Control method for ultrasonic drive devices applied to power facilities

Abstract

This invention relates to the field of power facility protection technology and discloses a control method for ultrasonic pest control equipment applied to power facilities. The method includes: real-time acquisition of environmental parameter data, construction of a multi-dimensional pest feature identification system and output of standardized signals; establishment of an ultrasonic parameter database based on frequency characteristic conversion technology, and generation of control commands through feature matching; construction of a historical pest control case information database; establishment of a multi-factor weight allocation model, calculation of priority scores and generation of task sequences; solving for the optimal control scheme using an improved genetic optimization algorithm, with the objectives of maximizing the pest control range, minimizing energy consumption, and shortening response time; and outputting control results and tracking feedback through an interactive platform. This method achieves accurate identification of pest activity, dynamic parameter matching, and multi-objective optimized control, improving the efficiency and adaptability of pest control for power facilities.

Description

Technical Field

[0001] This invention relates to the field of power facility protection technology, specifically to a control method for ultrasonic destructive devices applied to power facilities. Background Technology

[0002] During the operation of power systems, damage to power facilities by various pests (such as birds, rodents, and insects) has always been a significant issue affecting the safe and stable operation of power systems. The activities of these pests can lead to serious consequences such as power line short circuits, equipment failures, and line tripping, causing not only huge economic losses but also potentially triggering large-scale power outages and disrupting normal social production and daily life.

[0003] Currently, there are various methods for mitigating damage to power facilities, including physical mitigation, chemical control, and ultrasonic mitigation. Among these, ultrasonic mitigation technology is gradually becoming an important means of protecting power facilities due to its non-contact, pollution-free, and minimal impact on humans and the environment. However, existing ultrasonic mitigation equipment control methods have many shortcomings.

[0004] In pest identification, traditional methods often rely on a single sensor to collect data, making it difficult to comprehensively and accurately capture pest activity characteristics. Furthermore, they are easily affected by environmental background clutter, resulting in low pest identification accuracy. At the same time, the lack of a comprehensive multi-dimensional pest feature identification system makes it impossible to effectively distinguish between different types and activity states of pests, affecting the targeted nature of subsequent pest control measures.

[0005] In selecting ultrasonic parameters, existing technologies mostly rely on experience to set fixed parameters without establishing a systematic ultrasonic parameter database. This makes it difficult to dynamically adjust parameters according to specific pest types, power facility types, and environmental conditions, resulting in unstable pest control effects. Furthermore, when faced with multiple pest control tasks, the lack of a scientific prioritization mechanism prevents the rational allocation of resources based on factors such as facility importance and pest hazard level, potentially leading to critical facilities not receiving timely protection.

[0006] Existing control methods often only consider a single objective (such as maximizing pest control coverage) when optimizing equipment control schemes, while ignoring factors such as equipment energy consumption and response time, resulting in poor overall control performance. At the same time, the lack of effective utilization of historical pest control cases makes it impossible to learn from past experience and optimize control strategies, making it difficult to continuously improve the operating efficiency of pest control equipment.

[0007] With the continuous development and increasing intelligence of power systems, higher demands are being placed on the accuracy, adaptability, and efficiency of ultrasonic pest control methods. Therefore, developing an ultrasonic pest control method that can comprehensively consider multiple factors and achieve intelligent control has become an urgent problem to be solved in the field of power facility protection. Summary of the Invention

[0008] The purpose of this invention is to provide a control method for ultrasonic destructive devices applied to power facilities, in order to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides a control method for ultrasonic destructive devices applied to power facilities, the method comprising:

[0010] Real-time collection of environmental parameter data around power facilities; construction of a multi-dimensional pest characteristic identification system; and output of standardized pest activity signals through a signal processing module.

[0011] An ultrasonic parameter database is established based on frequency characteristic conversion technology. When a pest control requirement is detected, the target parameters are located and equipment control commands are generated through feature matching algorithm.

[0012] Establish a database of historical power plant damage cases and dynamically record usage records of different types of power facilities, environmental conditions, and ultrasonic parameters.

[0013] A multi-factor weight allocation model is established. By combining the importance of facilities, the level of hazard of hazardous substances, the degree of environmental interference and the urgency of tasks, the Delphi method is used to assign weights to each indicator. A weighted summation method is used to construct a control model to calculate the control priority score, thereby generating an equipment control priority sequence.

[0014] An improved genetic optimization algorithm is used to solve for the optimal equipment control scheme, with optimization objectives set in the direction of maximizing pest control coverage, minimizing equipment energy consumption, and shortening response time;

[0015] The interactive control platform outputs the final equipment control results and continuously tracks feedback on the actual pest control effect.

[0016] Preferably, the real-time acquisition of environmental parameter data around power facilities, the construction of a multi-dimensional pest characteristic identification system, and the output of standardized pest activity signals through a signal processing module are specifically as follows:

[0017] Based on array-type ultrasonic sensors, the activity signal data of harmful objects around power facilities are obtained. The signal acquisition terminal receives the raw signal from the sensor and performs noise suppression, time domain analysis, frequency domain conversion preprocessing to remove environmental background clutter.

[0018] Based on the biological characteristics of pest activity and the ultrasonic reflection pattern, correlation analysis was used to screen parameters closely related to pest characteristics as candidate features.

[0019] Candidate features are divided into frequency features, amplitude features, and time-frequency features, and a multi-level feature system is constructed. The overall pest identification rate is used as the top-level feature, and the top-level feature is decomposed into several first-level features. Each first-level feature is further subdivided into several second-level features.

[0020] The signal processing module is trained using a wavelet transform model based on a multi-level feature system.

[0021] The location information of power facilities is associated with the harmful characteristics of the pre-processed signals to establish a location-feature mapping relationship;

[0022] Signal analysis software was selected as the pest control platform, and the area was marked on the interface according to the actual layout of the power facilities. Different labels were used for visualization based on the differences in the characteristics of the pests. Areas with large differences in characteristics were marked with prominent labels, while areas with small differences in characteristics were marked with conventional labels.

[0023] Preferably, the establishment of the ultrasonic parameter database based on frequency characteristic conversion technology specifically includes:

[0024] Collect sensitive frequency data for common types of hazards and types of power facilities, including standard parameters such as fundamental frequency values, frequency range values, and intensity thresholds;

[0025] Obtain the physical attribute information of the parameters, including device model, emission angle, and effective distance parameters;

[0026] Parameters are abstracted into data nodes, and the frequency similarity between parameters is abstracted into data associations, thus constructing a node-association data structure;

[0027] Establish the database architecture, including defining data tables, setting fields, and establishing data relationships, storing parameter information, attribute information, and associations;

[0028] The preprocessed data is imported into the database of the signal analysis software to establish an ultrasonic parameter database.

[0029] Preferably, the step of locating the target parameters and generating device control commands through a feature matching algorithm specifically includes:

[0030] Based on the ultrasonic parameter database, each parameter node is regarded as a sample point in the dataset, and the frequency correlation between parameters is regarded as the similarity between samples.

[0031] Using the characteristics of pests in the area to be repelled as query points, the feature distance to each sample point is calculated using a fuzzy matching algorithm. When the sample point with the highest matching degree is reached, the parameter information of that sample point is recorded.

[0032] The characteristic distances obtained from the analysis and calculation are combined with information on power facility type and environmental interference to determine the matching range of the target parameters.

[0033] Extract the basic frequency value, frequency range value, and intensity threshold parameter of the target parameter from the ultrasonic parameter database to generate equipment control parameters;

[0034] The target parameters, control parameters, facility type, and environmental conditions are combined into a label and then visualized on the signal analysis software interface.

[0035] Preferably, the construction of a historical hazard case information database, dynamically recording usage records of different power facility types, environmental conditions, and ultrasonic parameters, specifically includes:

[0036] The main dimensions for constructing the information database include different types of power facilities, environmental conditions, and usage records of ultrasonic parameters.

[0037] At the same time, specific information fields are planned for each dimension, including facility model, ambient temperature and humidity, frequency of use, actual pest control effect, and user feedback and evaluation.

[0038] By connecting the power monitoring and management system with the user feedback platform, we continuously update the usage parameters, effect feedback, and equipment operation information of historical pest control cases.

[0039] A database of historical hazard cases is constructed based on key dimensions and data fields.

[0040] Preferably, the establishment of the multi-factor weight allocation model involves combining the importance of the facility, the hazard level of the hazardous substances, the degree of environmental disturbance, and the urgency of the task. The Delphi method is used to assign weights to each indicator, and a weighted summation method is employed to construct a control model to calculate the control priority score, thereby generating an equipment control priority sequence. Specifically:

[0041] Based on the importance of the facilities and the hazard level of the pollutants, the control importance level is divided into different levels. The requirements of critical facilities or high-hazard pollutants are set as high importance level, while the requirements of non-critical facilities or low-hazard pollutants are set as low importance level.

[0042] The degree of environmental interference is divided into strong interference environment, medium interference environment, and weak interference environment.

[0043] The urgency factors of tasks are categorized into urgent tasks, routine tasks, and delayed tasks;

[0044] Based on the Delphi method, weights are assigned to the control importance level, the degree of environmental disturbance, and the urgency of the task.

[0045] A weighted summation method is used to construct the control model. Each index value is multiplied by its corresponding weight and then summed to obtain the control priority score for each device control task.

[0046] Based on the control priority scores calculated by the control model, all equipment control tasks are sorted, with tasks with higher scores having higher control priority, thus generating an equipment control priority sequence.

[0047] Preferably, the step of using an improved genetic optimization algorithm to solve for the optimal equipment control scheme sets the optimization objective in the direction of maximizing the pest control coverage, minimizing equipment energy consumption, and shortening response time, specifically as follows:

[0048] A set of candidate solutions is randomly generated as the initial population;

[0049] Calculate the fitness value for each candidate solution based on the optimization objective;

[0050] The optimal solution is selected as the parent individual based on the fitness value, and the crossover and mutation probabilities of other individuals are updated.

[0051] By simulating the gene recombination and mutation behavior of the population, the candidate solutions are continuously updated iteratively until the termination condition is met;

[0052] Select a set of optimal solutions from the final population as the optimal equipment control scheme.

[0053] Preferably, the control priority score is specifically:

[0054] The control priority score is jointly determined by the control importance level weight, environmental disturbance degree weight, task urgency weight, and equipment status weight. The control importance level weight corresponds to the degree of influence of the facility's importance and the level of hazard level; the environmental disturbance degree weight corresponds to the degree of influence of environmental conditions; the task urgency weight corresponds to the degree of urgency of the hazard removal requirement; and the equipment status weight corresponds to the degree of influence of the equipment's current operating status. The final control priority score is obtained by multiplying each weight by its corresponding level score and then summing them up.

[0055] Preferably, the optimization objective is specifically:

[0056] The optimization objectives include maximizing the pest control coverage, minimizing equipment energy consumption, and shortening response time. The pest control coverage objective is achieved by calculating the overlap between candidate solutions and the pest activity area. The equipment energy consumption objective is achieved by limiting the equipment power and operating time range of candidate solutions. The response time objective is achieved by reducing the number of algorithm iterations and instruction transmission time. The final optimization objective is the comprehensive optimization of the three sub-objectives.

[0057] Preferably, the step of continuously updating the usage parameters, effect feedback, and equipment operation information of historical pest control cases through data integration with the power monitoring and management system and user feedback platform specifically involves:

[0058] Establish a data interface protocol and define the data transmission format between the power monitoring and management system and the user feedback platform, including structured usage parameter fields and semi-structured feedback evaluation fields;

[0059] Set a scheduled task to synchronize data daily from 11:00 to 13:00, and obtain data on newly added pest control cases in the previous 24 hours;

[0060] The validity of the synchronized data is verified, and invalid data that is missing facility models or does not record the actual pest control effect is removed;

[0061] The verified data is categorized and stored according to facility type and environmental conditions, and the usage records in the historical hazard case information database are updated.

[0062] Compared with the prior art, the beneficial effects of the present invention are:

[0063] By collecting real-time environmental parameter data around power facilities, a multi-dimensional pest feature identification system is constructed. Combined with array-type ultrasonic sensors to acquire pest activity signals, and preprocessing such as noise suppression and time-domain analysis, interference from environmental background clutter is effectively removed, improving the accuracy of pest activity signals. Simultaneously, candidate features are categorized into frequency features, amplitude features, and time-frequency features, constructing a multi-level feature system. Wavelet transform models are used to train the signal processing module, enhancing the ability to identify different pest features and making pest identification more accurate, providing a reliable basis for subsequent pest control.

[0064] An ultrasonic parameter database is established based on frequency characteristic conversion technology, collecting information such as sensitive frequency data of common pests and power facilities. When a pest control requirement is detected, the target parameters can be quickly located through a feature matching algorithm, generating appropriate equipment control commands. This overcomes the drawback of traditional methods that rely on experience to set fixed parameters, enabling dynamic adjustment of ultrasonic parameters, improving the adaptability of pest control equipment to different scenarios, and ensuring the stability of the pest control effect.

[0065] A historical pest control case database was built to dynamically record usage data under different conditions. This database is continuously updated through integration with the power monitoring and management system and user feedback platform, enabling control methods to learn from historical data and optimize control strategies. Simultaneously, a multi-factor weighting allocation model was established. Combining factors such as facility importance and pest hazard level, the Delphi method was used to assign weights and calculate control priority scores, generating an equipment control priority sequence. This ensures that when multiple pest control tasks exist simultaneously, important and urgent tasks are prioritized, resources are allocated rationally, and overall pest control efficiency is improved.

[0066] An improved genetic optimization algorithm was employed to find the optimal equipment control scheme. With the optimization objectives of maximizing pest control coverage, minimizing equipment energy consumption, and shortening response time, the overall optimality of the equipment control scheme was achieved through multi-objective optimization. This not only improved the coverage capability of the pest control equipment, ensuring effective protection for more areas, but also reduced equipment energy consumption and energy waste, while shortening response time to ensure timely removal of pests, further enhancing the safety of power facilities.

[0067] By outputting control results through an interactive control platform and continuously tracking feedback, a closed-loop control process is formed, facilitating timely problem detection and adjustment of control strategies to continuously improve pest control effectiveness. In summary, this control method significantly improves pest identification accuracy, parameter adaptability, resource allocation rationality, and control scheme optimization, providing more reliable and efficient protection for power facilities and ensuring the safe and stable operation of the power system. Attached Figure Description

[0068] Figure 1 This is a flowchart of the ultrasonic damage control method for power facilities as described in this invention;

[0069] Figure 2 A flowchart for harmful substance feature identification and signal processing;

[0070] Figure 3 A flowchart for building an ultrasonic parameter database;

[0071] Figure 4 Flowchart for feature matching and device control command generation;

[0072] Figure 5 A flowchart for building and updating a database of historical homicide cases. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] Please see Figures 1-5 This invention provides a control method for ultrasonic destructive devices applied to power facilities, the specific implementation steps of which are as follows:

[0075] Real-time collection of environmental parameter data around power facilities is used to construct a multi-dimensional pest characteristic identification system, and standardized pest activity signals are output through a signal processing module.

[0076] An ultrasonic parameter database is established based on frequency characteristic conversion technology. When a pest control requirement is detected, the target parameters are located and equipment control commands are generated through feature matching algorithm.

[0077] Establish a database of historical power plant damage cases to dynamically record usage records of different power facility types, environmental conditions, and ultrasonic parameters.

[0078] A multi-factor weight allocation model is established. By combining the importance of facilities, the level of hazard of hazardous substances, the degree of environmental interference and the urgency of the task, the Delphi method is used to assign weights to each indicator. A weighted summation method is used to construct a control model to calculate the control priority score, thereby generating an equipment control priority sequence.

[0079] An improved genetic optimization algorithm is used to solve for the optimal equipment control scheme, with optimization objectives set in the direction of maximizing pest control coverage, minimizing equipment energy consumption, and shortening response time.

[0080] The interactive control platform outputs the final equipment control results and continuously tracks feedback on the actual pest control effect.

[0081] Example 1:

[0082] In this embodiment, while real-time acquisition of environmental parameter data around power facilities is conducted to construct a multi-dimensional pest characteristic identification system, and standardized pest activity signals are output through a signal processing module, an array of ultrasonic sensors is used to acquire pest activity signal data around the power facilities. These array of ultrasonic sensors are deployed around the power facilities, enabling them to acquire signals from multiple angles and positions in the surrounding environment, ensuring the comprehensiveness and accuracy of the acquired signals. A signal acquisition terminal is connected to these sensors to receive the raw signals transmitted from them.

[0083] After the raw signal is transmitted to the signal acquisition terminal, it needs to undergo a series of preprocessing operations. Preprocessing includes noise suppression, time-domain analysis, and frequency-domain transformation. Noise suppression aims to remove environmental background clutter, as various interference signals exist in the surrounding environment of actual power facilities, such as operating noise from other equipment and sounds from the natural environment. These noises can interfere with the identification of harmful activity signals, so they must be suppressed using appropriate algorithms and techniques. Time-domain analysis helps to understand the characteristics of the signal in the time dimension, such as how the signal amplitude changes over time. Frequency-domain transformation converts the time-domain signal to the frequency domain for better analysis of the signal's frequency components.

[0084] After pretreatment, it is necessary to screen parameters closely related to pest characteristics as candidate features based on correlation analysis, according to the biological characteristics of pest activity and the ultrasonic reflection patterns. Different pests have different biological characteristics, and the ultrasonic signals they generate during activity will also have different characteristics. Furthermore, the reflection patterns of ultrasonic waves when encountering pests are also related to the properties of the pests. Through correlation analysis, parameters with a high degree of correlation with pest characteristics can be identified; these parameters can more accurately reflect the presence and activity of pests.

[0085] Candidate features are categorized into frequency features, amplitude features, and time-frequency features, and a multi-level feature system is constructed. In this system, the overall pest identification rate is used as the top-level feature, which is then decomposed into several first-level features, and each first-level feature is further subdivided into several second-level features. This hierarchical structure provides a more systematic and comprehensive description of pest characteristics, resulting in more accurate and detailed pest identification.

[0086] Based on the established multi-level feature system, a wavelet transform model is used to train the signal processing module. The wavelet transform model possesses excellent time-frequency localization characteristics, enabling multi-resolution signal analysis and making it suitable for processing signals with complex time-frequency characteristics, such as pest activity signals. Through training, the signal processing module can better identify and process pest activity signals.

[0087] This method associates the location information of power facilities with the characteristics of pests in preprocessed signals to establish a location-feature mapping relationship. Power facilities have specific locations in the actual environment, and the activity of pests may be related to the location of power facilities. Establishing this mapping relationship can clarify the characteristics of pests at different locations, providing more accurate location information for subsequent pest control operations.

[0088] Signal analysis software was selected as the pest control platform, and the interface was marked with areas according to the actual layout of the power facilities. Different labels were used for visualization based on the differences in pest characteristics; areas with significant differences in characteristics were marked with prominent labels, while areas with minor differences were marked with standard labels. This visualization method allows operators to intuitively understand the characteristics of pests in different areas around the power facilities, facilitating appropriate decision-making and operations.

[0089] Example 2:

[0090] In this embodiment, when establishing an ultrasonic parameter database based on frequency characteristic conversion technology, it is necessary to collect sensitive frequency data for common pest types and power facility types. Common pest types encompass various organisms that may harm power facilities, such as birds and rodents. For different pests, standard parameters such as their corresponding base frequency values, frequency range values, and intensity thresholds are obtained. Simultaneously, for different types of power facilities, such as substations and transmission lines, it is also necessary to collect sensitive frequency data applicable during pest control. This data forms the foundation for establishing the database.

[0091] After collecting sensitive frequency data, it is also necessary to obtain the physical attribute information of the parameters, including equipment model, emission angle, and effective distance. Different equipment models may have different performance and parameters. The emission angle determines the coverage range of the ultrasonic waves, while the effective distance affects the range of the repellent effect. This physical attribute information is crucial for the accurate use of ultrasonic equipment.

[0092] The collected parameters are abstracted into data nodes, and the frequency relationships between parameters are abstracted into data associations, thus constructing a node-association data structure. Each parameter is an independent data node, and nodes are associated with each other through frequency relationships. This structure clearly reflects the connections between different parameters, facilitating subsequent queries and matching.

[0093] Establishing the database architecture includes defining data tables, setting fields, and establishing data relationships. When defining data tables, they need to be categorized according to the type and attributes of the parameters. For example, tables for pest types, power facility types, and ultrasonic parameters can be created. The fields in each data table must accurately reflect the specific information of the parameters. For instance, the pest type table should include fields such as pest name and category, while the ultrasonic parameter table should include fields such as base frequency value, frequency range value, intensity threshold, and equipment model. Simultaneously, the relationships between data must be clearly defined, such as the correspondence between pest types and ultrasonic parameters, to ensure the consistency and integrity of the data in the database.

[0094] After establishing the database architecture, the preprocessed data is imported into the signal analysis software's database, thus creating an ultrasonic parameter database. The preprocessing process includes cleaning, organizing, and validating the collected data, removing invalid and erroneous data to ensure the accuracy and reliability of the data imported into the database.

[0095] When a hazard removal requirement is detected, a feature matching algorithm is used to locate the target parameters and generate equipment control commands. First, based on the established ultrasonic parameter database, each parameter node is considered a sample point in the dataset, and the frequency correlation between parameters is considered the similarity between samples. In this way, the entire database constitutes a dataset containing multiple sample points and their similarity relationships.

[0096] Using the pest characteristics of the area to be treated as query points, a fuzzy matching algorithm is used to calculate the feature distances from the query point to each sample point. The fuzzy matching algorithm can handle feature matching problems with uncertainty and fuzziness, and is suitable for matching pest characteristics with ultrasonic parameters. During the process of traversing all sample points, when the sample point with the highest matching degree is found, the parameter information of that sample point is recorded; this parameter information constitutes the initially matched target parameters.

[0097] The calculated feature distances, combined with information on power facility type and environmental interference, determine the matching range of the target parameters. The magnitude of the feature distance reflects the similarity between the query point and the sample point; analyzing the feature distances helps determine the possible range of the target parameters. Furthermore, different power facility types and levels of environmental interference can influence the selection of ultrasonic parameters; therefore, these factors need to be comprehensively considered to further narrow down the matching range and ensure the accuracy of the target parameters.

[0098] The fundamental frequency value, frequency range value, intensity threshold, and other parameters of the target parameters are extracted from the ultrasonic parameter database to generate equipment control parameters. These equipment control parameters are the key basis for controlling the operation of ultrasonic equipment, determining the operating status of the ultrasonic equipment, such as the frequency and intensity of its emission.

[0099] The target parameters, control parameters, facility types, and environmental conditions are combined into a single label and visualized on the signal analysis software interface. This visualization allows operators to intuitively understand the equipment control parameters required for the current pest control task, as well as the corresponding facility types and environmental conditions, facilitating quick and accurate equipment control operations.

[0100] Example 3:

[0101] In this embodiment, when constructing a historical hazard case information database and dynamically recording usage records of different power facility types, environmental conditions, and ultrasonic parameters, it is necessary to obtain the main dimensions for constructing the database. These main dimensions include usage records of different power facility types, environmental conditions, and ultrasonic parameters. Power facility types cover various categories such as substations, transmission lines, and power distribution equipment; environmental conditions involve factors such as temperature, humidity, wind speed, and sunlight; and ultrasonic parameter usage records include the ultrasonic frequency, intensity, and emission time used.

[0102] Specific information fields need to be planned for each dimension, including facility model, ambient temperature and humidity, usage frequency, actual pest control effect, and user feedback. Facility model specifies the particular electrical equipment, as different models may differ in pest control requirements and effects. Ambient temperature and humidity are important environmental parameters that can affect pest activity and ultrasonic wave propagation. Usage frequency records the operating frequency of the ultrasonic equipment, which is directly related to the pest control effect. Actual pest control effect describes changes in pest activity after the control operation. User feedback collects subjective evaluations of the pest control effect from operators or relevant personnel.

[0103] By connecting the power monitoring and management system with the user feedback platform, the system continuously updates the usage parameters, effect feedback, and equipment operation information of historical hazard mitigation cases. This process first requires establishing a data interface protocol and defining the data transmission format between the power monitoring and management system and the user feedback platform. This includes structured usage parameter fields and semi-structured feedback evaluation fields. Structured usage parameter fields, such as facility model, ambient temperature and humidity, and usage frequency, have clear formats and specifications; semi-structured feedback evaluation fields are more flexible and used to record users' subjective evaluations.

[0104] A scheduled task is set up to synchronize data daily from 11:00 to 13:00, retrieving data on newly added pest control cases from the previous 24 hours. This scheduled task ensures the timeliness and regularity of data updates, enabling the timely inclusion of the latest pest control case data into the database.

[0105] After acquiring the synchronized data, it is necessary to verify its validity and remove invalid data that is missing facility models or does not record actual pest control effects. Validity verification is crucial to ensure the accuracy and completeness of the data in the database; data lacking critical information will affect subsequent analysis and application.

[0106] The verified data is categorized and stored according to facility type and environmental conditions, and the usage records in the historical pest control case database are updated. Categorized storage facilitates data management and retrieval; relevant pest control case data can be quickly located based on facility type and environmental conditions.

[0107] To more clearly represent the structure of data storage, the following formula is introduced:

[0108]

[0109] in, This represents a record in the historical persecution case database; This indicates the type of power facility, including substations, transmission lines, etc. This indicates environmental conditions, including parameters such as temperature and humidity; Indicates ultrasonic parameters, such as frequency and intensity; Indicates the actual effect of repelling pests; This indicates user feedback and ratings.

[0110] Through the above steps, a historical pest control case database is constructed based on key dimensions and data fields. This database dynamically records usage records for different types of power facilities, environmental conditions, and ultrasonic parameters, providing reference and basis for subsequent pest control work. In the process of building the database, each step is closely linked. From determining the dimensions and fields to data integration, synchronization, verification, and storage, all procedures are strictly followed to ensure the accuracy, completeness, and usability of the data in the database. By establishing such a historical pest control case database, past pest control experience can be summarized and analyzed, providing support for optimizing the control methods of ultrasonic pest control equipment and improving the efficiency and effectiveness of pest control work.

[0111] Example 4:

[0112] In this embodiment, when establishing a multi-factor weight allocation model and generating a device control priority sequence, it is necessary to first combine the importance of the facility, the level of hazard from harmful substances, the degree of environmental interference, and the urgency of the task, and then use the Delphi method to assign weights to each indicator. Finally, a control model is constructed by weighted summation to calculate the priority score. Taking a regional power grid as an example, this area contains different types of power facilities such as substations, high-voltage transmission lines, and ordinary power distribution equipment, and faces various threats from harmful substances such as birds nesting and rodents gnawing on cables.

[0113] When classifying control importance levels, the needs of critical facilities or highly hazardous substances are set at the high importance level based on the importance of the facilities and the hazard level of the pests. For example, as a core node in the regional power supply, the normal operation of a substation directly affects the power stability of the entire region. If the substation faces the threat of nesting by large birds such as eagles, which may cause short circuits in the equipment, the hazard level is high, and the pest control requirement should be set at the high importance level. On the other hand, ordinary power distribution equipment is relatively less important. If it faces minor disturbances from pests such as squirrels, which will not cause serious damage to the equipment in the short term, the pest control requirement should be set at the low importance level.

[0114] The degree of environmental interference is categorized into strong interference, medium interference, and weak interference environments. For example, in areas with power transmission lines near highways, noise and vibration generated by vehicles constitute a strong interference environment; in areas with power facilities near suburban farmland, environmental interference is relatively low, classifying it as a medium interference environment; while in isolated power facility areas in remote mountainous regions, human activity is minimal, resulting in a low degree of environmental interference, which is considered a weak interference environment.

[0115] The urgency of a task is categorized into emergency tasks, routine tasks, and delayed tasks. When harmful activity is detected on a section of a transmission line, causing abnormal equipment heating and potentially triggering a tripping accident, the task of removing the harmful substance is considered an emergency task. If only a small amount of harmful activity is detected during routine inspections and has not yet significantly affected equipment operation, this is a routine task. If the harmful activity is in its early stages and poses a relatively small potential threat to the equipment, the task can be postponed and is therefore considered a delayed task.

[0116] When assigning weights to control importance level, environmental disturbance level, and task urgency factors based on the Delphi method, an expert team composed of power equipment operation and maintenance experts and ultrasonic pest control engineers was formed. Questionnaires were distributed to the experts, inviting them to score the importance of each indicator. Based on their experience and professional knowledge, the experts assessed the impact of different indicators on pest control decisions. After multiple rounds of feedback and adjustments, the final weights of each indicator were determined. Assuming that after the Delphi method assessment, the weight of control importance level is 0.4, the weight of environmental disturbance level is 0.3, the weight of task urgency factors is 0.2, and the weight of equipment status is 0.1 (the equipment status weight corresponds to the degree of influence of the equipment's current operating status, such as whether the equipment is under maintenance or whether the operating power is within the normal range).

[0117] A weighted summation method is used to construct the control model. Each index value is multiplied by its corresponding weight and then summed to obtain the control priority score for each equipment control task. Taking a substation pest control task as an example, this substation is a critical facility with a control importance score of 90 points (out of 100). Its environment is a moderately disturbed environment near an industrial area, with an environmental disturbance score of 70 points. The current hazardous activity has caused partial discharge in the equipment, classifying it as an urgent task, with an urgency factor score of 95 points and a good equipment condition score of 85 points. Therefore, the control priority score for this task is calculated as: 90×0.4+70×0.3+95×0.2+85×0.1=36+21+19+8.5=84.5 points.

[0118] Another example is a pest control task for a general power distribution equipment. This equipment is a non-critical facility, with a control importance score of 60 points. It is located in a suburban, low-interference environment, with an environmental interference score of 80 points. The pest activity is a small number of rodents discovered during routine inspections. The task urgency factor score is 60 points, and the equipment's normal condition score is 90 points. Its control priority score is calculated as: 60×0.4+80×0.3+60×0.2+90×0.1=24+24+12+9=69 points.

[0119] Based on the control priority scores calculated by the control model, all equipment control tasks are ranked, with tasks scoring higher having higher control priority, thus generating an equipment control priority sequence. In the example above, the substation hazard removal task scored 84.5 points, while the ordinary power distribution equipment hazard removal task scored 69 points. Therefore, the substation hazard removal task has a higher priority than the ordinary power distribution equipment task, and the substation hazard removal needs should be prioritized during resource allocation and scheduling.

[0120] Throughout the implementation process, the impact of various factors on the pest control task must be comprehensively considered. Through scientific weighting and scoring calculations, the priority sequence must reflect the actual importance and urgency of the task. For example, if a task is in a weakly disturbed environment but poses an urgent threat to a critical facility, its high score in control importance and urgency will place it at the top of the priority sequence. Conversely, if a task has a high degree of environmental disturbance but is a routine task for a non-critical facility, its overall score may be lower, resulting in a relatively lower priority. This multi-factor weighting model and priority calculation method provides a reasonable decision-making basis for the control of ultrasonic pest control equipment, enabling more optimized allocation and utilization of equipment resources and improving the overall efficiency and targeting of pest control work.

[0121] Example 5:

[0122] In this embodiment, when using an improved genetic optimization algorithm to solve for the optimal equipment control scheme, the optimization objectives are to maximize the pest control coverage, minimize equipment energy consumption, and shorten response time. A set of candidate solutions is randomly generated as the initial population. These candidate solutions can be understood as different combinations of ultrasonic equipment control schemes. For example, for an area containing multiple power facilities, each candidate solution in the initial population may correspond to different numbers of devices activated, different device transmission frequency settings, and different device operating times. Assuming there are 5 ultrasonic pest control devices in the area, each with different frequency adjustment levels and operating time settings, a candidate solution might be a combination such as device 1 using frequency A and operating for 2 hours, device 2 using frequency B and operating for 3 hours, etc. The initial population then contains multiple similar randomly generated combinations.

[0123] The fitness value of each candidate solution is calculated based on the optimization objectives. These objectives include maximizing pest control coverage, minimizing equipment energy consumption, and shortening response time. The pest control coverage objective is achieved by calculating the overlap between the candidate solution and the pest's activity area. For example, if early monitoring determines that the pest mainly operates in a specific area around power facilities, and the emission angle and effective distance of the ultrasonic equipment in each candidate solution determine its coverage area, then calculating the percentage of overlap between this coverage area and the pest's activity area yields the candidate solution's performance in terms of pest control coverage. The equipment energy consumption objective is achieved by limiting the equipment power and operating time range of the candidate solutions. Different equipment power and operating times result in different energy consumption. Summing the energy consumption of all equipment in each candidate solution gives the solution's energy consumption value; a smaller value indicates lower energy consumption. The response time objective is achieved by reducing the number of algorithm iterations and command transmission time. When calculating fitness, the ability to quickly generate and transmit control commands corresponding to the candidate solution is considered. Although initially reflected in the evaluation of the simplicity and efficiency of the solution itself, this affects the overall fitness value.

[0124] The optimal solution is selected as the parent individual based on its fitness value, and the crossover and mutation probabilities of other individuals are updated. Candidate solutions with higher fitness values—that is, solutions that are closer to maximizing pest control coverage, minimizing energy consumption, and shortening response time—are more likely to be selected as parents. For example, among the initial 100 candidate solutions, the top 20 solutions in terms of fitness are prioritized as parents. The updates to the crossover and mutation probabilities are based on the fitness of the parent individuals. Generally, the higher the fitness of the parent, the more favorable its crossover and mutation probabilities will be adjusted to values ​​that are more conducive to producing superior offspring. For example, the crossover probability might be increased from 0.6 to 0.8, and the mutation probability from 0.01 to 0.02, to increase the diversity and optimization potential of new solutions.

[0125] By simulating the gene recombination and mutation behavior of the population, candidate solutions are iteratively updated until the termination condition is met. Gene recombination is a crossover operation, such as selecting two parent candidate solutions and exchanging some of their control parameters to generate new offspring candidate solutions. For example, parent 1 is device 1 operating at frequency A for 2 hours and device 2 operating at frequency B for 3 hours; parent 2 is device 1 operating at frequency C for 4 hours and device 2 operating at frequency A for 2 hours. After crossover, offspring 1 may be device 1 operating at frequency A for 4 hours and device 2 operating at frequency B for 2 hours. Mutation behavior involves randomly changing individual parameters of a candidate solution, such as changing the operating frequency of a device from level A to level B, or changing the operating time from 2 hours to 3 hours, to increase population diversity and avoid the algorithm getting trapped in local optima. During the iteration process, after each new offspring population is generated, the fitness value of each solution is recalculated, and the selection, crossover, and mutation operations are repeated until the preset termination conditions are met, such as the number of iterations reaching 1000, or the fitness value of the optimal solution not significantly improving in 50 consecutive iterations.

[0126] A set of optimal solutions is selected from the final population as the optimal equipment control scheme. After multiple rounds of iterative optimization, the candidate solution with the highest fitness value among the final population is the optimal scheme. For example, after iteration, if a candidate solution achieves 90% overlap of the pest activity area in terms of pest control coverage, has the lowest equipment energy consumption of 1500 watt-hours among all solutions, and has a short generation and transmission time for control commands, and has the highest overall fitness value, then this solution will be selected as the optimal equipment control scheme.

[0127] The interactive control platform outputs the final equipment control results and continuously tracks feedback on the actual pest control effect. The platform intuitively displays the equipment control parameters of the optimal solution, such as the operating frequency, emission intensity, and running time of each device, allowing operators to control the equipment based on platform prompts. Simultaneously, the platform collects real-time equipment operation data and pest activity monitoring data to track the actual pest control effect. For example, it records whether the activity frequency of pests around power facilities decreases after implementing the solution, and whether the equipment's energy consumption matches expectations. While this feedback data is not used for hypothetical effect claims, it serves as a reference for subsequent optimization algorithms and solutions, helping to continuously improve the control methods of ultrasonic pest control equipment.

[0128] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0129] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A control method for ultrasonic destructive devices applied to power facilities, characterized in that, Includes the following steps: Real-time collection of environmental parameter data around power facilities; construction of a multi-dimensional pest characteristic identification system; and output of standardized pest activity signals through a signal processing module. An ultrasonic parameter database is established based on frequency domain conversion technology. When a pest control requirement is detected, the target parameters are located and equipment control commands are generated through feature matching algorithm. Establish a database of historical power plant damage cases and dynamically record usage records of different types of power facilities, environmental conditions, and ultrasonic parameters. A multi-factor weight allocation model is established. By combining the importance of facilities, the level of hazard of hazardous substances, the degree of environmental interference and the urgency of tasks, the Delphi method is used to assign weights to each indicator. A weighted summation method is used to construct a control model to calculate the control priority score, thereby generating an equipment control priority sequence. An improved genetic optimization algorithm is used to solve for the optimal equipment control scheme, with optimization objectives set in the direction of maximizing pest control coverage, minimizing equipment energy consumption, and shortening response time; The final equipment control results are output through the interactive control platform, and the actual pest control effect feedback is continuously tracked. The system collects real-time environmental parameter data around power facilities, constructs a multi-dimensional pest characteristic identification system, and outputs standardized pest activity signals through a signal processing module. Specifically: Based on array-type ultrasonic sensors, the activity signal data of harmful objects around power facilities are obtained. The signal acquisition terminal receives the raw signal from the sensor and performs noise suppression, time domain analysis, frequency domain conversion preprocessing to remove environmental background clutter. Based on the biological characteristics of pest activity and the ultrasonic reflection pattern, correlation analysis was used to screen parameters closely related to pest characteristics as candidate features. Candidate features are divided into frequency features, amplitude features, and time-frequency features, and a multi-level feature system is constructed. The overall pest identification rate is used as the top-level feature, and the top-level feature is decomposed into several first-level features. Each first-level feature is further subdivided into several second-level features. The signal processing module is trained using a wavelet transform model based on a multi-level feature system. The location information of power facilities is associated with the harmful characteristics of the pre-processed signals to establish a location-feature mapping relationship; Signal analysis software was selected as the pest control platform, and the area was marked on the interface according to the actual layout of the power facilities. Different labels were used for visualization based on the differences in the characteristics of the pests. Areas with large differences in characteristics were marked with prominent labels, while areas with small differences in characteristics were marked with conventional labels. The establishment of the ultrasonic parameter database based on frequency domain transformation technology is specifically as follows: Collect sensitive frequency data for common types of hazards and types of power facilities, including standard parameters such as fundamental frequency values, frequency range values, and intensity thresholds; Obtain the physical attribute information of the parameters, including device model, emission angle, and effective distance parameters; Parameters are abstracted into data nodes, and the frequency similarity between parameters is abstracted into data associations, thus constructing a node-association data structure; Establish the database architecture, including defining data tables, setting fields, and establishing data relationships, storing parameter information, attribute information, and associations; The preprocessed data is imported into the database of the signal analysis software to establish an ultrasonic parameter database; The step of locating target parameters and generating device control commands through feature matching algorithms specifically involves: Based on the ultrasonic parameter database, each parameter node is regarded as a sample point in the dataset, and the frequency correlation between parameters is regarded as the similarity between samples. Using the characteristics of pests in the area to be repelled as query points, the feature distance to each sample point is calculated using a fuzzy matching algorithm. When the sample point with the highest matching degree is reached, the parameter information of that sample point is recorded. The characteristic distances obtained from the analysis and calculation are combined with information on power facility type and environmental interference to determine the matching range of the target parameters. Extract the basic frequency value, frequency range value, and intensity threshold parameter of the target parameter from the ultrasonic parameter database to generate equipment control parameters; The target parameters, control parameters, facility type, and environmental conditions are combined into a label and then visualized on the signal analysis software interface.

2. The control method for ultrasonic destructive equipment applied to power facilities according to claim 1, characterized in that, The aforementioned construction of a historical disaster prevention case database dynamically records usage records for different types of power facilities, environmental conditions, and ultrasonic parameters. Specifically: The main dimensions for constructing the information database include different types of power facilities, environmental conditions, and usage records of ultrasonic parameters. At the same time, specific information fields are planned for each dimension, including facility model, ambient temperature and humidity, frequency of use, actual pest control effect, and user feedback and evaluation. By connecting the power monitoring and management system with the user feedback platform, we continuously update the usage parameters, effect feedback, and equipment operation information of historical pest control cases. A database of historical hazard cases is constructed based on key dimensions and data fields.

3. The control method for ultrasonic destructive equipment applied to power facilities according to claim 2, characterized in that, The multi-factor weight allocation model is established by combining the importance of facilities, the level of hazard from hazardous substances, the degree of environmental disturbance, and the urgency of the task. The Delphi method is used to assign weights to each indicator, and a weighted summation method is employed to construct a control model to calculate the control priority score, thereby generating an equipment control priority sequence. Specifically: Based on the importance of the facilities and the hazard level of the pollutants, the control importance level is divided into different levels. The requirements of critical facilities or high-hazard pollutants are set as high importance level, while the requirements of non-critical facilities or low-hazard pollutants are set as low importance level. The degree of environmental interference is divided into strong interference environment, medium interference environment, and weak interference environment. The urgency factors of tasks are categorized into urgent tasks, routine tasks, and delayed tasks; Based on the Delphi method, weights are assigned to the control importance level, the degree of environmental disturbance, and the urgency of the task. A weighted summation method is used to construct the control model. Each index value is multiplied by its corresponding weight and then summed to obtain the control priority score for each device control task. Based on the control priority scores calculated by the control model, all equipment control tasks are sorted, with tasks with higher scores having higher control priority, thus generating an equipment control priority sequence.

4. The control method for ultrasonic destructive equipment applied to power facilities according to claim 3, characterized in that, The improved genetic optimization algorithm is used to solve for the optimal equipment control scheme. The optimization objectives are set in the direction of maximizing the pest control coverage, minimizing equipment energy consumption, and shortening response time. Specifically: A set of candidate solutions is randomly generated as the initial population; Calculate the fitness value for each candidate solution based on the optimization objective; The optimal solution is selected as the parent individual based on the fitness value, and the crossover and mutation probabilities of other individuals are updated. By simulating the gene recombination and mutation behavior of the population, the candidate solutions are continuously updated iteratively until the termination condition is met; Select a set of optimal solutions from the final population as the optimal equipment control scheme.

5. The control method for ultrasonic destructive equipment applied to power facilities according to claim 4, characterized in that, The control priority score is specifically as follows: The control priority score is jointly determined by the control importance level weight, environmental disturbance degree weight, task urgency weight, and equipment status weight. The control importance level weight corresponds to the degree of influence of the facility's importance and the level of hazard level; the environmental disturbance degree weight corresponds to the degree of influence of environmental conditions; the task urgency weight corresponds to the degree of urgency of the hazard removal requirement; and the equipment status weight corresponds to the degree of influence of the equipment's current operating status. The final control priority score is obtained by multiplying each weight by its corresponding level score and then summing them up.

6. The control method for ultrasonic destructive equipment applied to power facilities according to claim 5, characterized in that, The optimization objective is specifically as follows: The optimization objectives include maximizing the pest control coverage, minimizing equipment energy consumption, and shortening response time. The pest control coverage objective is achieved by calculating the overlap between candidate solutions and the pest activity area. The equipment energy consumption objective is achieved by limiting the equipment power and operating time range of candidate solutions. The response time objective is achieved by reducing the number of algorithm iterations and instruction transmission time. The final optimization objective is the comprehensive optimization of the three sub-objectives.

Images