to the actual inspection results of the cable, avoiding the tediousness and inaccuracy of manual planning; at the same time, through the training and optimization of the model, the accuracy and precision of the prediction can be improved, making the inspection plan more reasonable and efficient; the inspection robot control method based on deep learning is scalable and flexible; with the increase of data sets and the development of technology, the model can be continuously optimized and improved to improve the accuracy and precision of the prediction; at the same time, this method can also be applied to different types of cables and inspection robots, and has broad application prospects; in short, step S5 is a key step of using a deep learning model to analyze the cable status and predict the next inspection time node, realizing real-time monitoring and prediction of the cable status, improving inspection efficiency and accuracy, reducing costs and risks, and having scalability and flexibility.

[0038] S6. Compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the cable in the inspection plan: if the inspection time node output by the cable status analysis model is earlier than the time node of the next inspection of the cable in the inspection plan, the next inspection time will be the inspection time node output by the cable status analysis model, and the inspection plan will be updated accordingly; if the inspection time node output by the cable status analysis model is later than the time node of the next inspection of the cable in the inspection plan, the next inspection time will still be based on the time node in the inspection plan.

[0039] The purpose of step S6 is to compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the corresponding cable in the inspection plan to determine the actual time of the next inspection. Specifically, step S6 includes the following contents: S61, obtaining the inspection time node output by the cable status analysis model: In step S5, the cable status feature matrix is ​​input into the cable status analysis model corresponding to the voltage level, and the model outputs the predicted inspection time node; S62, obtaining the inspection time node in the inspection plan: in the inspection plan, for each cable and voltage level, there is a predetermined inspection time node; S63, comparing the two time nodes: comparing the inspection time node output by the cable status analysis model with the time node in the inspection plan; specifically, the comparison result is processed in the following two cases: A. The inspection time output by the cable status analysis model is earlier than the time in the inspection plan: If the output of the cable status analysis model indicates that the next inspection is required in a shorter time, for example, because the cable status deteriorates rapidly or there is an emergency, then this new time will be adopted to ensure the safety and reliability of the cable; in this case, the inspection plan should be updated to set the next cable inspection time to the inspection time output by the cable status analysis model; B. The inspection time output by the cable status analysis model is later than the time in the inspection plan: If the cable status analysis model believes that the next inspection is required in a longer period of time, the original time in the inspection plan is still valid; in this case, the inspection plan should be maintained and the original time in the inspection plan should be maintained without making any changes.

[0040] In this step, by comparing the output of the cable status analysis model with the inspection plan, the actual time of the next inspection can be updated in time; this enables the inspection plan to be dynamically adjusted according to the latest cable status, improving the timeliness and flexibility of implementing the inspection; incorporating the prediction results of the cable status analysis model into the decision-making process can more accurately reflect the actual status of the cable; this can avoid the situation where the inspection plan based on fixed time intervals is too early or too late; When the inspection time output by the cable status analysis model is earlier than the time in the inspection plan, possible problems with the cable can be discovered in time, thereby reducing potential safety risks and improving the reliability of the cable system. When the cable status deteriorates rapidly or an emergency occurs, necessary measures can be taken quickly to ensure the normal operation of the cable system. When the inspection time output by the cable status analysis model is later than the time in the inspection plan, maintaining the original inspection plan can avoid excessive intervention and maintain the stability of the system. In general, step S6 makes the inspection strategy more targeted and real-time by combining the output of the cable status analysis model with the predetermined inspection plan, thereby improving the operating efficiency and safety of the cable system. Example

[0041] like Figure 4 As shown, the inspection robot control system based on deep learning of the present invention specifically includes the following modules: A division processing module is used to divide the cables that need to be inspected into units to obtain multiple cable units; The inspection control module is used to control the inspection robot to inspect the cables according to the inspection plan time, and use the inspection robot to obtain the status information of each cable unit; the status information includes the number of cable strand breaks in the cable unit, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; The data processing module is used to arrange the status information of several cable units collected by the inspection robot in the order of connection, and align the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance, respectively, to obtain a cable status feature matrix; The model calling module is used to obtain the voltage level of the cable and, according to the voltage level of the cable, call the cable state analysis model corresponding to the voltage level; A state analysis module is used to input the cable state characteristic matrix into a cable state analysis model corresponding to the voltage level, and the cable state analysis model outputs the time node for the next cable inspection; The inspection plan update module is used to compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the cable in the inspection plan: If the inspection time node output by the cable status analysis model is earlier than the time node of the next inspection of the cable in the inspection plan, the next inspection time will be the inspection time node output by the cable status analysis model, and the inspection plan will be updated accordingly; If the inspection time node output by the cable status analysis model is later than the time node of the next inspection of the cable in the inspection plan, the next inspection time will still be carried out according to the time node in the inspection plan.

[0042] In this embodiment, the system can autonomously control the actions of the inspection robot and adjust the inspection plan according to the actual status of the cable, rather than relying solely on a pre-made plan; this improves the automation level of the system, reduces human intervention, and thus improves efficiency; through the status information collected by the inspection robot, the system can understand the status of each cable unit in real time, including the number of cable strand breaks, temperature, image features, etc., so that cable problems can be detected more quickly and the possibility of missed faults can be reduced; The data processing module can integrate and align the status information of multiple cable units collected to construct a cable status feature matrix; this helps to comprehensively analyze the overall status of the cable system rather than a single cable unit, thereby better understanding the health of the system; the system can select the corresponding analysis model according to the voltage level of different cables through the model calling module and the status analysis module, so it is suitable for high-voltage overhead cables of various voltage levels; The system can output the next inspection time node according to the actual status of the cable, making the inspection plan more flexible and optimized; this helps to detect problems in a timely manner, reduce maintenance costs, and improve the reliability and safety of the power system; by reducing human risks and providing real-time status information, the system can improve the safety of the inspection process and reduce the risk of personnel needing to come into contact with high-voltage cables.

[0043] The various variations and specific embodiments of the deep learning-based inspection robot control method in the aforementioned embodiment 1 are also applicable to the deep learning-based inspection robot control system in this embodiment. Through the aforementioned detailed description of the deep learning-based inspection robot control method, those skilled in the art can clearly understand the implementation method of the deep learning-based inspection robot control system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0044] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0045] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A control method for an inspection robot based on deep learning, characterized in that: The method comprises: The cables to be inspected are divided into units to obtain multiple cable units; According to the inspection plan time, the inspection robot is controlled to inspect the cables, and the status information of each cable unit is obtained by the inspection robot; the status information includes the number of cable strand breaks in the cable unit, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; The status information of several cable units collected by the inspection robot is arranged in the order of connection, and the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image and the maximum sag distance are aligned to obtain the cable status feature matrix; Obtaining the voltage level of the cable, and according to the voltage level of the cable, calling up a cable state analysis model corresponding to the voltage level; The cable status feature matrix is ​​input into the cable status analysis model corresponding to the voltage level, and the cable status analysis model outputs the time node for the next cable inspection; Compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the cable in the inspection plan: If the inspection time node output by the cable status analysis model is earlier than the time node of the next inspection of the cable in the inspection plan, the next inspection time will be the inspection time node output by the cable status analysis model, and the inspection plan will be updated accordingly; If the inspection time node output by the cable status analysis model is later than the time node of the next inspection of the cable in the inspection plan, the next inspection time will still be carried out according to the time node in the inspection plan.

2. The inspection robot control method based on deep learning according to claim 1, characterized in that: The cable status characteristic matrix is: Among them, B n Indicates the number of times the wire strands are broken in the nth cable unit; the T n represents the average temperature of the nth cable unit; n represents the average grayscale value of the image of the nth cable unit; n represents the maximum sag distance of the nth cable unit; the elements of each row in the cable state characteristic matrix represent different state quantified characteristic values ​​of the cable segment between two adjacent cable towers.

3. The inspection robot control method based on deep learning according to claim 1, characterized in that: The method for dividing cables that need to be inspected into units includes: Divide the cables between two adjacent cable towers into one cable unit; Each cable unit is marked with a unique identifier for inspection robots to identify and track; The geographical location information of each cable unit is associated with the corresponding cable unit and saved for navigation and positioning by the inspection robot.

4. The inspection robot control method based on deep learning as claimed in claim 3, characterized in that: The method for controlling an inspection robot to inspect a cable comprises: According to the pre-established inspection plan, determine the time to inspect each cable unit; According to the geographical location information of the cable unit, the inspection robot is positioned to the cable unit that needs to be inspected; Control the inspection robot to collect status information of the cable unit.

5. The inspection robot control method based on deep learning as claimed in claim 4, characterized in that: The method for controlling the inspection robot to collect status information of the cable unit comprises: Use image sensors to obtain images of the cable unit surface, use image processing technology to compare the cable line in the image with the line in normal state to perform difference analysis to detect broken strands in the image, and use algorithms to mark and count the number of broken strands to obtain the number of broken strands in the cable unit; Use a temperature sensor or a thermal imager to measure the surface temperature of multiple locations on the cable unit, and obtain the average temperature of the cable unit by statistically averaging the measured values; An image sensor is used to obtain a cable unit surface image, pixel values ​​of the cable unit surface image are converted into gray values, and an average gray value of all pixel gray values ​​is calculated to obtain an average gray value of the cable unit image; Use laser ranging technology to determine the lowest point of the cable unit, and calculate the continuous vertical distances between the lowest point of the cable unit and the two end points of the cable unit to obtain the maximum sag distance of the cable unit.

6. The inspection robot control method based on deep learning as claimed in claim 2, characterized in that: The method for constructing the cable status analysis model comprises: Collect high-voltage overhead cable inspection sample data, including cable status information under different states, wherein the cable status information includes the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; Preprocess the collected sample data, including data cleaning, standardization, and normalization; and convert the sample data into the cable status feature matrix format; Based on the evaluation of domain experts, the cable status information is associated with the inspection time, and it is determined that inspection is required when each cable status information item reaches the set threshold; Based on deep learning technology, a model architecture is built to handle cable status analysis problems; Use the preprocessed sample data to train and optimize the selected deep learning model; The cable status analysis model is deployed in an actual inspection system for testing to verify the model's predictive and generalization capabilities.

7. The inspection robot control method based on deep learning according to claim 6, characterized in that: The model architecture of the cable status analysis model selects a convolutional neural network, a recurrent neural network or a long short-term memory network.

8. A patrol robot control system based on deep learning, characterized in that: The system comprises: A division processing module is used to divide the cables that need to be inspected into units to obtain multiple cable units; The inspection control module is used to control the inspection robot to inspect the cables according to the inspection plan time, and use the inspection robot to obtain the status information of each cable unit; the status information includes the number of cable strand breaks in the cable unit, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance; The data processing module is used to arrange the status information of several cable units collected by the inspection robot in the order of connection, and align the number of cable strand breaks, the average temperature of the cable, the average gray value of the cable image, and the maximum sag distance, respectively, to obtain a cable status feature matrix; The model calling module is used to obtain the voltage level of the cable and, according to the voltage level of the cable, call the cable state analysis model corresponding to the voltage level; A state analysis module is used to input the cable state characteristic matrix into a cable state analysis model corresponding to the voltage level, and the cable state analysis model outputs the time node for the next cable inspection; The inspection plan update module is used to compare the inspection time node output by the cable status analysis model with the time node of the next inspection of the cable in the inspection plan: If the inspection time node output by the cable status analysis model is earlier than the time node of the next inspection of the cable in the inspection plan, the next inspection time will be the inspection time node output by the cable status analysis model, and the inspection plan will be updated accordingly; If the inspection time node output by the cable status analysis model is later than the time node of the next inspection of the cable in the inspection plan, the next inspection time will still be carried out according to the time node in the inspection plan.

9. A patrol robot control electronic device based on deep learning, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

Images