Finished product detection system and method of wind power blade, storage medium and electronic equipment

Abstract

The invention provides a finished product detection system and method for wind power blades, a storage medium and electronic equipment, and relates to the technical field of wind power blade material testing, and the system comprises a detection module which is used for detecting a plurality of wind power blades in a finished product area based on artificial intelligence detection body networking; the analysis module is used for carrying out working rational analysis on artificial intelligence detection body networking in real time; the optimization module is used for carrying out continuous optimization iteration on the artificial intelligence detection body networking based on the working rational analysis result; and the relay module is used for carrying out relay detection on each wind power blade on the basis of networking of the optimized and iterated artificial intelligence detection body after each optimization iteration. When a finished product of the wind power blade is subjected to material detection, damage and defects of the wind power blade are automatically identified through artificial intelligence control, and the detection efficiency is improved. And continuous optimization iteration is carried out on artificial intelligence detection body networking, so that the working capability is optimized, and the data processing level of a system for carrying out material testing on a finished product of the wind power blade is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine blade material testing, and particularly to a finished product detection system, method, storage medium and electronic device for wind turbine blades. Background Art

[0002] Currently, as a core component of wind turbine units, wind turbine blades directly affect the operation stability and power generation efficiency of wind turbines. Wind turbine blades usually need to withstand strong winds, extreme climates and high-load working environments, and their manufacturing quality is crucial. Once quality defects occur, they will not only affect the long-term operation of wind turbines, but may also pose safety hazards. Therefore, the finished product detection of wind turbine blades is particularly important, and material testing of them is a key link to ensure the normal operation of wind turbines.

[0003] Currently, the finished product detection of wind turbine blades still mainly relies on manual operations. Workers conduct inspections through visual inspection, manual measurement and traditional instruments. Although this traditional manual detection method can, to a certain extent, detect surface defects and problems, there are some significant deficiencies. First, manual detection relies on a large number of workers to cooperate, with low efficiency and unable to comprehensively detect large-scale produced wind turbine blades in a short time. Second, manual operations are easily affected by human factors, such as improper operations and fatigue work, which will lead to large errors in detection results and even missed detection of some potential defects. Finally, traditional detection methods cannot meet the large-scale, fast and dynamically changing requirements in the production scenario of wind turbine blades, lacking real-time performance and flexibility. Overall, the existing technology has insufficient data processing level for material testing of the finished products of wind turbine blades.

[0004] Therefore, there is an urgent need for a more efficient, accurate and intelligent solution for material testing of the finished products of wind turbine blades. Summary of the Invention

[0005] One of the objectives of the present invention is to provide a finished product detection system for wind turbine blades. When conducting material detection on the finished products of wind turbine blades, an artificial intelligence detection body network is introduced, and a variety of detection bodies such as robots, drones and AGV vehicles are set up. Through artificial intelligence control, damage and defects of wind turbine blades are automatically identified, greatly improving the detection efficiency and enabling comprehensive detection of large-scale produced wind turbine blades in a short time. At the same time, through real-time working rational analysis, the system continuously optimizes and iterates the artificial intelligence detection body network, continuously optimizing the working ability of the artificial intelligence detection body network, greatly improving the accuracy of the finished product detection of wind turbine blades, better adapting to large-scale, fast and dynamically changing production requirements, enhancing the real-time performance and flexibility of the system, and further enhancing the data processing level of the system for material testing of the finished products of wind turbine blades.

[0006] A finished product detection system for a wind turbine blade provided by an embodiment of the present invention includes: A detection module for detecting multiple wind turbine blades in the finished product area based on an artificial intelligence detection body network; An analysis module for performing real-time working rationality analysis on the artificial intelligence detection body network; An optimization module for continuously optimizing and iterating the artificial intelligence detection body network based on the working rationality analysis result; A relay module for, after each optimization iteration, detecting each wind turbine blade in a relay manner based on the optimized and iterated artificial intelligence detection body network.

[0007] Optionally, the analysis module performing real-time working rationality analysis on the artificial intelligence detection body network includes: Selecting a reference divergence point from a sequence of decision divergence points generated by the artificial intelligence detection body network within a recently preset time; Retrospectively generating a historical working map of the target detection body corresponding to the reference divergence point within a target time period; wherein, the start and end times of the target time period are respectively the generation times of the decision divergence points before and after the reference divergence point in the sequence of decision divergence points; Based on the traversal sequence corresponding to the divergence point type of the reference divergence point, sequentially traversing each map feature in the historical working map; wherein, the map features at least include: the movement change of the minimum bounding box of the target detection body and the working interaction record, and the global change trend of the historical working map; Each time when traversing, when the traversed map feature matches the trigger map feature, stop continuing to traverse the next map feature, and obtain multiple groups of corresponding replay mechanisms and verification rules corresponding to the trigger map feature that matches the traversed map feature; Based on any one of the replay mechanisms, controlling the historical working map to be replayed, and executing the corresponding verification rule during the replay of the historical working map, and taking the execution result of the corresponding verification rule as the working rationality analysis result.

[0008] Optionally, the selecting a reference divergence point from a sequence of decision divergence points generated by the artificial intelligence detection body network within a recently preset time includes: Respectively determining a first target and a second target from the sequence of decision divergence points; Taking the first-generated one of the first target and the second target as the reference divergence point; Wherein, the determination step of the first target is as follows: Taking the jth decision divergence point in the sequence of decision divergence points as the first target; the sum of the cost values of the first j-1 decision divergence points in the sequence of decision divergence points is closest to the cost value sum threshold; Wherein, the determination step of the second target is as follows: Take the decision-making divergence point in the decision-making divergence point sequence whose first divergence point type is the same as the standard divergence point type as the second target.

[0009] Optionally, based on the work rationality analysis result, the optimization module continuously optimizes and iterates the artificial intelligence detection body network, including: Describe the characteristics of the work rationality analysis result to obtain a feature description vector; Determine the optimization iteration strategy corresponding to the feature description vector from the optimization iteration strategy library; Based on the optimization iteration strategy, continuously optimize and iterate the artificial intelligence detection body network.

[0010] Optionally, the finished product detection system of the wind turbine blade further includes: A visualization monitoring module for: Generate a visualization monitoring model of the artificial intelligence detection body network after optimization and iteration; Based on the visualization monitoring model, assist the management personnel to monitor the work of the artificial intelligence detection body network after optimization and iteration.

[0011] Optionally, the visualization monitoring module generates a visualization monitoring model of the artificial intelligence detection body network after optimization and iteration, including: Based on the visualization template corresponding to the personnel portrait of the management personnel, generate a visualization monitoring model according to the multi-modal work information of the artificial intelligence detection body network after optimization and iteration.

[0012] Optionally, the visualization monitoring module based on the visualization monitoring model assists the management personnel to monitor the work of the artificial intelligence detection body network after optimization and iteration, including: Detect multiple work monitoring timing events occurring in the visualization monitoring model; Based on each work monitoring timing event, generate a monitoring task quick selection table; Assist the management personnel to quickly select monitoring tasks from the monitoring task quick selection table and execute them accordingly.

[0013] A method for detecting the finished product of a wind turbine blade provided by an embodiment of the present invention includes: Based on the artificial intelligence detection body network, detect multiple wind turbine blades in the finished product area; Perform real-time work rationality analysis on the artificial intelligence detection body network; Based on the work rationality analysis result, continuously optimize and iterate the artificial intelligence detection body network; After each optimization and iteration, based on the artificial intelligence detection body network after optimization and iteration, relay to detect each wind turbine blade.

[0014] Optionally, the real-time work rationality analysis of the artificial intelligence detection body network includes: Select a reference divergence point from the sequence of decision divergence points generated within the most recent preset time of the artificial intelligence detection body network; Trace and generate the historical working map of the target detection body corresponding to the reference divergence point within the target time period; wherein, the start and end times of the target time period are respectively the generation times of the decision divergence points before and after the reference divergence point in the decision divergence point sequence; Based on the traversal order corresponding to the divergence point type of the reference divergence point, sequentially traverse each map feature in the historical working map; wherein, the map features at least include: the movement change of the minimum bounding box of the target detection body and the work interaction record, and the global change trend of the historical working map; Each time when traversing, when the traversed map feature matches the trigger map feature, stop continuing to traverse the next map feature, and obtain multiple groups of corresponding replay mechanisms and verification rules corresponding to the trigger map feature that match the traversed map feature; Based on any one of the replay mechanisms, control the historical working map to replay, and execute the corresponding verification rules during the replay of the historical working map, and use the execution result of the corresponding verification rule as the work rationality analysis result.

[0015] Optionally, the selecting a reference divergence point from the sequence of decision divergence points generated within the most recent preset time of the artificial intelligence detection body network includes: Respectively determine a first target and a second target from the decision divergence point sequence; Take the one that is generated first among the first target and the second target as the reference divergence point; Wherein, the determining step of the first target is as follows: Take the jth decision divergence point in the decision divergence point sequence as the first target; the sum of the cost values of the first j-1 decision divergence points in the decision divergence point sequence is closest to the cost value sum threshold; Wherein, the determining step of the second target is as follows: Take the decision divergence point whose first divergence point type is the same as the standard divergence point type in the decision divergence point sequence as the second target.

[0016] Optionally, the optimization module continuously optimizes and iterates the artificial intelligence detection body network based on the work rationality analysis result, including: Perform feature description on the work rationality analysis result to obtain a feature description vector; Determine the optimization iteration strategy corresponding to the feature description vector from the optimization iteration strategy library; Based on the optimization iteration strategy, continuously optimize and iterate the artificial intelligence detection body network.

[0017] Optionally, the finished product detection method of the wind turbine blade further includes: Generate a visual monitoring model for the network of AI detection bodies after optimization and iteration; Based on the visual monitoring model, assist the management personnel to monitor the work of the network of AI detection bodies after optimization and iteration.

[0018] Optionally, the generation of the visual monitoring model for the network of AI detection bodies after optimization and iteration includes: Based on the visual template corresponding to the personnel portrait of the management personnel, generate a visual monitoring model according to the multi-modal work information of the network of AI detection bodies after optimization and iteration.

[0019] Optionally, the assisting the management personnel to monitor the work of the network of AI detection bodies after optimization and iteration based on the visual monitoring model includes: Detect multiple work monitoring timing events occurring in the visual monitoring model; Based on each work monitoring timing event, generate a monitoring task quick selection table; Assist the management personnel to quickly select monitoring tasks from the monitoring task quick selection table and perform corresponding executions.

[0020] A computer-readable storage medium provided by an embodiment of the present invention, on which a computer program is stored, and the processor executes the computer program to implement the above method.

[0021] An electronic device provided by an embodiment of the present invention, the electronic device includes a memory and a processor, a computer program is stored in the memory, and the processor executes the computer program to implement the above method.

[0022] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written specification and the drawings.

[0023] The technical solution of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0024] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the drawings: Figure 1 It is a schematic diagram of a finished product detection system for a wind turbine blade in an embodiment of the present invention; Figure 2 It is a specific implementation schematic diagram of a finished product detection system for a wind turbine blade in an embodiment of the present invention; Figure 3Schematic diagram of a finished product inspection method for a wind turbine blade in an embodiment of the present invention. Detailed implementation manners

[0025] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for the purpose of illustrating and explaining the present invention, and are not used to limit the present invention.

[0026] An embodiment of the present invention provides a finished product inspection system for a wind turbine blade, as Figure 1 shown, including: A detection module 1, configured to detect multiple wind turbine blades in the finished product area based on an artificial intelligence detection body network. An analysis module 2, configured to perform real-time working rationality analysis on the artificial intelligence detection body network. An optimization module 3, configured to continuously optimize and iterate the artificial intelligence detection body network based on the working rationality analysis result. A relay module 4, configured to, after each optimization iteration, relay to detect each wind turbine blade based on the optimized and iterated artificial intelligence detection body network.

[0027] The artificial intelligence detection body network is a network of multiple detection bodies for performing wind turbine blade detection operations based on artificial intelligence technology. The detection body at least includes: robots, unmanned aerial vehicles, AGV cars, etc. It independently performs detection operation task planning, running trajectory planning, on-site obstacle avoidance control, etc. based on artificial intelligence technology. Specifically, each detection body is equipped with an advanced machine vision camera, and combined with a deep learning algorithm, it can automatically identify damages or defects of wind turbine blades; when performing detection operation task planning, it intelligently plans future detection operation objects based on the environmental layout and wind turbine blade distribution in the finished product area; the detection body uses environmental sensor data to real-time identify obstacles in the environment and perform automatic obstacle avoidance path adjustment to ensure operation safety.

[0028] Specifically, as Figure 2 shown, in specific implementation, the detection robot 1.1, the bottom detection device 1.2, and the upper detection device 1.3 in the detection module 1 cooperate to detect the wind turbine blade. The detection result is uploaded to the analysis module 2 for analysis, and the further analysis result is transmitted to the optimization module 3 to continuously optimize and iterate the artificial intelligence detection body network. Finally, it is handed over to the relay module 4 to relay and control the detection module 1 to detect each wind turbine blade based on the optimized and iterated artificial intelligence detection body network.

[0029] The finished product area is an area where the finished wind turbine blades are stacked after production and manufacturing.

[0030] During the process of using an artificial intelligence detection body network to detect multiple wind turbine blades in the finished product area, perform a rational analysis of its work to determine whether continuous optimization and iteration are required.

[0031] Next, based on the results of the rational analysis of the work, continuously optimize and iterate the artificial intelligence detection body network.

[0032] Finally, after each optimization and iteration, based on the optimized artificial intelligence detection body network, continue to detect each wind turbine blade in relay.

[0033] In this application, when detecting the materials of the finished wind turbine blades, an artificial intelligence detection body network is introduced, and a variety of detection bodies such as robots, drones, and AGV vehicles are set up. Through artificial intelligence control, the damage and defects of the wind turbine blades are automatically identified, greatly improving the detection efficiency, and enabling a comprehensive detection of a large number of wind turbine blades produced in a short time. At the same time, through real-time rational analysis of the work, the artificial intelligence detection body network is continuously optimized and iterated, continuously optimizing the working ability of the artificial intelligence detection body network, greatly improving the accuracy of the detection of the finished wind turbine blades, better adapting to the production requirements of large-scale, fast, and dynamically changing, enhancing the real-time and flexibility of the system, and further enhancing the data processing level of the system for testing the materials of the finished wind turbine blades.

[0034] In one embodiment, the analysis module performs a rational analysis of the artificial intelligence detection body network in real time, including: Select a reference divergence point from the sequence of decision divergence points generated by the artificial intelligence detection body network within the most recent preset time; Trace and generate the historical work map of the target detection body corresponding to the reference divergence point within the target time period; where the start and end times of the target time period are respectively the generation times of the decision divergence points before and after the reference divergence point in the sequence of decision divergence points; Based on the traversal order corresponding to the divergence point type of the reference divergence point, sequentially traverse each map feature in the historical work map; where the map features at least include: the movement change of the minimum bounding box of the target detection body and the work interaction record, and the global change trend of the historical work map; Each time when traversing, when the traversed map feature matches the trigger map feature, stop traversing the next map feature, and obtain multiple groups of one-to-one corresponding replay mechanisms and verification rules corresponding to the trigger map feature that matches the traversed map feature; Based on any replay mechanism, control the historical work map to be replayed, and execute the corresponding verification rules during the replay of the historical work map, and use the execution result of the corresponding verification rule as the result of the rational analysis of the work.

[0035] When the artificial intelligence detection body is networked to inspect the finished products of wind turbine blades, each detection body needs to continuously make task planning decisions, path planning decisions and obstacle avoidance control decisions. Every time a decision divergence occurs, a decision divergence point will be formed (for example: multiple planning tasks are decided, and then one of them is selected for final execution, then the multiple planning tasks decided and one of the planning tasks finally executed form a decision divergence point). The decision divergence points generated by the artificial intelligence detection body network within the most recent preset time (for example: 100 seconds) are sorted in chronological order to obtain a decision divergence point sequence.

[0036] First, a benchmark divergence point is selected from the decision divergence point sequence, and the benchmark divergence point is used as the benchmark for rational analysis of work. Then, the historical work map is generated retroactively. The historical work map is a map that reflects the working conditions of the target detection body in the target period of wind turbine blade finished product inspection from the time and space dimensions. The corresponding to the benchmark divergence point means that the benchmark divergence point is generated by the target detection body. The benchmark divergence point is used as the benchmark for rational analysis of work. The target time can be taken as the period between the generation time of the previous and next decision divergence points, that is, the first and last moments of the target period are the generation time of the previous and next decision divergence points respectively. Then, each map feature in the historical work map is traversed in turn. Each time the traversal is reached, when the traversed map feature matches the trigger map feature, it means that the current replay verification can determine the ideal analysis result of the work. Finally, the replay mechanism and verification rules are indicated by the matching trigger map features, and the mechanism execution and corresponding verification are carried out in turn, and the execution results are used as the results of rational analysis of work.

[0037] Specifically, the divergence point type of the reference divergence point corresponds to a traversal order, in which there is a traversal priority order corresponding to different map features. In the traversal order, the more likely it is to reflect the current replay verification to determine the ideal analysis result of the work, the higher it is ranked. For example: if the divergence point type of the reference divergence point is the obstacle avoidance path decision divergence, then in the corresponding traversal order, the map feature with the minimum bounding box movement change can determine whether the obstacle avoidance is successful in the end, and the map feature will be traversed first. In addition, the triggering map feature can be, for example, the minimum bounding box movement change of the target detection body is unstable, and accordingly, the replay mechanism and verification rules can be, for example, controlling the historical working map to perform obstacle avoidance replay to verify whether the obstacle avoidance process of the target detection body is smooth.

[0038] When the embodiment of the present invention conducts a working rationality analysis on the artificial intelligence detection body networking, a decision divergence point sequence is introduced, and a benchmark divergence point for conducting the working rationality analysis is selected from it, which improves the working rationality analysis efficiency; traces and generates a historical working map, and uses the generation times of the benchmark divergence point and the previous and next decision divergence points in the decision divergence point sequence as the start and end times to accurately generate the target time period, which improves the accuracy of tracing and generating the historical working map, reduces the tracing and generating resources, and improves the tracing and generating efficiency; based on the traversal order, traverses each map feature in the historical working map in turn, and the more likely it is to reflect the current replay verification, the more priority is given to traversing the working ideal analysis result, which improves the efficiency of determining the timing of the replay verification; matches the trigger map features that conform to the replay mechanism and verification rules, and executes one by one, and takes the execution result as the working rationality analysis result, which further improves the working rationality analysis efficiency, and overall improves the accuracy and comprehensiveness of the working rationality analysis.

[0039] In one embodiment, the selecting a benchmark divergence point from the decision divergence point sequence generated within the most recent preset time of the artificial intelligence detection body networking includes: Determine a first target and a second target from the decision divergence point sequence respectively; Take the one that is generated first among the first target and the second target as the benchmark divergence point; Among them, the determination steps of the first target are as follows: Take the jth decision divergence point in the decision divergence point sequence as the first target; the sum of the cost values of the first j - 1 decision divergence points in the decision divergence point sequence is closest to the cost value sum threshold; j is a positive integer greater than 1; Among them, the determination steps of the second target are as follows: Take the decision divergence point whose first divergence point type in the decision divergence point sequence is the same as the standard divergence point type as the second target.

[0040] The cost value of the decision divergence point represents the degree of influence of generating the decision divergence point on the working efficiency of the detection body; the cost value sum threshold is preset by the technical personnel; setting the constraint that the sum of the cost values of the first j - 1 decision divergence points in the decision divergence point sequence is closest to the cost value sum threshold can make the value of j unique, and the selected jth decision divergence point as the first target can be used as a candidate for the benchmark of the working rationality analysis.

[0041] The standard divergence point type is the type of the decision divergence point that can be used as a candidate for the benchmark of the working rationality analysis. For example: there is a divergence where it is difficult to choose more than 10 planning tasks.

[0042] To improve the accuracy of selecting the benchmark for the working rationality analysis, take the one that is generated first among the first target and the second target as the benchmark divergence point.

[0043] In one embodiment, the optimization module continuously optimizes and iterates the artificial intelligence detection body network based on the work rationality analysis result, including: Describing the characteristics of the work rationality analysis result to obtain a feature description vector; Determining the optimization iteration strategy corresponding to the feature description vector from the optimization iteration strategy library; Based on the optimization iteration strategy, continuously optimize and iterate the artificial intelligence detection body network.

[0044] The work rationality analysis result is characterized into a feature description vector. There are optimization iteration strategies corresponding to different feature description vectors in the optimization iteration strategy library. The corresponding optimization iteration strategy is determined by querying the library, and the artificial intelligence detection body network is continuously optimized and iterated based on the optimization iteration strategy. The optimization iteration strategy is a strategy indicating how to continuously optimize and iterate the artificial intelligence detection body network. For example, if the work rationality analysis result indicates that a certain detection body's obstacle avoidance accuracy is not suitable for the current complex finished product area site, the corresponding optimization iteration strategy for the feature description vector it is characterized into is to improve the obstacle avoidance accuracy.

[0045] In one embodiment, the finished product detection system of the wind turbine blade further includes: A visualization monitoring module for: Generating a visualization monitoring model of the optimized and iterated artificial intelligence detection body network; Based on the visualization monitoring model, assisting the management personnel to monitor the work of the optimized and iterated artificial intelligence detection body network; The visualization monitoring module generates a visualization monitoring model of the optimized and iterated artificial intelligence detection body network, including: Generating a visualization monitoring model based on the visualization template corresponding to the personnel portrait of the management personnel and according to the multi-modal work information of the optimized and iterated artificial intelligence detection body network.

[0046] When generating the visualization monitoring model, there is a visualization template corresponding to the personnel portrait of the management personnel (the visualization template arranges the content that the personnel portrait of the management personnel reflects as the content that the management personnel need to view first to the front). According to the multi-modal work information of the optimized and iterated artificial intelligence detection body network, a visualization monitoring model is generated, and a targeted visualization monitoring model is generated to improve the efficiency of the management personnel to monitor the work of the optimized and iterated artificial intelligence detection body network based on it.

[0047] In one embodiment, the visualization monitoring module, based on the visualization monitoring model, assists the management personnel to monitor the work of the optimized and iterated artificial intelligence detection body network, including: Detecting multiple work monitoring timing events occurring in the visualization monitoring model; Generate a quick selection table of monitoring tasks based on each work monitoring timing event; Assist the management personnel to quickly select monitoring tasks from the quick selection table of monitoring tasks and perform corresponding executions.

[0048] The work monitoring timing event refers to the timing event that requires the management personnel to intervene in work monitoring. For example, a certain detection body crashes and needs to be maintained. There are multiple monitoring tasks generated according to each work monitoring timing event in the quick selection table of monitoring tasks, which can be quickly selected and executed by the management personnel.

[0049] A finished product detection method for a wind turbine blade provided by an embodiment of the present invention, as Figure 3 shown, includes: S1. Detect multiple wind turbine blades in the finished product area based on the artificial intelligence detection body network; S2. Analyze the working rationality of the artificial intelligence detection body network in real time; S3. Continuously optimize and iterate the artificial intelligence detection body network based on the working rationality analysis result; S4. After each optimization and iteration, based on the optimized and iterated artificial intelligence detection body network, relay to detect each wind turbine blade.

[0050] The real-time analysis of the working rationality of the artificial intelligence detection body network includes: Select a reference divergence point from the sequence of decision divergence points generated by the artificial intelligence detection body network in the most recent preset time; Trace and generate the historical work map of the target detection body corresponding to the reference divergence point within the target time period; wherein, the start and end moments of the target time period are respectively the generation moments of the decision divergence points before and after the reference divergence point in the sequence of decision divergence points; Based on the traversal sequence corresponding to the divergence point type of the reference divergence point, sequentially traverse each map feature in the historical work map; wherein, the map features at least include: the movement change of the minimum bounding box of the target detection body and the work interaction record, and the global change trend of the historical work map; Each time when traversing, when the traversed map feature matches the trigger map feature, stop continuing to traverse the next map feature, and obtain multiple groups of corresponding replay mechanisms and verification rules corresponding to the trigger map feature that matches the traversed map feature; Based on any replay mechanism, control the historical work map to replay, and execute the corresponding verification rules during the replay of the historical work map, and use the execution result of the corresponding verification rule as the working rationality analysis result.

[0051] The selection of the reference divergence point from the sequence of decision divergence points generated by the artificial intelligence detection body network in the most recent preset time includes: Determine a first target and a second target respectively from the decision divergence point sequence; Take the one that is generated first among the first target and the second target as the benchmark divergence point; Among them, the steps for determining the first target are as follows: Take the j-th decision divergence point in the decision divergence point sequence as the first target; the sum of the cost values of the first j-1 decision divergence points in the decision divergence point sequence is closest to the cost value sum and the threshold; Among them, the steps for determining the second target are as follows: Take the decision divergence point whose first divergence point type in the decision divergence point sequence is the same as the standard divergence point type as the second target.

[0052] Based on the work rationality analysis result, the optimization module continuously optimizes and iterates the artificial intelligence detection body networking, including: Perform feature description on the work rationality analysis result to obtain a feature description vector; Determine the optimization iteration strategy corresponding to the feature description vector from the optimization iteration strategy library; Based on the optimization iteration strategy, continuously optimize and iterate the artificial intelligence detection body networking.

[0053] The finished product detection method of the wind turbine blade further includes: Generate a visual monitoring model of the optimized and iterated artificial intelligence detection body networking; Based on the visual monitoring model, assist the management personnel to monitor the work of the optimized and iterated artificial intelligence detection body networking.

[0054] The generation of the visual monitoring model of the optimized and iterated artificial intelligence detection body networking includes: Based on the visual template corresponding to the personnel portrait of the management personnel, generate a visual monitoring model according to the multi-modal work information of the optimized and iterated artificial intelligence detection body networking.

[0055] The assisting the management personnel to monitor the work of the optimized and iterated artificial intelligence detection body networking based on the visual monitoring model includes: Detect multiple work monitoring timing events occurring in the visual monitoring model; Based on each work monitoring timing event, generate a monitoring task quick selection table; Assist the management personnel to quickly select monitoring tasks from the monitoring task quick selection table and perform corresponding executions.

[0056] A computer-readable storage medium provided by an embodiment of the present invention, on which a computer program is stored, and a processor executes the computer program to implement the above method.

[0057] An electronic device provided by an embodiment of the present invention, the electronic device includes a memory and a processor, a computer program is stored in the memory, and the processor executes the computer program to implement the above method.

[0058] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A finished product inspection system for a wind turbine blade, characterized in that, Including: A detection module for detecting multiple wind turbine blades in the finished product area based on an artificial intelligence detection body network; An analysis module for performing real-time working rationality analysis on the artificial intelligence detection body network; An optimization module for continuously optimizing and iterating the artificial intelligence detection body network based on the working rationality analysis results; A relay module for, after each optimization iteration, detecting each wind turbine blade in relay based on the optimized and iterated artificial intelligence detection body network.

2. The finished product detection system for a wind power blade according to claim 1, wherein The analysis module performs real-time working rationality analysis on the artificial intelligence detection body network, including: Selecting a reference divergence point from the sequence of decision divergence points generated by the artificial intelligence detection body network within the most recent preset time; Retrospectively generating a historical working map of the target detection body corresponding to the reference divergence point within the target time period; wherein, the start and end times of the target time period are respectively the generation times of the decision divergence points before and after the reference divergence point in the sequence of decision divergence points; Based on the traversal order corresponding to the divergence point type of the reference divergence point, sequentially traversing each map feature in the historical working map; wherein, the map features at least include: the movement change of the minimum bounding box of the target detection body and the working interaction record, and the global change trend of the historical working map; Each time when traversing, when the traversed map feature matches the trigger map feature, stop continuing to traverse the next map feature, and obtain multiple groups of one-to-one corresponding replay mechanisms and verification rules corresponding to the trigger map feature that matches the traversed map feature; Based on any one of the replay mechanisms, controlling the historical working map to be replayed, and executing the corresponding verification rule during the replay process of the historical working map, and taking the execution result of the corresponding verification rule as the working rationality analysis result.

3. The finished product inspection system for a wind turbine blade according to claim 2, wherein The selecting a reference divergence point from the sequence of decision divergence points generated by the artificial intelligence detection body network within the most recent preset time includes: Respectively determining a first target and a second target from the sequence of decision divergence points; Taking the first-generated one of the first target and the second target as the reference divergence point; Wherein, the determining step of the first target is as follows: Taking the jth decision divergence point in the sequence of decision divergence points as the first target; the sum of the cost values of the first j - 1 decision divergence points in the sequence of decision divergence points is closest to the cost value sum threshold; Wherein, the determining step of the second target is as follows: Taking the decision divergence point whose first divergence point type is the same as the standard divergence point type in the sequence of decision divergence points as the second target.

4. The finished product inspection system for a wind turbine blade according to claim 1, characterized in that, The optimization module continuously optimizes and iterates the artificial intelligence detection body network based on the working rationality analysis results, including: Performing feature description on the working rationality analysis results to obtain a feature description vector; Determining an optimization iteration strategy corresponding to the feature description vector from the optimization iteration strategy library; Based on the optimization iteration strategy, continuously optimizing and iterating the artificial intelligence detection body network.

5. The finished product detection system for a wind power blade according to claim 1, characterized in that, It further includes: A visualization monitoring module for: Generating a visualization monitoring model of the optimized and iterated artificial intelligence detection body network; Based on the visualization monitoring model, assisting management personnel to monitor the work of the optimized and iterated artificial intelligence detection body network.

6. The finished product inspection system for a wind power blade according to claim 5, characterized in that, The visualization monitoring module generating a visualization monitoring model of the optimized and iterated artificial intelligence detection body network includes: Based on the visualization template corresponding to the personnel portrait of the management personnel, a visualization monitoring model is generated according to the multi-modal work information of the optimized and iterated artificial intelligence detection body network.

7. The finished product inspection system for a wind power blade according to claim 5, wherein, The visualization monitoring module, based on the visualization monitoring model, assists the management personnel in monitoring the work of the optimized and iterated artificial intelligence detection body network, including: Detecting multiple work monitoring timing events occurring in the visualization monitoring model; Generating a monitoring task quick selection table based on each work monitoring timing event; Assisting the management personnel in quickly selecting monitoring tasks from the monitoring task quick selection table and performing corresponding executions.

8. A finished product inspection method for a wind power blade, characterized in that, Including: Based on the artificial intelligence detection body network, detecting multiple wind power blades in the finished product area; Performing real-time work rationality analysis on the artificial intelligence detection body network; Based on the work rationality analysis results, continuously optimizing and iterating the artificial intelligence detection body network; After each optimization and iteration, based on the optimized and iterated artificial intelligence detection body network, continuously detecting each wind power blade.

9. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and the processor executes the computer program to implement the method according to claim 8.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor. A computer program is stored in the memory, and the processor executes the computer program to implement the method according to claim 8.

Images