Wind turbine generator intelligent inspection defect identification system based on big data analysis

Abstract

The invention provides an intelligent wind turbine generator inspection defect identification system based on big data analysis, which is characterized by comprising a data acquisition module, an anomaly detection module, an inspection control module and a defect identification module, the anomaly detection module is used for detecting the abnormal state in the collected information, the inspection control module is used for controlling the unmanned aerial vehicle to inspect the wind turbine generator, and the defect identification module identifies the defects of the wind turbine generator based on the inspection information; the system firstly performs big data analysis on the operation data of the wind turbine generator, and assigns the unmanned aerial vehicle to perform targeted inspection based on the analysis result, so that the inspection efficiency can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of electrical digital data processing, and in particular to a wind turbine intelligent inspection defect recognition system based on big data analysis. Background Art

[0002] With the rapid development of the wind power industry, the number and installed capacity of wind turbines continue to increase, and the importance of their operation and maintenance has become increasingly prominent. However, the operating environment of wind turbines is complex. They are exposed to harsh conditions such as high wind speed, low temperature, and high humidity for a long time, which can easily cause problems such as component aging, structural damage, and system failure. If these problems are not discovered and handled in a timely manner, they may cause unplanned equipment shutdowns, reduce power generation efficiency, and even cause safety accidents. Although traditional manual inspection methods can detect unit defects to a certain extent, they are difficult to meet the needs of refined operation and maintenance of modern wind farms due to their high labor costs, low efficiency, and incomplete data coverage.

[0003] Many wind turbine inspection systems have been developed. After a lot of searching and reference, it is found that the existing inspection systems include the system disclosed in the publication number CN111708380B. These system methods generally include: Step 1, the drone determines the flight route of the drone according to the status of the wind turbine of the wind turbine; Step 2, the drone automatically inspects along the flight route and shoots the video stream of the wind turbine of the wind turbine; Step 3, according to the current network connection status, the drone is selected to push the video stream to the platform in real time or the drone pushes the video stream to the platform after the flight is completed. However, this system does not conduct an in-depth analysis of the status of the wind turbine, and only conducts inspections based on simple status information. It still requires a higher inspection frequency, and the inspection efficiency needs to be improved. Summary of the invention

[0004] The purpose of the present invention is to propose a wind turbine intelligent inspection defect identification system based on big data analysis in response to the existing deficiencies.

[0005] The present invention adopts the following technical solution:

[0006] An intelligent inspection and defect recognition system for wind turbines based on big data analysis, comprising a data acquisition module, an anomaly detection module, an inspection control module and a defect recognition module;

[0007] The data acquisition module is used to collect data information of the wind turbine generator set, the abnormality detection module is used to detect abnormal conditions in the collected information, the inspection control module is used to control the drone to inspect the wind turbine generator set, and the defect recognition module recognizes defects in the wind turbine generator set based on the inspection information;

[0008] The data acquisition module includes an operation monitoring unit, a data sorting unit and a data storage unit, wherein the operation monitoring unit is used to monitor the operation data of the wind turbine generator set, the data sorting unit is used to sort the collected monitoring data, and the data storage unit is used to store the sorted data;

[0009] The anomaly detection module includes a big data analysis unit, a real-time proofreading unit and an anomaly output unit. The big data analysis unit is used to perform feature analysis on a large amount of historical data collected. The real-time proofreading unit is used to proofread the real-time monitoring data with the feature data. The anomaly output unit is used to output the anomaly information appearing in the proofreading to the inspection control module.

[0010] The inspection control module includes an abnormality analysis unit, an inspection execution unit and an image feedback unit. The abnormality analysis unit is used to analyze the abnormal information to obtain inspection parameters. The inspection execution unit assigns the UAV to the corresponding area for inspection based on the inspection parameters. The image feedback unit is used to capture the image information of the wind turbine and feed it back to the defect recognition module.

[0011] The defect recognition module includes a defect information storage unit, an image feature extraction unit and a proofreading and recognition unit. The defect information storage unit is used to store defect information of the wind turbine set, the image feature extraction unit is used to extract feature information in the image information, and the proofreading and recognition unit is used to proofread the feature information and the defect information to identify defects in the image.

[0012] Furthermore, the big data analysis unit includes a data statistics processor, a data classification processor and a feature extraction processor. The data statistics processor is used to perform statistical processing on historical data, the data classification processor classifies the data based on the statistical results, and the feature extraction processor extracts feature information based on the classification results.

[0013] Furthermore, the data classification processor calculates the classification judgment index P(i) according to the following formula:

[0014] P(i)=log m ·(1+m·a(i)-1 2 );

[0015] Among them, m is the number of statistical intervals for each item, and a(i) is the probability in the i-th statistical interval;

[0016] The data classification processor classifies the statistical interval whose classification judgment index is greater than the classification threshold as a trace interval, and classifies the statistical interval whose classification judgment index is less than or equal to the classification threshold as a flux interval.

[0017] Furthermore, the real-time proofreading unit includes a data receiving processor, a feature matching processor and a feature proofreading processor, wherein the data receiving processor is used to receive real-time data, the feature matching processor is used to match the real-time data with the feature items, and the feature proofreading processor is used to proofread the real-time data with the matched feature data;

[0018] The feature calibration processor is used to calibrate whether the real-time data is between the first critical point and the second critical point of the corresponding item. If so, the calibration value of the item is set to 1, and if not, the calibration value of the item is set to 0.

[0019] Further, the abnormal output unit includes a proofreading judgment processor, an abnormal summary processor and a packaging output processor, wherein the proofreading judgment processor is used to judge whether the proofreading result is abnormal, the abnormal summary processor is used to summarize the abnormal proofreading information, and the packaging output processor is used to package the summarized abnormal information and output it to the abnormal analysis unit;

[0020] The proofreading judgment processor is provided with items included in all exception types, and when the proofreading values ​​of the included items are all 0, it is judged that an exception of this type exists.

[0021] The beneficial effects achieved by the present invention are:

[0022] This system first performs big data analysis on the operating data of the wind turbine to detect abnormal information, then controls the drone to conduct targeted inspections based on the abnormal information, and then identifies defects in the image information taken during the inspection. This can effectively increase the probability of detecting defects during the inspection process and improve inspection efficiency.

[0023] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0025] Figure 2 This is a schematic diagram of the data acquisition module of the present invention;

[0026] Figure 3 This is a schematic diagram of the abnormality detection module of the present invention;

[0027] Figure 4 This is a schematic diagram of the patrol control module of the present invention;

[0028] Figure 5 This is a schematic diagram of the defect recognition module of the present invention;

[0029] Figure 6 This is a comparison chart of the inspection effect data of the present invention. DETAILED DESCRIPTION

[0030] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual sizes. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

[0031] Embodiment 1.

[0032] This embodiment provides a wind turbine intelligent inspection defect recognition system based on big data analysis, combined with Figure 1 , including data acquisition module, anomaly detection module, inspection control module and defect identification module;

[0033] The data acquisition module is used to collect data information of the wind turbine generator set, the abnormality detection module is used to detect abnormal conditions in the collected information, the inspection control module is used to control the drone to inspect the wind turbine generator set, and the defect recognition module recognizes defects in the wind turbine generator set based on the inspection information;

[0034] The data acquisition module includes an operation monitoring unit, a data sorting unit and a data storage unit, wherein the operation monitoring unit is used to monitor the operation data of the wind turbine generator set, the data sorting unit is used to sort the collected monitoring data, and the data storage unit is used to store the sorted data;

[0035] The anomaly detection module includes a big data analysis unit, a real-time proofreading unit and an anomaly output unit. The big data analysis unit is used to perform feature analysis on a large amount of historical data collected. The real-time proofreading unit is used to proofread the real-time monitoring data with the feature data. The anomaly output unit is used to output the anomaly information appearing in the proofreading to the inspection control module.

[0036] The inspection control module includes an abnormality analysis unit, an inspection execution unit and an image feedback unit. The abnormality analysis unit is used to analyze the abnormal information to obtain inspection parameters. The inspection execution unit assigns the UAV to the corresponding area for inspection based on the inspection parameters. The image feedback unit is used to capture the image information of the wind turbine and feed it back to the defect recognition module.

[0037] The defect recognition module includes a defect information storage unit, an image feature extraction unit and a proofreading and recognition unit. The defect information storage unit is used to store defect information of the wind turbine set, the image feature extraction unit is used to extract feature information in the image information, and the proofreading and recognition unit is used to proofread the feature information and the defect information to identify defects in the image.

[0038] The big data analysis unit includes a data statistics processor, a data classification processor and a feature extraction processor. The data statistics processor is used to perform statistical processing on historical data, the data classification processor performs classification processing on data based on statistical results, and the feature extraction processor extracts feature information based on the classification results.

[0039] The data classification processor calculates the classification judgment index P(i) according to the following formula:

[0040] P(i)=log m ·(1+m·a(i)-1 2 );

[0041] Among them, m is the number of statistical intervals for each item, and a(i) is the probability in the i-th statistical interval;

[0042] The data classification processor classifies the statistical interval whose classification judgment index is greater than the classification threshold as a trace interval, and classifies the statistical interval whose classification judgment index is less than or equal to the classification threshold as a flux interval.

[0043] The real-time proofreading unit includes a data receiving processor, a feature matching processor and a feature proofreading processor, wherein the data receiving processor is used to receive real-time data, the feature matching processor is used to match the real-time data with feature items, and the feature proofreading processor is used to proofread the real-time data with the matched feature data;

[0044] The feature calibration processor is used to calibrate whether the real-time data is between the first critical point and the second critical point of the corresponding item. If so, the calibration value of the item is set to 1, and if not, the calibration value of the item is set to 0.

[0045] The abnormal output unit includes a proofreading judgment processor, an abnormal summary processor and a packaging output processor, wherein the proofreading judgment processor is used to judge whether the proofreading result is abnormal, the abnormal summary processor is used to summarize the abnormal proofreading information, and the packaging output processor is used to package the summarized abnormal information and output it to the abnormal analysis unit;

[0046] The proofreading judgment processor is provided with items included in all exception types, and when the proofreading values ​​of the included items are all 0, it is judged that an exception of this type exists.

[0047] Embodiment 2.

[0048] This embodiment includes all the contents of the first embodiment, and provides a wind turbine intelligent inspection defect identification system based on big data analysis, including a data acquisition module, an abnormality detection module, an inspection control module and a defect identification module;

[0049] The data acquisition module is used to collect data information of the wind turbine generator set, the abnormality detection module is used to detect abnormal conditions in the collected information, the inspection control module is used to control the drone to inspect the wind turbine generator set, and the defect recognition module recognizes defects in the wind turbine generator set based on the inspection information;

[0050] Combination Figure 2 The data acquisition module includes an operation monitoring unit, a data sorting unit and a data storage unit. The operation monitoring unit is used to monitor the operation data of the wind turbine generator set, the data sorting unit is used to sort the collected monitoring data, and the data storage unit is used to store the sorted data;

[0051] Combination Figure 3 The anomaly detection module includes a big data analysis unit, a real-time proofreading unit and an anomaly output unit. The big data analysis unit is used to perform feature analysis on a large amount of historical data collected. The real-time proofreading unit is used to proofread the real-time monitoring data with the feature data. The anomaly output unit is used to output the anomaly information that appears in the proofreading to the inspection control module.

[0052] Combination Figure 4 The inspection control module includes an abnormality analysis unit, an inspection execution unit and an image feedback unit. The abnormality analysis unit is used to analyze the abnormal information to obtain inspection parameters. The inspection execution unit assigns the UAV to the corresponding area for inspection based on the inspection parameters. The image feedback unit is used to capture the image information of the wind turbine and feed it back to the defect recognition module.

[0053] Combination Figure 5 The defect recognition module includes a defect information storage unit, an image feature extraction unit and a proofreading and recognition unit. The defect information storage unit is used to store defect information of the wind turbine generator set, the image feature extraction unit is used to extract feature information from the image information, and the proofreading and recognition unit is used to proofread the feature information and the defect information to identify defects in the image;

[0054] The operation monitoring unit includes a sensor detection processor, a signal conversion processor and a data reporting processor, wherein the sensor detection processor is used to detect the operation signal of the wind turbine generator set, the signal conversion processor is used to convert the collected signal into data information, and the data reporting processor is used to report the data information to the data sorting unit;

[0055] The data sorting unit includes a device information memory, a device mapping processor and a data classification processor, wherein the device information memory is used to store the address information of the monitored device, the device mapping processor is used to map the received reported data with the address, and the data classification processor classifies the reported data based on the mapping relationship;

[0056] The data storage unit includes a historical data register, a real-time data register and a data conversion processor, wherein the historical data register is used to store historical data, the real-time data register is used to store real-time data, and the data transfer processor is used to transfer the data in the real-time data register to the historical data register;

[0057] The big data analysis unit includes a data statistics processor, a data classification processor and a feature extraction processor. The data statistics processor is used to perform statistical processing on historical data. The data classification processor performs classification processing on data based on statistical results. The feature extraction processor extracts feature information based on the classification results.

[0058] The data statistics processor sets a statistical interval for each item, and the data statistics processor counts the number of times falling in each statistical interval and converts it into a percentage to obtain the probability information of each item in each statistical interval;

[0059] The data classification processor calculates the classification judgment index P(i) according to the following formula:

[0060] P(i)=log m ·(1+m·a(i)-1 2 );

[0061] Among them, m is the number of statistical intervals for each item, and a(i) is the probability in the i-th statistical interval;

[0062] The data classification processor classifies the statistical interval whose classification judgment index is greater than the classification threshold as a trace interval, and classifies the statistical interval whose classification judgment index is less than or equal to the classification threshold as a flux interval;

[0063] The feature extraction processor uses the minimum value and the maximum value in the flux interval as feature information of the corresponding item, which are called the first critical point and the second critical point respectively;

[0064] The real-time proofreading unit includes a data receiving processor, a feature matching processor and a feature proofreading processor, wherein the data receiving processor is used to receive real-time data, the feature matching processor is used to match the real-time data with feature items, and the feature proofreading processor is used to proofread the real-time data with the matched feature data;

[0065] The feature check processor is used to check whether the real-time data is between the first critical point and the second critical point of the corresponding item. If so, the check value of the item is set to 1, and if not, the check value of the item is set to 0;

[0066] The abnormal output unit includes a proofreading judgment processor, an abnormal summary processor and a packaging output processor, wherein the proofreading judgment processor is used to judge whether the proofreading result is abnormal, the abnormal summary processor is used to summarize the abnormal proofreading information, and the packaging output processor is used to package the summarized abnormal information and output it to the abnormal analysis unit;

[0067] The proofreading judgment processor is provided with items included in all exception types. When the proofreading values ​​of the included items are all 0, it is judged that an exception of this type exists.

[0068] The abnormality analysis unit includes a main body positioning processor, an abnormality positioning processor and a parameter formulation processor, wherein the main body positioning processor is used to locate the position information of the abnormal wind turbine group, the abnormality positioning processor is used to locate the position information of the abnormal component in the wind turbine group, and the parameter formulation processor formulates the inspection parameters of the drone based on the positioning information;

[0069] The inspection execution unit includes a flight control processor, a safety adjustment processor and a shooting activation processor. The flight control processor controls the flight path of the UAV based on the village construction parameters. The safety adjustment processor is used to adjust the route to avoid obstacles during flight. The shooting activation processor is used to send a shooting signal.

[0070] The image feedback unit includes a signal receiving processor, an image shooting processor and an information feedback processor, wherein the signal receiving processor is used to receive a shooting signal, the image shooting processor is used to shoot the wind turbine when receiving the shooting signal, and the information feedback processor is used to perform feedback processing on the shot image;

[0071] The defect information storage unit includes a defect classification manager, a defect feature register and a defect retrieval processor, wherein the defect classification manager is used to classify and manage defects, the defect feature register is used to store feature information of each defect type, and the defect retrieval processor is used to retrieve defect types;

[0072] The image feature extraction unit includes a pixel statistics processor, a feature rule register and a rule execution processor, wherein the pixel statistics processor is used to count pixel information in the image, the feature rule register is used to store rule information for calculating features, and the rule execution processor extracts feature data from the image based on the rule information;

[0073] The proofreading and identifying unit includes a feature comparison processor and a defect recognition processor, wherein the feature comparison processor is used to compare feature information, and the defect recognition processor identifies corresponding defects based on the comparison result;

[0074] The feature comparison processor performs comparison processing on the feature data according to the following formula:

[0075] ΔV=|V 1 -V 0 |;

[0076] Among them, V 1 is the extracted eigenvalue, V 0 is the standard value in the characteristic information, and ΔV represents the comparison result;

[0077] The defect recognition processor calculates the defect judgment value Q according to the following formula:

[0078]

[0079] Among them, n is the number of features contained in the corresponding defect type, k i represents the weight coefficient of the i-th feature, ΔV i Represents the comparison result of the i-th feature;

[0080] When Q is greater than the defect threshold, it indicates that the corresponding defect exists;

[0081] The i's that appear in the above text are ordinal numbers used to indicate serial numbers and have no actual meaning.

[0082] Part of the code of this system is described as follows:

[0083]

[0084]

[0085]

[0086]

[0087] Now, this system compares and counts the number of defects identified in each inspection and the number of anomalies found in big data analysis, and obtains Figure 6 The effect diagram shown.

[0088] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. A wind turbine intelligent inspection defect recognition system based on big data analysis, characterized in that: It includes data acquisition module, anomaly detection module, inspection control module and defect identification module; The data acquisition module is used to collect data information of the wind turbine generator set, the abnormality detection module is used to detect abnormal conditions in the collected information, the inspection control module is used to control the drone to inspect the wind turbine generator set, and the defect recognition module recognizes defects in the wind turbine generator set based on the inspection information; The data acquisition module includes an operation monitoring unit, a data sorting unit and a data storage unit, wherein the operation monitoring unit is used to monitor the operation data of the wind turbine generator set, the data sorting unit is used to sort the collected monitoring data, and the data storage unit is used to store the sorted data; The anomaly detection module includes a big data analysis unit, a real-time proofreading unit and an anomaly output unit. The big data analysis unit is used to perform feature analysis on a large amount of historical data collected. The real-time proofreading unit is used to proofread the real-time monitoring data with the feature data. The anomaly output unit is used to output the anomaly information appearing in the proofreading to the inspection control module. The inspection control module includes an abnormality analysis unit, an inspection execution unit and an image feedback unit. The abnormality analysis unit is used to analyze the abnormal information to obtain inspection parameters. The inspection execution unit assigns the UAV to the corresponding area for inspection based on the inspection parameters. The image feedback unit is used to capture the image information of the wind turbine and feed it back to the defect recognition module. The defect recognition module includes a defect information storage unit, an image feature extraction unit and a proofreading and recognition unit. The defect information storage unit is used to store defect information of the wind turbine set, the image feature extraction unit is used to extract feature information in the image information, and the proofreading and recognition unit is used to proofread the feature information and the defect information to identify defects in the image.

2. The wind turbine intelligent inspection defect identification system based on big data analysis according to claim 1, characterized in that: The big data analysis unit includes a data statistics processor, a data classification processor and a feature extraction processor. The data statistics processor is used to perform statistical processing on historical data, the data classification processor performs classification processing on data based on statistical results, and the feature extraction processor extracts feature information based on the classification results.

3. The wind turbine intelligent inspection defect identification system based on big data analysis as claimed in claim 2, characterized in that: The data classification processor calculates the classification judgment index P(i) according to the following formula: P(i)=log m ·(1+m·a(i)-1 2 ); Among them, m is the number of statistical intervals for each item, and a(i) is the probability in the i-th statistical interval; The data classification processor classifies the statistical interval whose classification judgment index is greater than the classification threshold as a trace interval, and classifies the statistical interval whose classification judgment index is less than or equal to the classification threshold as a flux interval.

4. The wind turbine intelligent inspection defect identification system based on big data analysis as claimed in claim 3, characterized in that: The real-time proofreading unit includes a data receiving processor, a feature matching processor and a feature proofreading processor, wherein the data receiving processor is used to receive real-time data, the feature matching processor is used to match the real-time data with feature items, and the feature proofreading processor is used to proofread the real-time data with the matched feature data; The feature calibration processor is used to calibrate whether the real-time data is between the first critical point and the second critical point of the corresponding item. If so, the calibration value of the item is set to 1, and if not, the calibration value of the item is set to 0.

5. The wind turbine intelligent inspection defect identification system based on big data analysis as claimed in claim 4, characterized in that: The abnormal output unit includes a proofreading judgment processor, an abnormal summary processor and a packaging output processor, wherein the proofreading judgment processor is used to judge whether the proofreading result is abnormal, the abnormal summary processor is used to summarize the abnormal proofreading information, and the packaging output processor is used to package the summarized abnormal information and output it to the abnormal analysis unit; The proofreading judgment processor is provided with items included in all exception types, and when the proofreading values ​​of the included items are all 0, it is judged that an exception of this type exists.

Images