Iron tower foundation winter intelligent maintenance method fused with environment perception algorithm

Abstract

The invention provides an iron tower foundation winter intelligent maintenance method fused with an environment perception algorithm, and relates to the technical field of data processing. Images captured by a visual camera are recognized through a computer vision technology, scene data of a target iron tower are obtained, and environment data perception is performed on the target iron tower through an environment sensor array; obtaining basic environment data of the target iron tower; the scene data and the basic environment data are fused, an iron tower basic state database is input into a simulation prediction space for iron tower state prediction, an iron tower maintenance scheme is generated according to an iron tower prediction result in combination with historical maintenance records, and intelligent iron tower maintenance is carried out. The technical problem that in the prior art, the iron tower environment in winter is changeable, preventive maintenance cannot be conducted in time, and consequently the maintenance work efficiency is low is solved. The technical effects that the intelligent preventive maintenance of the iron tower in winter is achieved, the maintenance efficiency is improved, and the safety risk caused by changeable winter environments is reduced are achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a winter intelligent maintenance method for a tower foundation integrating an environmental perception algorithm. Background Art

[0002] With the continuous growth of energy demand in modern society and the increasing complexity of power transmission networks, the safety and stability of the operation of iron towers, as important supporting structures for power transmission and communication networks, have attracted much attention. Especially in winter, severe environmental conditions such as severe cold, wind and snow have a significant impact on the structure of iron towers. Traditional iron tower maintenance mainly relies on regular inspections and manual maintenance, which not only consumes a lot of manpower and material resources, but also makes it difficult to achieve real-time monitoring and preventive maintenance due to the variability of environmental conditions.

[0003] In the prior art, there are technical problems that the tower environment is changeable in winter, maintenance mainly relies on regular inspections and manual maintenance, and preventive maintenance cannot be carried out in time, resulting in low efficiency of maintenance work. Summary of the invention

[0004] The present application provides an intelligent winter maintenance method for tower foundations that integrates an environmental perception algorithm, which is used to solve the technical problems in the prior art that the tower environment in winter is changeable, maintenance mainly relies on regular inspections and manual maintenance, and preventive maintenance cannot be performed in a timely manner, resulting in low efficiency of maintenance work.

[0005] In view of the above problems, the present application provides a winter intelligent maintenance method for a tower foundation that integrates an environmental perception algorithm, the method comprising: identifying images captured by a visual camera through computer vision technology to obtain scene data of a target tower, the scene data including the tower structure and the tower appearance status; performing environmental data perception on the target tower through an environmental sensor array to obtain basic environmental data of the target tower; fusing the scene data and the basic environmental data to form a tower foundation status database; inputting the tower foundation status database into a simulation prediction space to perform tower status prediction to obtain a tower prediction result; generating a tower maintenance plan based on the tower prediction result combined with historical maintenance records to perform intelligent tower maintenance.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The method provided in the embodiment of the present application uses computer vision technology to identify the image captured by the visual camera to obtain the scene data of the target tower, and the scene data includes the structural features of the tower and the appearance status of the tower; the environmental data of the target tower is sensed through the environmental sensor array to obtain the basic environmental data of the target tower; the scene data and the basic environmental data are integrated to form a tower basic status database; the tower basic status database is input into the simulation prediction space to predict the tower status and obtain the tower prediction result; according to the tower prediction result combined with the historical maintenance records, a tower maintenance plan is generated to perform intelligent tower maintenance. The intelligent winter preventive maintenance of the tower is achieved, the maintenance efficiency is improved, and the technical effect of reducing the safety risks caused by the changeable winter environment is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A flow chart of the winter intelligent maintenance method for the tower foundation integrating the environmental perception algorithm provided in this application;

[0010] Figure 2 A schematic diagram of the process of forming a tower foundation status database in the winter intelligent maintenance method of the tower foundation that integrates the environmental perception algorithm provided in this application. DETAILED DESCRIPTION

[0011] This application provides a winter intelligent maintenance method for iron tower foundations that integrates an environmental perception algorithm, which is used to solve the technical problem in the prior art that the iron tower environment is changeable in winter, maintenance mainly relies on regular inspections and manual maintenance, and preventive maintenance cannot be carried out in time, resulting in low efficiency of maintenance work. The intelligent preventive maintenance of iron towers in winter is achieved, the maintenance efficiency is improved, and the technical effect of reducing the safety risks caused by the changeable winter environment is achieved.

[0012] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0013] like Figure 1 As shown, the present application provides a winter intelligent maintenance method for a tower foundation integrating an environmental perception algorithm, the method comprising:

[0014] The images captured by the visual camera are recognized by computer vision technology to obtain the scene data of the target tower, wherein the scene data includes the tower structure and the appearance status of the tower.

[0015] Specifically, the high-definition visual cameras deployed around the tower are used to automatically capture information related to the target tower. For example, the visual camera is an infrared camera to enhance visibility at night or in severe winter weather. Then, the collected original image is preprocessed to remove image noise, adjust brightness contrast, etc., so that the structure and appearance details of the tower are more obvious. The target detection algorithm in computer vision technology is used to identify the preprocessed image, extract and analyze the key features in the image, wherein the geometric features of the tower structure in the image are extracted, including angle, length, surface curvature, etc., to form the tower structure features. The tower structure feature data can help determine whether the tower is deformed, broken or has other structural problems. For the appearance state of the target tower, the surface texture, color change and rust of the target tower are analyzed, and the tower appearance state features are extracted. The tower appearance state can reflect whether the tower has surface damage, rust extension and other problems. All the above-identified feature data are integrated to generate complete scene data of the target tower, and the scene data includes the tower structure features and the tower appearance state. Based on the scene data captured by the camera image, the structure and appearance of the tower can be accurately identified and analyzed, which can provide comprehensive basic information for subsequent intelligent maintenance decisions and ensure the safe operation of the tower.

[0016] Through the environmental sensor array, the environmental data of the target tower is sensed to obtain the basic environmental data of the target tower.

[0017] Furthermore, environmental data of the target tower is sensed through an environmental sensor array to obtain basic environmental data of the target tower, including: determining monitoring data, the monitoring data including temperature, humidity, wind speed and snowfall; determining sensor types based on the monitoring data, the sensor types including temperature sensors, humidity sensors, anemometers and snow meters; planning the layout of multiple sensors based on the structural characteristics of the tower and the monitoring data to generate the environmental sensor array; based on the environmental sensor array, real-time perception and collection of environmental data of the target tower are performed to form the basic environmental data.

[0018] Specifically, the key environmental monitoring data of the target tower is first determined. In winter, low temperature may increase the brittleness of the tower foundation materials (such as concrete and steel), which are prone to frost cracking or deformation. By monitoring the temperature, the low temperature environment can be discovered and warned in time. Humidity indicates the moisture content in the air, which may affect the corrosion condition of the tower, and wind speed and snowfall have a direct impact on the structural stability of the tower, such as wind load and snow pressure monitoring data. Therefore, the monitoring data is determined to include temperature, humidity, wind speed and snowfall. Then, according to the determined monitoring data type, the appropriate sensor type is selected. The sensor type refers to the equipment for monitoring specific environmental parameters. For example, temperature monitoring requires a temperature sensor, humidity monitoring uses a humidity sensor, an anemometer is used to measure wind speed and wind direction, and a snow gauge is used to measure snowfall. Then, according to the structural characteristics of the target tower and the monitoring data, the specific installation position and arrangement of the sensor are formulated to ensure the comprehensiveness and accuracy of the data. After completing the layout of multiple sensors, an environmental sensor array is generated. The sensor array refers to a collaborative monitoring network composed of multiple sensors in a specific layout and connection method, which can obtain environmental data around the tower in real time. After the sensor array is deployed, the environment of the target tower is monitored and data is collected in real time. Real-time perception means that the sensor can continuously and uninterruptedly obtain environmental data to ensure that any changes can be immediately sensed and recorded. Data collection includes environmental parameters in multiple dimensions such as temperature, humidity, wind speed, wind direction and snowfall. Based on the environmental sensor array, real-time monitoring and data collection of the environment around the target tower are achieved to form the basic environmental data. By obtaining basic environmental data, the winter climate conditions and their changing trends in the environment where the target tower is located can be reflected, which helps to understand the impact of the environment on the tower status and identify environmental factors that may cause defects.

[0019] The scene data and the basic environment data are integrated to form a tower basic status database.

[0020] Specifically, the visual information and environmental information are comprehensively considered, that is, the obtained scene data and the basic environmental data are integrated to form a tower basic status database that comprehensively reflects the status of the target tower, providing data support for the subsequent intelligent maintenance of the tower.

[0021] In one embodiment, Figure 2As shown, the scene data and the basic environment data are merged to form a tower basic status database, including: synchronously sorting the scene data and the basic environment data according to the acquisition timestamp; analyzing the scene data to obtain tower scene features, the tower scene features include tower status, defect type, and defect location; analyzing the basic environment data to obtain basic environment features, the basic environment features include environmental parameter values ​​and change trends; according to the timestamp information after synchronous sorting, the tower scene features and the basic environment features are correlated and analyzed to generate correlation analysis results under multiple time frames; the tower scene features, the basic environment features and the correlation analysis results under multiple time frames are stored in the tower basic status database.

[0022] Specifically, according to the timestamp information of the collected data, that is, the specific time point when the data is collected, the scene data and the basic environment data are time-stamped and synchronously sorted to ensure that the two are aligned in time and avoid data analysis errors caused by time inconsistency. After synchronous sorting, the scene data is deeply analyzed, and scene feature information is extracted from the scene data, including the tower state, defect type and defect location, wherein the tower state is such as the overall health status of the tower, the defect type is such as rust, cracks, deformation, etc., and the defect location is the specific part of the tower where the problem occurs. By obtaining the tower scene features, basic data is provided for the subsequent tower state evaluation. Then, the basic environment data is analyzed, and basic environment features are extracted from the basic environment data, including environmental parameter values ​​and change trends, the environmental parameter values ​​are such as specific values ​​of the current temperature, humidity, wind speed, etc., and the change trends are such as whether the temperature gradually increases, whether the humidity changes periodically, etc., and the basic environment features are obtained to provide a data basis for the tower operation environment and its impact on the tower state.

[0023] Next, by analyzing the data of the scene features and the basic environmental features at the same timestamp, the impact of environmental changes on the state of the tower is identified, for example, the impact of changes in the environment such as temperature, humidity, snowfall, etc. on the state of the tower, such as aggravated corrosion and crack expansion. During the analysis process, the correlation analysis results under multiple time frames are generated. A time frame refers to a data segment divided in time units. In each time frame, the correlation analysis will show how the state of the tower and the environmental features interact with each other and how they change over time, thereby identifying the key influencing factors and changing trends. Finally, the tower scene features, the basic environmental features, and the correlation analysis results under multiple time frames are stored in the basic state database of the tower. By correlating the tower scene features and environmental features through timestamps, the relationship between the change of the tower state and the environmental conditions can be revealed. The tower scene features, the basic environmental features, and the correlation analysis results are stored to form a basic state database of the tower, which provides a basis for the subsequent state prediction of the tower and improves the accuracy and real-time performance of risk prediction.

[0024] The tower foundation status database is input into the simulation prediction space to perform tower status prediction and obtain tower prediction results.

[0025] Furthermore, the fused tower foundation status database is input into the simulation prediction space to perform tower status prediction to obtain a tower prediction result. The simulation prediction space includes a time series analysis technology, which includes: extracting historical tower scene features, historical basic environmental features and historical correlation analysis results from the tower foundation status database; dividing the historical tower scene features and the historical basic environmental features into multiple time series data sets in chronological order according to a predefined time window; using the multiple time series data sets to train a prediction model to obtain a time series prediction model; and deploying the trained time series prediction model in the simulation prediction space.

[0026] Specifically, the tower basic status database is a comprehensive database that stores the status information of the tower in different time periods, including the scene characteristics, environmental characteristics of the tower and the correlation analysis results between the two. Relevant historical data is extracted from the tower basic status database, including historical tower scene characteristics, historical basic environmental characteristics and historical correlation analysis results. According to the predefined time window, the continuous historical data is divided into independent data sets in multiple time periods in chronological order, that is, multiple time series data sets. The time window is the time period used to segment the data, and the length of the time window depends on the frequency of the tower state change and the prediction requirements, such as one week, half a month, etc. The time series data set is a data set segmented according to the time window, and each data set contains the tower scene characteristics, environmental characteristics and correlation analysis results in a time period.

[0027] The prediction model is trained using the multiple time series data sets after segmentation. The prediction model is a data model specifically used to predict time-based changes. By learning the time series patterns in historical data, the state of the future tower can be predicted. Optionally, the prediction model includes ARIMA, LSTM, etc. The prediction model training is a process of data input and model parameter adjustment. By repeatedly inputting the time series data set, the parameters of the prediction model are adjusted so that it can accurately capture and predict the time dependency and change trend in the data, continuously minimize the prediction error, and ensure that the model can effectively predict the future tower state until the model output reaches convergence, and the trained time series prediction model is obtained. The trained time series prediction model is deployed in the simulation prediction space, so that the time series prediction model can perform prediction tasks in practical applications. The deployed model can receive new data in real time, perform state prediction, and generate prediction results for the future tower state. The time dependency in the data is captured through time series analysis, ensuring that the prediction model can effectively predict the future tower state.

[0028] The tower foundation status database is input into the simulation prediction space, and the tower status is predicted using the deployed time series prediction model to finally obtain the tower prediction result. The tower prediction result is the output generated by the time series prediction model based on the current tower scene data and environmental basic data, including a specific description of the future tower status (such as possible defect types and defect locations) and a prediction of the impact of environmental conditions on the tower. By training the time prediction model and deploying it in the simulation prediction space, the future tower status can be effectively predicted, the prediction accuracy and reliability can be improved, and timely decision support can be provided for the winter maintenance of the tower, thereby ensuring the safe operation of the tower.

[0029] Based on the tower prediction results and historical maintenance records, a tower maintenance plan is generated to carry out intelligent tower maintenance.

[0030] Furthermore, a tower maintenance plan is generated based on the tower prediction results combined with historical maintenance records, including: collecting and organizing the tower's historical maintenance records, the historical maintenance records including historical maintenance, historical repairs and inspection reports; generating a tower maintenance plan using a multi-objective optimization algorithm in combination with the tower prediction results and the historical maintenance records; dividing the tower maintenance plan into multiple stages, each stage including specific maintenance tasks and schedules; and allocating resources and priorities to the maintenance plan for each stage based on the tower prediction results, the scenario data and the basic environmental data.

[0031] Specifically, database query tools are used to extract and organize the historical maintenance records of the tower from electronic databases such as the power company database and the tower maintenance management database. The historical maintenance records are mainly divided into three categories: historical maintenance, historical repairs, and inspection reports. Historical maintenance refers to regular maintenance work, such as cleaning, lubrication, tightening and other daily maintenance operations. These records help understand the regular maintenance status of the tower. Historical repairs involve repair operations performed on the tower in the past, including repairing cracks, replacing damaged parts, etc., which are used to reflect the specific problems that occurred in the tower at different time points and their repair status. The inspection report is a detailed report that evaluates the status of the tower. It is usually generated by professionals after regular inspections and contains information such as the structural condition of the tower and potential risks.

[0032] After obtaining the historical maintenance records of the tower, combined with the tower prediction results obtained through the simulation prediction space, a multi-objective optimization algorithm is used to comprehensively consider multiple goals such as risk, time, resources, and cost, and generate a tower maintenance plan while meeting all goals. For example, the multi-objective optimization algorithm will balance the relationship between repair costs and maintenance effects to ensure the best maintenance effect within limited resources. The tower maintenance plan is a plan generated by the optimization algorithm, which includes all maintenance work arrangements for the tower in the future. The tower maintenance plan not only takes into account the current status and future predictions of the tower, but also integrates the lessons learned from historical maintenance records to ensure the comprehensiveness and rationality of the plan.

[0033] In order to facilitate the orderly implementation of the maintenance plan, the tower maintenance plan is divided into multiple stages, each with specific maintenance tasks and schedules. For example, the first stage may focus on emergency repairs and preventive maintenance, and the second stage may focus on daily inspections and small-scale repairs. The maintenance tasks of each stage are formulated according to the specific conditions and forecast results of the tower, including inspection, repair, monitoring and other activities, and a detailed timetable is formulated for each stage to clarify when and what maintenance activities will be carried out, to ensure that various tasks are carried out as planned, and to avoid resource conflicts and time delays.

[0034] Finally, according to the tower prediction results, scenario data and basic environmental data, resources are allocated and priorities are set for the maintenance plan for each stage. Maintenance work requires the investment of various resources, such as manpower, materials, equipment, etc. According to the task requirements of each stage, these resources are reasonably allocated to ensure that key tasks are adequately supported by resources. During the maintenance process, different tasks have different importance and urgency. According to the prediction results and environmental data, priorities are set for the tasks at each stage to ensure that the most critical and urgent tasks are given priority. For example, if the prediction results show that a component may be about to fail, the repair work for this component should be given priority. By combining the historical maintenance records with the future prediction results of the tower, a multi-objective optimization algorithm is used to generate a phased maintenance plan, and resources are reasonably allocated and priorities are set for each stage, so that preventive maintenance can be carried out in a timely manner, and maintenance work can be more organized, ensuring that key tasks are given priority and resources are optimally configured, improving the timeliness and comprehensiveness of the maintenance plan, and ensuring the effectiveness of the implementation of the maintenance plan.

[0035] Furthermore, in combination with the tower prediction result and the historical maintenance record, a tower maintenance plan is generated using a multi-objective optimization algorithm, including: obtaining tower maintenance demand information based on the tower prediction result; based on the tower maintenance demand information, traversing the historical maintenance records to generate N maintenance plans, where N is a positive integer greater than 1; based on the N maintenance plans, determining the optimal maintenance plan using a multi-objective optimization algorithm to obtain the tower maintenance plan.

[0036] Specifically, the prediction information of the future state of the tower obtained by the simulation prediction model includes possible defects, structural damage, environmental impact, etc. The specific maintenance requirements extracted from the prediction results indicate which parts need maintenance, repair or further inspection. For example, the prediction results may show that a key node is at risk of fatigue damage, then the corresponding maintenance requirements may include structural reinforcement or regular monitoring of the node. After obtaining the maintenance demand information, all relevant historical maintenance records, including repairs, maintenance, inspections, etc., are retrieved and analyzed one by one to extract useful information from them. For example, if a certain repair method has achieved good results in similar situations in the past, then this information can be used to generate a new maintenance plan. According to the results after traversal combined with the current maintenance demand information, multiple different maintenance plans are generated (N is a positive integer greater than 1), each of which may be based on different strategies, resource allocation methods or technical methods. For example, some plans focus on preventive maintenance, while others may focus more on emergency repairs. After generating N maintenance plans, a multi-objective optimization algorithm is used to determine the optimal maintenance plan. For example, we can make trade-offs among multiple goals such as cost, time, risk, and resource utilization, analyze the performance of each maintenance plan under these goals, and find the plan that best balances all goals. After screening by the optimization algorithm, the optimal plan is finally determined, namely the tower maintenance plan. The tower maintenance plan can complete the task with the lowest cost, shortest time, and optimal resource allocation while meeting all maintenance needs. Determining the optimal maintenance plan based on N maintenance plans using a multi-objective optimization algorithm can significantly improve the efficiency and pertinence of the winter intelligent maintenance of the tower foundation, optimize resource allocation, and improve the level of intelligence.

[0037] Furthermore, according to the structural characteristics of the tower and the monitoring data, the layout of multiple sensors is planned to generate the environmental sensor array, including: obtaining the structural characteristics of the target tower according to the scene data; dividing the priorities of the tower monitoring points according to the influence of the structural characteristics and the monitoring data; and deploying multiple sensors according to the division results to generate the environmental sensor array.

[0038] Specifically, according to the identified scene data, the structural features of the target tower are extracted. The structural features refer to the physical properties and geometric characteristics of the tower, including the height, node position, support points, distribution of key parts (such as connection nodes, tower top, tower base, etc.) of the tower, etc. The structural features can directly affect the stability and safety of the tower. After obtaining the structural features of the tower, the degree of influence of the monitoring data (such as wind speed, temperature, humidity, etc.) on the structural features (different parts) of the tower is determined. For example, strong winds may have a greater impact on the top of the tower, while temperature changes may have a greater impact on the stress distribution of the connection nodes. By analyzing the monitoring data, it can be determined which environmental factors have the most significant impact on which structural parts. According to the degree of influence of the structural features and monitoring data, it is determined which monitoring points are more critical, and the priority of the tower monitoring points is divided according to the degree of influence. The monitoring points with high priority are usually those parts that bear greater pressure in the tower structure or are more susceptible to environmental factors, such as the tower top, the main load-bearing structure, and the key connection points. According to the priority division results of the monitoring points, the sensors are arranged, and finally the environmental sensor array is generated. The environmental sensor array is a complete monitoring network, consisting of multiple sensors deployed at various key locations of the tower. These sensors can collect environmental data in real time, ensuring comprehensive monitoring of the tower structure, thereby improving the comprehensiveness and accuracy of the monitoring data and providing reliable data support for subsequent winter maintenance of the tower.

[0039] Furthermore, according to the structural characteristics and the degree of influence of the monitoring data, the priorities of the tower monitoring points are divided, including: dividing the priorities of the tower monitoring points according to the stress concentration areas of the tower structure, the parts that are easily affected by environmental factors, and the historical fault-prone points; the priorities include three levels, among which the first-level monitoring points are located at the key parts of the tower, including the tower feet, connection nodes, and the tower top; the second-level monitoring points are located at the parts with variable environment, including the windward side and the snow-covered side; the third-level monitoring points are located at the historical fault-prone points of the tower.

[0040] Specifically, based on the design and structural analysis of the tower, the stress concentration areas of the tower structure are identified. The stress concentration areas of the tower structure are the parts of the tower structure that bear the greatest stress or strain, including key parts such as the tower foot, connection nodes, and the tower top. These areas are the most prone to fatigue and damage due to their large bearing capacity and concentrated load. Therefore, these key parts are identified as the first-level priority points for tower monitoring. Next, based on the environmental data analysis and historical environmental conditions, the parts of the tower that are susceptible to environmental factors are determined. The parts that are susceptible to environmental factors are located on the windward side and snow-covered side of the tower. The windward side refers to the part of the tower that directly faces the wind direction, which is exposed to strong winds all year round and may cause wind erosion or structural fatigue. The snow-covered surface refers to the area covered with snow in winter, and the weight of the snow may cause additional loads on the structure. Therefore, these environmental abnormality sites are classified as secondary monitoring points. By analyzing historical fault data, the historical fault-prone points on the tower are identified. The historical fault-prone points refer to the parts where the tower frequently encountered problems in past use. These faults may be related to specific environmental conditions, structural defects, or long-term fatigue accumulation. Although these points may not be in areas with the greatest structural stress or the strongest environmental impact, they are classified as level 3 monitoring points because historical data shows that these parts have a higher probability of failure. Monitoring of these points helps prevent the same problem from happening again. After the priority division is completed, the sensors are deployed and monitoring begins. At the same time, the priority and monitoring intensity of the monitoring points are dynamically adjusted according to the real-time monitoring data and the operating status of the tower. By dividing the priorities of the tower monitoring points, the allocation of sensors for monitoring resources can be optimized to ensure that resources are used most effectively. In addition, it is possible to achieve a comprehensive and real-time grasp of the tower status, obtain more accurate and comprehensive monitoring data, provide strong data support for the winter maintenance and management of the tower, and improve the accuracy and timeliness of management.

[0041] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0042] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. The winter intelligent maintenance method of the iron tower foundation integrating the environmental perception algorithm is characterized by: The method comprises: The image captured by the visual camera is recognized by computer vision technology to obtain scene data of the target tower, wherein the scene data includes the structural features of the tower and the appearance status of the tower; Through the environmental sensor array, the environmental data of the target tower is sensed to obtain the basic environmental data of the target tower; The scene data and the basic environment data are integrated to form a tower foundation status database; Inputting the tower foundation status database into the simulation prediction space to perform tower status prediction and obtain tower prediction results; Based on the tower prediction results and historical maintenance records, a tower maintenance plan is generated to carry out intelligent tower maintenance.

2. The winter intelligent maintenance method for the iron tower foundation integrating the environment perception algorithm as claimed in claim 1 is characterized in that: Through the environmental sensor array, the target tower is perceived for environmental data, and the basic environmental data of the target tower is obtained, including: Determining monitoring data, wherein the monitoring data includes temperature, humidity, wind speed and snowfall; Determine the sensor type according to the monitoring data, the sensor type including a temperature sensor, a humidity sensor, an anemometer and a snow gauge; According to the structural characteristics of the tower and the monitoring data, planning the layout of multiple sensors and generating the environmental sensor array; Based on the environmental sensor array, the environmental data of the target tower is sensed and collected in real time to form the basic environmental data.

3. The winter intelligent maintenance method for the iron tower foundation integrating the environment perception algorithm as claimed in claim 2 is characterized in that: According to the structural characteristics of the tower and the monitoring data, the layout of multiple sensors is planned to generate the environmental sensor array, including: According to the scene data, the structural characteristics of the target tower are obtained; Prioritizing tower monitoring points according to the structural characteristics and the degree of influence of the monitoring data; A plurality of sensors are arranged according to the division result to generate the environmental sensor array.

4. The winter intelligent maintenance method for iron tower foundation integrating environment perception algorithm as claimed in claim 3 is characterized in that: According to the structural characteristics and the degree of influence of the monitoring data, the priorities of the tower monitoring points are divided, including: Prioritize tower monitoring points based on stress concentration areas of the tower structure, parts that are susceptible to environmental factors, and historical fault-prone points; The priority level includes three levels, among which the first-level monitoring points are located at the key parts of the tower, including the tower base, connection nodes, and tower top; the second-level monitoring points are located at parts with variable environment, including the windward side and snow-covered side; the third-level monitoring points are located at points where the tower has a history of frequent failures.

5. The winter intelligent maintenance method for iron tower foundation integrating environment perception algorithm as claimed in claim 1 is characterized in that: The scene data and the basic environment data are integrated to form a tower basic status database, including: Synchronously sorting the scene data and the basic environment data according to the acquisition timestamp; Analyze the scene data to obtain tower scene features, where the tower scene features include tower status, defect type, and defect location; Analyze the basic environmental data to obtain basic environmental characteristics, wherein the basic environmental characteristics include environmental parameter values ​​and change trends; According to the synchronously sorted timestamp information, correlation analysis is performed on the tower scene features and the basic environment features to generate correlation analysis results under multiple time frames; The tower scene features, the basic environment features and the associated analysis results under multiple time frames are stored in the tower basic status database.

6. The winter intelligent maintenance method for iron tower foundation integrating environment perception algorithm as claimed in claim 5 is characterized in that: The fused tower foundation status database is input into the simulation prediction space to perform tower status prediction to obtain tower prediction results, which includes: Extracting historical tower scene features, historical basic environment features and historical correlation analysis results from the tower foundation status database; According to a predefined time window, the historical tower scene features and the historical basic environment features are divided into multiple time series data sets in chronological order; Using the multiple time series data sets, training a prediction model to obtain a time series prediction model; The trained time series prediction model is deployed in the simulation prediction space.

7. The winter intelligent maintenance method for iron tower foundation integrating environment perception algorithm as claimed in claim 1 is characterized in that: Based on the tower prediction results and historical maintenance records, a tower maintenance plan is generated, including: Collect and organize the historical maintenance records of the tower, including historical maintenance, historical repair and inspection reports; Combining the tower prediction results with the historical maintenance records, generating a tower maintenance plan using a multi-objective optimization algorithm; Dividing the tower maintenance program into multiple stages, each stage including specific maintenance tasks and schedules; Resources and priorities are allocated to the maintenance plan for each stage based on the tower prediction results, the scenario data and the basic environment data.

8. The winter intelligent maintenance method for iron tower foundation integrating environment perception algorithm as claimed in claim 7 is characterized in that: Combining the tower prediction results and the historical maintenance records, a tower maintenance plan is generated using a multi-objective optimization algorithm, including: Obtaining tower maintenance demand information based on the tower prediction result; Based on the tower maintenance demand information, traverse the historical maintenance records and generate N maintenance plans, where N is a positive integer greater than 1; Based on the N maintenance plans, a multi-objective optimization algorithm is used to determine the optimal maintenance plan to obtain the tower maintenance plan.

Images