CN112801432B - Intelligent inspection system for fan set blades and inspection method for fan set blades - Google Patents

Abstract

The invention discloses an intelligent inspection system for fan blades and a related inspection method for fan blades. The intelligent inspection system and the inspection method for the blades of the fan set can effectively conduct unified planning on the blade inspection tasks of a plurality of wind fields, achieve blade inspection information management, automatic blade image analysis and intelligent inspection of the blades of the fan, greatly improve the utilization rate of personnel and equipment, reduce the technical threshold of the personnel for inspecting the blades of the fan set, achieve automation of inspection of the blades of the fan set, data analysis and intellectualization, and information management digitization, and conduct health information tracking of the whole life cycle of the blades of the fan.

Description

Intelligent inspection system for fan set blades and inspection method for fan set blades

Technical Field

The invention relates to the field of blade inspection and maintenance of wind generating sets, in particular to an intelligent inspection system and an inspection method for fan set blades.

Background

The blade is an important component of a wind generating set (hereinafter referred to as a fan set for short), and because the fan is in a severe environment, the blade is damaged by natural factors such as sand, rain, snow, lightning and the like when running in a severe environment, so that defects such as surface falling, sand holes, lightning strokes, blade edge abrasion and the like are formed, and the blade needs to be periodically checked and maintained to prevent accidents caused by the defects. Traditional blade inspection adopts manual mode more, and operation intensity is big, shut down cycle is long and accompanied personnel safety risk.

With the development of technology, unmanned aerial vehicles are increasingly widely applied to inspection work. The current common inspection mode in the industry is that an operator observes the surface state of a blade through a real-time image displayed by a ground station, and when a suspicious point is found, the unmanned aerial vehicle is remotely operated to acquire images at various angles so as to further perform detailed inspection. The blade inspection mode has certain limitations when solving the inspection efficiency and personnel safety problems:

1. Wind fields are distributed in a scattered manner, and the traditional blade inspection work lacks uniform task management, so that the inspection work has high transition pressure and low efficiency; meanwhile, the inspection result is usually kept in the form of a paper report, which is not beneficial to retrospective analysis of the leaf problems in the later stage.

2. The blade inspection needs to be carried out by special operation and maintenance personnel, and the operation threshold of the unmanned aerial vehicle is relatively high, so that a batch of operation and maintenance personnel which can grasp the knowledge of the blade defects and have the operation capability of the unmanned aerial vehicle needs to be cultivated, and the labor cost is increased intangibly.

3. The unmanned aerial vehicle is used for checking the blades of the fan, so that the number of the photos of the blades is greatly increased, but the current general inspection mode is that a blade expert performs manual screening on the image data of the blades, so that the workload is high and the efficiency is low; meanwhile, an effective image data management mechanism is not provided, so that the blade images obtained by inspection are difficult to store in different categories, and the blade images become a stopper for establishing a blade life cycle health management system. Therefore, intelligent blade inspection systems have become a new trend.

Disclosure of Invention

Aiming at the defects of the prior art, the invention provides an intelligent inspection system for fan set blades and an inspection method for fan set blades.

The invention provides an intelligent inspection system for fan set blades, which comprises the following components: a task management system, a task processing system, a data analysis system and a data acquisition system,

Wherein,

The task management system is in signal connection with the task processing system and is used for carrying out unified planning management on the inspection task and issuing the inspection task to the task processing system;

The task processing system is in signal connection with the data analysis system and the data acquisition system, and is used for distributing the inspection task to the data acquisition system and receiving, managing and auditing the analysis result of the data analysis system;

the data acquisition system is used for arranging and executing blade inspection work according to the inspection task and uploading stored blade image data to the data analysis system;

The data analysis system is in signal connection with the data acquisition system and is used for automatically analyzing according to the inspection image data of the data acquisition system and uploading analysis results;

the number of the task processing systems is equal to or greater than 1, and each task processing system is in signal connection with the corresponding data analysis system and data acquisition system.

Further, the task management system comprises a wind field information management module, a patrol task management module and a patrol report management module,

Wherein,

The wind field information management module is used for creating or importing wind field information, wherein the wind field information comprises a wind field code, a wind field name, a wind field geographic position, a unit model and blade information;

the inspection task management module is used for issuing all the inspection tasks and auditing the inspection task information reported by all the wind farms according to the unified planning of the area, and checking the in-process inspection task and historical task information;

The patrol report management module is used for checking the detail of the patrol report uploaded by the task processing system, auditing and uniformly managing the patrol report;

The wind field information management module comprises a manual wind field information introduction functional unit, a template wind field information introduction functional unit and a wind field information editing functional unit, wherein the manual wind field information introduction functional unit, the template wind field information introduction functional unit and the wind field information editing functional unit are in signal connection with the inspection task management module;

The manual wind field information importing functional unit is provided with a wind field creating interface, and new wind field information is manually created by manually inputting corresponding information into the wind field creating interface;

the template importing wind field information function unit is used for filling in template data, selecting importing template information and creating wind field information;

The wind field information editing functional unit is used for inputting and modifying the wind field information;

The inspection task management module comprises an inspection task issuing functional unit, an inspection task auditing functional unit and an inspection task list functional unit;

the patrol task issuing functional unit and the patrol task auditing functional unit are both in signal connection with the patrol task list functional unit and the task processing system;

the patrol task list functional unit is in signal connection with the patrol report management module;

The inspection task issuing functional unit is used for an administrator to issue the inspection task aiming at a certain wind field;

The inspection task auditing function unit is used for auditing the inspection task submitted by the wind farm by a manager;

the patrol task list functional unit is used for displaying all patrol plans;

and the patrol report management module is in signal connection with the task processing system.

Further, the task processing system comprises a patrol plan management module, a patrol workbench module and a patrol information summarization module,

Wherein,

The inspection plan management module is in signal connection with the inspection workbench module, the task management system and the data acquisition system;

the inspection workbench module is in signal connection with the inspection information summarizing module, the data acquisition system and the data analysis system;

The patrol information summarizing module is in signal connection with the task management system;

the inspection plan management module is used for making and filling an inspection plan according to the received inspection task and distributing the inspection plan to each data acquisition system;

the inspection workbench module is used for inputting the inspection information of the blade, has temporary storage and preview functions, and inquires, modifies and generates an inspection report;

the inspection information summarizing module is used for checking and managing the inspection reports of all fans in the wind field range uploaded by the inspection workbench module;

The inspection plan management module comprises an inspection calendar functional unit, an inspection plan editing functional unit and an inspection plan inquiring functional unit;

the inspection calendar functional unit and the inspection plan editing functional unit are connected with the inspection plan inquiring functional unit through signals;

the inspection calendar functional unit is used for displaying an inspection plan of the latest date;

The inspection plan editing functional unit is used for filling an inspection plan, and submitting the inspection plan to archiving after filling;

the inspection plan inquiring functional unit is used for receiving and managing the inspection plan information uploaded by the inspection calendar functional unit and the inspection plan editing functional unit;

the inspection workbench module comprises an inspection site information input functional unit and a query, modification and report generation functional unit;

The inspection site information input functional unit is used for inputting the blade inspection information uploaded by the data acquisition system;

The receiving and managing inspection calendar function unit is used for enabling a manager to inspect inspection analysis results uploaded by the data analysis system and generating the inspection report.

Further, the data analysis system comprises a data import/export module A, an image preprocessing module, a defect marking module, a defect classification module, a defect diagnosis module and a data import/export module B;

Wherein,

The data importing/exporting module A, the image preprocessing module, the defect marking module, the defect classifying module, the defect diagnosing module and the data importing/exporting module B are connected in sequence through signals;

The data import/export module A is in signal connection with the task processing system and the data acquisition system;

The data import/export module B is in signal connection with the task processing system;

the data import/export module A is used for uploading the image data of the data acquisition system;

the image preprocessing module is used for preprocessing the image data and determining a monitored area;

The defect marking module is used for marking the defects of the image;

the defect classification module is used for classifying the marked defects;

The defect diagnosis module is used for carrying out grading diagnosis on the identified defects;

The data import/export module B is used for uploading the analysis result of the completed image to the task processing system.

Further, the data analysis system is provided with a patrol end and a data center end,

Wherein,

The inspection end is operated on each inspection device and is used for analyzing the image data and giving out a preliminary analysis result including whether the inspected blade has defects and the defect positions;

the data center end is used for reading the blade image obtained by inspection uploaded by the inspection end, carrying out recognition analysis, and giving out detailed analysis results including whether the inspected blade has defects and defect positions, defect types and defect classification.

Further, the data acquisition system comprises a fan information acquisition module, a ground control module and an intelligent inspection unmanned aerial vehicle,

Wherein,

The fan information acquisition module is used for acquiring the direction angle and the phase angle of the fan impeller;

the ground control module is used for automatically generating a routing inspection route and monitoring a flight process of the unmanned aerial vehicle;

the fan information acquisition module comprises a handheld direction finding unit and an angle calculating unit, and the handheld direction finding unit is in signal connection with the angle calculating unit;

The handheld direction-finding unit is used for selecting each point of the blade tips of two blades on the tested fan, and measuring to obtain the geomagnetic angle of the connecting line between the current observation point and the tested point of the blade tips;

the angle calculating unit is used for calculating the impeller orientation angle through the geomagnetic angle;

The ground control module comprises an automatic route generating unit and a flight process monitoring unit;

The automatic route generating unit is in signal connection with the angle calculating unit;

The automatic route generating unit is used for directly generating a routing inspection route according to the direction of the impeller of the tested fan, the phase of the impeller, the height of the hub, the length of the blade and the coordinate information of the fan, and transmitting the routing inspection route to the intelligent routing inspection unmanned aerial vehicle for automatic routing inspection;

the flight process monitoring unit is used for receiving the shooting picture of the unmanned aerial vehicle in real time and monitoring in real time;

the intelligent inspection unmanned aerial vehicle comprises an unmanned aerial vehicle flight platform, a positioning unit, an image acquisition unit, a data storage unit, an information matching unit and an obstacle avoidance unit;

the unmanned aerial vehicle flight platform is in signal connection with the route automatic generation unit, the positioning unit, the image acquisition unit and the data storage unit;

the positioning unit is in signal connection with the image acquisition unit, the data storage unit and the information matching unit;

The image acquisition unit is in signal connection with the data storage unit and the information matching unit;

The information matching unit is in signal connection with the data storage unit;

The obstacle avoidance unit is in signal connection with the unmanned aerial vehicle flight platform;

the unmanned aerial vehicle flight platform is used for carrying all the function inspection function modules;

the positioning unit comprises an airborne positioning unit and a ground base station and is used for positioning the unmanned aerial vehicle and transmitting position information obtained by positioning to the information matching unit;

the image acquisition unit comprises a high-power camera, a self-stabilizing cradle head and a visual guiding unit, wherein the high-power camera is used for taking the pictures of the blades; the self-stabilizing cradle head is used for connecting the high-power camera signal to the unmanned aerial vehicle and counteracting high-power camera shake caused by the flight of the unmanned aerial vehicle; the vision guiding unit is used for assisting the high-power camera to aim at the blade in real time, and guaranteeing that the blade is positioned in the center of a view finding frame of the high-power camera;

The data storage unit is used for receiving and storing the image data of the information matching unit, and transmitting the image data to the data analysis system after the unmanned aerial vehicle finishes flying;

The information matching unit is used for reading and matching the picture information acquired by the image acquisition unit and the position information of the shooting point where the unmanned aerial vehicle is located so as to locate a defect position and transmit the defect position to the data storage unit;

The obstacle avoidance unit is used for scanning the surrounding environment of the unmanned aerial vehicle, and when an obstacle appears in a set distance, the obstacle avoidance unit gives an alarm, and the unmanned aerial vehicle suspends a route and hovers;

The inspection route sequentially traverses the two sides of each blade of the fan to be inspected.

The invention also provides a fan set blade inspection method based on the intelligent fan set blade inspection system, which comprises the following steps:

Step 1, creating a wind field: creating wind field information in the task management system, filling in wind field codes, wind field names, wind field geographic positions, unit models and blade information, and synchronizing the wind field codes, the wind field names, the wind field geographic positions, the unit models and the blade information to the task processing system;

Step 2, creating a task: generating a new inspection task in the task management system, and issuing the inspection task to the task processing system of a corresponding wind field or auditing a task application submitted by the task processing system, and allowing execution;

Step3, task planning: the task processing system receives the inspection task, makes an inspection plan, and plans the inspection time, the inspection personnel, the inspection equipment number and the construction period of each fan in a field;

Step 4, executing the job: the data acquisition system executes blade inspection work according to the task plan;

Step5, data analysis: the data analysis system analyzes the inspection data uploaded by the intelligent inspection unmanned aerial vehicle and uploads the result to the task processing system;

step 6, completing the task: the task processing system examines the data analysis result, records the execution condition of the inspection operation, generates an inspection report and uploads the inspection report to the task management system;

Step 7, task closing: and the task management system examines the task processing system to upload the inspection report, and the task is closed after no objection is confirmed.

Further, in the step 5, the data analysis system performs the steps of:

Step 1a0, reading image data: uploading fan blade image data shot by the unmanned aerial vehicle to a data analysis system;

Step 2a0, image preprocessing: removing the background of the fan blade image, and removing the influence of illumination on the fan blade image;

step 3a0, defect identification: screening the fan blade images, removing the non-defective blade images, and marking the defects of the blades by utilizing boxes with different colors;

step 4a0, synchronizing the data analysis results obtained through the steps 1a0-3a0 to the task processing system, giving a patrol report,

Or (b)

The following steps are performed:

step 1a, reading image data: uploading fan blade image data shot by the unmanned aerial vehicle to a data analysis system;

Step 2a, image preprocessing: removing the background of the fan blade image, and removing the influence of illumination on the fan blade image;

Step 3a, defect identification: screening the fan blade images, removing the non-defective blade images, and marking the defects of the blades by utilizing boxes with different colors;

step 4a, defect classification: extracting and identifying image features of the target area obtained by marking, and determining the category of the defect;

Step 5a, defect diagnosis: performing grading diagnosis on the identified defects to give the severity of the defects;

Step 6a, data uploading: and (3) synchronizing the data analysis results obtained in the steps 1a-5a to the task processing system for auditing, and ending the data analysis process.

Further, in the step 4, the following steps are performed:

step 1b, determining fan information: acquiring fan impeller orientation, impeller phase angle, hub height, blade length and fan coordinate information;

step 2b, generating an air route: inputting fan information into the ground control module, automatically generating route information data, and transmitting the route information data to the intelligent inspection unmanned aerial vehicle;

Step 3b, blade inspection: the unmanned aerial vehicle automatically flies along the route, and images of the blades are shot and stored;

step 4b, task end: the blade image data is imported into the data analysis system.

Further, the step 3b includes:

Matching the stored image data with GPS position information of the shooting points one by one;

In the shooting process, a high-power camera in the intelligent inspection unmanned aerial vehicle is aligned to the blade in real time by utilizing a vision auxiliary lens, so that the blade is ensured to be positioned in the center of a view finding frame of the high-power camera;

The obstacle avoidance unit on the unmanned aerial vehicle emits laser in real time to scan the surrounding environment of the unmanned aerial vehicle, so that safety is ensured.

Further, the step 1b further includes:

step 1c, turning a fan impeller, wherein the impeller is locked after rotating to a Y-shaped position;

Step 2c, measuring a measured point a of the tip of one blade above the impeller by using a direction-finding device, and recording a geomagnetic angle aa and a distance La of a connecting line between the current position of the direction-finding device and the measured point a;

step 3c, keeping the current position of the direction-finding device unchanged, measuring a measured point b of the tip of the other blade above the impeller, and recording a geomagnetic angle bb and a distance Lb of a connecting line between the current position and the measured point b;

step 4c, calculating the orientation angle of the impeller according to the geomagnetic angle 1, the geomagnetic angle 2, the distance La and the distance Lb,

Or (b)

The step 1b further includes:

step 1d, locking the impeller after the impeller of the fan rotates to a Y-shaped position;

Step 2d, vertically projecting the line between the observation point and a measured point a1 of a blade tip above the impeller to the ground to obtain a direction b1, and obtaining an included angle between a fixed direction and the direction b1 to obtain a geomagnetic angle c1 and a projection line length L1;

Step 3d, vertically projecting the line between the observation point and a measured point a2 of the tip of the other blade above the impeller to the ground to obtain a direction b2, and obtaining an included angle between the fixed direction and the direction b2 to obtain a geomagnetic angle c2 and a projection line length L2;

And 4d, calculating an included angle c3 between the fixed direction and the direction 3 through the geomagnetic angle c1, the geomagnetic angle c2, the projection line length L1 and the projection line length L2, and subtracting 90 degrees from the included angle c3 to obtain the orientation angle of the impeller, wherein the direction 3 is the direction of the ground projection connecting line of the measured points a1 and a 2.

According to the intelligent inspection system and the inspection method for the fan blades, provided by the invention, the factors of task scheduling, plan execution, automatic inspection, intelligent analysis, report inspection and the like in the inspection process of the fan blades are comprehensively considered, the inspection task arrangement of each wind field is uniformly planned, the wind field can submit task applications according to the self needs, the inspection results are inspected layer by layer, and the error probability is reduced; the automation and intelligent degree of the inspection process and the data processing are improved, the inspection efficiency of the blade is improved, and the inspection cost of the unmanned aerial vehicle is reduced. Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 shows a schematic diagram of a fan blade intelligent inspection system in accordance with an embodiment of the present invention;

FIG. 2 shows a schematic diagram of a task management system in a fan blade intelligent patrol system according to an embodiment of the present invention;

FIG. 3 shows a schematic diagram of a task processing system in a fan blade intelligent patrol system according to an embodiment of the present invention;

FIG. 4 shows a schematic diagram of a data analysis system in a fan blade intelligent inspection system according to an embodiment of the present invention;

FIG. 5 shows a schematic diagram of a data acquisition system in a fan blade intelligent inspection system according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a method for acquiring fan orientation information according to an embodiment of the present invention;

fig. 7 is a schematic diagram of a method for calculating fan orientation information according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Fig. 1 shows a structure of an intelligent inspection system for fan blades according to the present invention, and as can be seen from fig. 1, the intelligent inspection system for fan blades according to the present invention includes a task management system 1, a task processing system 2, a data analysis system 3, and a data acquisition system 4.

Wherein,

The number of the task processing systems 2 is equal to or greater than 1, and each task processing system 2 is in signal connection with a corresponding data analysis system 3 and a corresponding data acquisition system 4;

The task management system 1 is in signal connection with the task processing system 2, and the task management system 1 performs unified planning management on the inspection tasks and issues specific inspection tasks to the task processing system 2;

The task processing system 2 is simultaneously connected with the data analysis system 3 and the data acquisition system 4 in a signal manner, distributes the inspection task to the data acquisition system 4 and receives, manages and audits the analysis result of the data analysis system 3;

The data acquisition system 4 is used for executing blade inspection work according to the inspection task arrangement and uploading the stored blade image data to the data analysis system 3;

The data analysis system 3 is in signal connection with the data acquisition system 4, and automatically analyzes according to the inspection image data of the data acquisition system 4 and uploads the analysis result.

As shown in fig. 2, the task management system 1 includes a wind farm information management module 11, a patrol task management module 12, and a patrol report management module 13.

Wherein,

The wind field information management module 11 is used for creating and managing wind field information, and sharing the wind field information with the inspection task management module 12, wherein the wind field information comprises wind field codes, wind field names, wind field geographic positions, unit models, blade information and the like, the wind field information management module 11 comprises a manual wind field information introduction functional unit 111, a template wind field information introduction functional unit 112 and a wind field information editing functional unit 113, and all three units are in signal connection with the inspection task management module 12;

the manual wind field information importing function unit 111 is provided with a wind field creating interface, and by manually inputting corresponding information at the wind field creating interface, new wind field information is manually created successfully and a prompt is given;

A template importing wind field information function unit 112 for filling in template data, selecting importing template information, and creating wind field information after the unit 112 displays that importing is successful;

The wind field information editing function unit 113 is provided with a wind field information list, and can enter a modification interface by selecting a wind field code in the wind field information list, input modification information, and edit the wind field information.

The inspection task management module 12 is configured to issue inspection tasks of each wind farm and audit the inspection task information reported by each wind farm according to a unified planning of an area, and view the in-progress inspection task and history task information, where the inspection task management module 12 includes an inspection task issuing function unit 121, an inspection task audit function unit 122, and an inspection task list function unit 123;

The patrol task issuing functional unit 121 and the patrol task auditing functional unit 122 are both in signal connection with the patrol task list functional unit 123 and the task processing system 2;

the patrol task list functional unit 123 is in signal connection with the patrol report management module 13;

a patrol task issuing function unit 121, configured to issue a patrol task for a certain wind farm by a manager, where the task is to be pushed to the wind farm, and the wind farm performs patrol according to details of the task, and uploads a patrol report;

The inspection task auditing function unit 122 is used for auditing the inspection task submitted by the wind farm by the manager, if the inspection task is audited, the wind farm personnel can inspect the wind farm, and meanwhile, the data of the wind farm is locked for inspection, and any data of the wind farm is forbidden to be modified;

The patrol task list function unit 123, said unit 123 is provided with a patrol task list page, said page will display all patrol plans, including historical patrol plans, the latest patrol plan will be ranked in front, the patrol plan to be audited will be ranked in front.

The inspection report management module 13 is configured to check the inspection report details uploaded by the task processing system 2, audit and uniformly manage the reports, where the inspection report management module 13 displays all inspection reports, including historical inspection reports, and the latest inspection report is ranked in front and the inspection report to be audited is ranked in front in the report list of the module 13.

As shown in fig. 3, the task processing system 2 includes a patrol plan management module 21, a patrol table module 22, and a patrol information summary module 23.

Wherein,

The inspection plan management module 21 is in signal connection with the inspection workbench module 22, the task management system 1 and the data acquisition system 4;

the inspection workbench module 22 is in signal connection with the inspection information summarizing module 23, the data acquisition system 4 and the data analysis system 3;

the patrol information summary module 23 is in signal connection with the task management system 1.

The inspection plan management module 21 is configured to make an inspection plan according to the received inspection task and allocate the inspection plan to each data acquisition system 4, where the module 21 includes an inspection calendar function unit 211, an inspection plan editing function unit 212, and an inspection plan query function unit 213;

The inspection calendar function unit 211 and the inspection plan editing function unit 212 are all in signal connection with the inspection plan inquiry function unit 213;

The inspection calendar function unit 211 is used for displaying the inspection plan of the latest date, the interface of the inspection calendar function unit 211 displays the inspection plan of the latest date, the inspection plan within 30 days is specially marked by adopting different colors, and the system and the message reminding are given;

The inspection plan editing function unit 212 is used for filling in an inspection plan, submitting and archiving after filling in, wherein the inspection plan editing function unit 212 selects an in-field fan number, fills in the inspection plan, and comprises inspection time, inspection personnel, inspection equipment number, planning period and the like, and submits and archiving after filling in;

The inspection plan query function unit 213 receives and manages the inspection plan information uploaded by the inspection calendar function unit 211 and the inspection plan editing function unit 212, and can query the inspection plan according to conditions, and can select the inspection plan in the query result to modify and delete.

The inspection workbench module 22 is used for inputting inspection information of the blade, has temporary storage and preview functions, inquires, modifies and generates an inspection report, and the inspection workbench module 22 comprises an inspection site information input function unit 221 and an inquiry, modification and generation report function unit 222;

The inspection site information input function unit 221 is used for inputting the blade inspection information uploaded by the data acquisition system 4, and providing temporary storage and preview functions so as to check results in work;

the receiving and managing inspection calendar function unit 211 is used for a manager to inspect the inspection analysis result uploaded by the data analysis system 3, and if the result has an error, the result can be manually modified and an inspection report can be generated.

The inspection information summarizing module 23 is used for checking inspection reports of all fans in the wind field uploaded by the management inspection workbench module 22, summarizing inspection result information of all fans in the wind field, providing a user inquiry function, and inquiring related fan inspection conditions by inputting inquiry conditions.

As shown in fig. 4, the data analysis system 3 includes a data import/export module 31 (i.e., a data import/export module a), an image preprocessing module 32, a defect marking module 33, a defect classifying module 34, a defect diagnosing module 35, and another data import/export module 31 (i.e., a data import/export module B) connected in sequence by signals, and is configured to analyze the fan image data uploaded by the data acquisition system 4, and upload the analysis result to the task processing system 2 for auditing.

Wherein,

The data import/export module 31 is in signal connection with the task processing system 2 and the data acquisition system 4;

The other data import/export module 31 is signally connected to the task processing system 2.

The data import/export module 31 uploads the image data of the data acquisition system 4 to the server in the module 31 in an automatic synchronization or manual import mode, the imported image can be checked and subjected to defect analysis after uploading, and the completed image analysis result can be automatically synchronized to the task processing system 2 by the other data import/export module 31 and also can be manually exported for comparison and check.

The image preprocessing module 32 preprocesses the image data, performs such processes as background rejection and illumination removal on elements of the fan blade for analysis against background, illumination and the like, and then performs region division on the elements by color band marks made on the surface of the blade to determine a monitored region for further analysis.

The defect marking module 33 marks the defects on the image by utilizing a box, the length and the width of the box can cover the whole defect position and be close to the defect boundary, and the defects are marked by adopting different colors according to the difference of the severity of the influence of the defect position on the quality of the blade.

The defect classification module 33 is configured to classify the marked defects, extract and identify image features of the marked target area of the blade image, establish a fault feature database to identify and classify the defect image, and give a defect class name.

The defect diagnosis module 35 is configured to perform grading diagnosis on the identified defects, establish grading rules based on comprehensive information such as damage position and size, and respectively select the length or area of the defects as the size index of the defect grading according to different defect types, wherein the defects are classified into four grades of slight damage, general damage, medium damage, heavy damage and fatal damage according to the severity.

The data analysis system 3 may be provided with a patrol end and a data center end,

The inspection end is operated on each inspection device such as a notebook computer and is used for analyzing the image data of the blade obtained by the inspection device from the data acquisition system 4 by using a trained model to give a preliminary analysis result including whether the inspected blade has defects and the defect positions;

The data center end is used for reading the inspection picture uploaded by the inspection end and carrying out recognition analysis, reading the inspection picture uploaded by the inspection end, realizing fine recognition analysis of the picture, and giving out detailed analysis results containing all information of whether the blade has defects, defect position labels, defect categories, defect classification and the like.

As shown in fig. 5, the data acquisition system 4 includes a fan information acquisition module 41, a ground control module 42, and an intelligent patrol unmanned aerial vehicle 43.

Wherein,

The fan information acquisition module 41 is used for acquiring information such as a fan impeller orientation angle, an impeller phase angle and the like on the ground, and the fan information acquisition module 41 comprises a handheld direction-finding unit 411 and an angle calculation unit 412 connected with the unit 411 in a signal manner;

The handheld direction-finding unit 411 can select each point of the blade tips of two blades above the tested fan, and measure to obtain the geomagnetic angle of the connecting line between the current observation point and the tested point of the blade tip;

The angle calculating unit 412 is installed on a computer, and can calculate the impeller orientation angle according to geomagnetic angle data measured by the handheld direction finding unit 411.

The ground control module 42 is used for automatically generating a routing inspection route and monitoring a flight process of the unmanned aerial vehicle 43, and the ground control module 42 comprises a route automatic generation unit 421 and a flight process monitoring unit 422;

The route automatic generation unit 421 is in signal connection with the angle calculation unit 412;

The route automatic generation unit 421 directly generates a routing inspection route according to the fan impeller direction, the impeller phase, the hub height, the blade length and the fan coordinate information, and transmits the routing inspection route to the intelligent routing inspection unmanned aerial vehicle 43 for automatic routing inspection;

The flight process monitoring unit 422 is connected with the intelligent inspection unmanned aerial vehicle 43 by utilizing Gao Qingtu transmission and data transmission equipment, and the flight process monitoring unit 422 display screen receives the shooting picture of the unmanned aerial vehicle camera in real time for real-time monitoring.

The intelligent inspection unmanned aerial vehicle 43 comprises an unmanned aerial vehicle flight platform 431, a positioning unit 432, an image acquisition unit 433, a data storage unit 434, an information matching unit 435 and an obstacle avoidance unit 436;

the unmanned aerial vehicle flying platform 431 is in signal connection with the route automatic generation unit 421, the positioning unit 432, the image acquisition unit 433 and the data storage unit 434;

the positioning unit 432 is in signal connection with the image acquisition unit 433, the data storage unit 434 and the information matching unit 435;

the image acquisition unit 433 is in signal connection with the data storage unit 434 and the information matching unit 435;

the information matching unit 435 is in signal connection with the data storage unit 434;

The obstacle avoidance unit 436 is in signal connection with the unmanned aerial vehicle flying platform 431;

the unmanned aerial vehicle flight platform 431 is an industrial unmanned aerial vehicle and is used for carrying all function inspection function modules;

The positioning unit 432 comprises an airborne positioning unit and a ground base station, and can realize centimeter-level positioning of the unmanned aerial vehicle 43 and transmit position information to the information matching unit 435;

The image acquisition unit 433 comprises a high-power camera, a self-stabilizing cradle head and a visual guiding unit, wherein the high-power camera is used for taking a blade picture, and the self-stabilizing cradle head is used for connecting the high-power camera signal to the unmanned aerial vehicle 43 and counteracting the high-power camera shake caused by the flight of the unmanned aerial vehicle 43; the vision guiding unit is used for assisting the high-power camera to aim at the blade in real time, and guaranteeing that the blade is positioned in the center of the view finding frame of the high-power camera;

the data storage unit 434 is configured to receive and store the image data of the information matching unit 435, and transmit the image data to the data analysis system 3 after the unmanned aerial vehicle 43 finishes flying;

the information matching unit 435 is configured to read and match the picture information acquired by the image acquisition unit 433 and the position information of the shooting point, so as to quickly locate the defect position in the data analysis process, and transmit the defect position to the data storage unit 434 after the completion;

The obstacle avoidance unit 436 is used for preventing the unmanned aerial vehicle from collision accident, the obstacle avoidance unit 436 is installed above the unmanned aerial vehicle 43 body, the surrounding environment of the unmanned aerial vehicle is scanned through laser 360 degrees, when the obstacle appears in the set distance, the obstacle avoidance unit 436 sends out an alarm, and the unmanned aerial vehicle 43 stops the route and hovers.

The invention provides a fan set blade inspection method based on the fan set intelligent blade inspection system, which comprises the following steps:

Step 1, creating a wind field: creating wind field information in the task management system 1, filling in wind field codes, wind field names, wind field geographic positions, unit models, blade information and the like, and synchronizing the wind field codes, the wind field names, the wind field geographic positions, the unit models, the blade information and the like to the task processing system 2;

Step 2, creating a task: generating a new inspection task in the task management system 1, and issuing the new inspection task to the task processing system 2 of the corresponding wind field or the task application submitted by the auditing task processing system 2, and allowing the execution;

Step 3, task planning: the task processing system 2 receives the inspection task, makes an inspection plan, and plans the inspection time, the inspection personnel, the inspection equipment number, the planned construction period and the like of each fan in the field;

step 4, executing the job: the data acquisition system 4 executes blade inspection work according to a task plan;

Step 5, data analysis, wherein the data analysis system 3 analyzes the inspection data uploaded by the intelligent inspection unmanned aerial vehicle 43 and uploads the result to the task processing system 2;

Step 6, completing the task: the task processing system 2 examines the data analysis result, records the execution condition of the inspection operation, generates an inspection report and uploads the inspection report to the task management system 1, and the task execution is completed;

step 7, task closing: the task management system 1 examines the task processing system to upload the inspection report, and the task is closed after no objection is confirmed.

Wherein,

1. In the step 5, the inspection terminal runs on each inspection device, such as a notebook computer, analyzes the image data of the fan set blade by using the trained model, and gives out a preliminary analysis result including whether the blade has a defect, a defect position mark or a mark, and the inspection terminal performs the following steps:

step 1a0, reading image data: uploading the image data of the fan blade shot by the unmanned aerial vehicle 43 to the data analysis system 3 through an automatic synchronization or manual introduction mode;

step 2a0, image preprocessing: firstly, removing the background of the fan blade image through an edge extraction algorithm, and then removing the influence of illumination on the fan blade image;

step 3a0, defect identification: screening the fan blade images, removing the non-defective blade images, and marking the defects of the blades by utilizing boxes with different colors;

And 4a0, synchronizing the data analysis results obtained through the steps 1a0-3a0 to the task processing system 2, and giving a patrol report.

The data center reads the blade image obtained by the unmanned aerial vehicle 43 by inspection uploaded by the inspection end, and uses server resources to realize fine recognition analysis of the image, and gives out detailed analysis results of all information including whether the blade has defects, defect position labels, defect categories, defect classification and the like, and the data center performs the following steps:

Step 1a, reading image data: uploading the fan blade image d data shot by the unmanned aerial vehicle 43 to the data analysis system 3 through an automatic synchronization or manual introduction mode;

step 2a, image preprocessing: firstly, removing the background of the fan blade image d through an edge extraction algorithm, and then removing the influence of illumination on the fan blade image d;

Step 3a, defect identification: screening the fan blade image d, removing a non-defective blade image, and marking the defects of the blades by utilizing boxes with different colors;

step 4a, defect classification: extracting and identifying image features of the target area marked by the fan blade image d, and determining the category of the defect;

Step 5a, defect diagnosis: performing grading diagnosis on the identified defects to give the severity of the defects;

step 6a, data uploading: and synchronizing the data analysis results obtained in the steps 1a-5a to the task processing system 2 for auditing, and ending the data analysis process.

2. In said step 4 the following steps are performed:

first, a patrol plan is received: receiving a patrol plan issued by the task processing system 2, carrying the data acquisition system 4 to a specified machine position according to the plan to execute tasks;

secondly, device debugging: the assembly and debugging fan information acquisition module 41, the ground control module 42 and the intelligent inspection unmanned aerial vehicle 43 ensure normal operation;

Step 1b, determining fan information: acquiring information such as fan impeller orientation, impeller phase angle, hub height, blade length, fan coordinates and the like;

Step 2b, generating an air route: inputting fan information into a ground control module 42, automatically generating route information data, and transmitting the route information data to an intelligent patrol unmanned aerial vehicle 43;

Step 3b, blade inspection: the intelligent inspection unmanned aerial vehicle 43 is started by one key, automatically flies along the route, photographs the blades and stores, and the unmanned aerial vehicle 43 automatically descends after the route is executed;

Step 4b, task end: the blade image data is imported into the data analysis system 3, the equipment is retracted, and the task is ended.

Wherein,

The step 3b includes:

Matching the stored image data with GPS position information of the shooting points one by one;

In the shooting process, the high-power camera in the intelligent inspection unmanned aerial vehicle 43 is aligned to the blade in real time by using a vision auxiliary lens, so that the blade is ensured to be positioned in the center of the view-finding frame of the high-power camera;

the obstacle avoidance unit 436 on the unmanned aerial vehicle 43 emits laser to scan the surrounding environment of the unmanned aerial vehicle in real time, so that safety is ensured.

Referring to fig. 6, the step 1b further includes:

Step 1c, turning a fan impeller, and locking the impeller after the impeller rotates to a Y-shaped position in the figure;

Step 2c, a patrol worker measures the tip of any blade (measured point one) above the impeller at any position below the fan tower by using a direction-finding device such as a handheld direction-finding unit 411, and records the geomagnetic angle 1 and the distance La of a connecting line between the current position of the direction-finding device and the measured point one;

step 3c, the patrol staff keeps the position unchanged, measures the tip of the other blade above the impeller (measured point II) by using a direction-finding device such as a handheld direction-finding unit 411, and records the geomagnetic angle 2 and the distance Lb of the connecting line between the current position of the direction-finding device and the measured point II;

Step 4c, inputting geomagnetic angle 1, geomagnetic angle 2, distance La and distance Lb into angle calculation unit 412, calculating impeller orientation angle,

Or referring to fig. 7, the step 1b further includes:

step 1d, locking the impeller of the wind turbine after the impeller rotates to a Y-shaped position;

step 2d, vertically projecting a connecting line between an observation point of a patrol worker and a measured point 1 of a blade tip above the impeller to the ground to obtain a direction 1, taking a fixed direction such as a clockwise included angle between the north direction and the direction 1 to obtain a geomagnetic angle c1, and obtaining the projection line length L1 of the projection;

Step 3d, vertically projecting a connecting line between an observation point of a patrol worker and a measured point 2 of the tip of the other blade above the impeller to the ground to obtain a direction 2, taking an included angle between the fixed direction and the direction 2 in a clockwise direction to obtain a geomagnetic angle c2, and obtaining the projection line length L2 of the projection;

And 4d, calculating a clockwise included angle c3 between the fixed direction and the direction 3 through the geomagnetic angle c1, the geomagnetic angle c2, the projection line length L1 and the projection line length L2, and subtracting 90 degrees from the included angle c3 to obtain the orientation angle of the impeller, wherein the direction 3 is the direction of a ground projection connecting line of the measured point 1 and the measured point 2.

The inspection route of the intelligent inspection unmanned aerial vehicle 43 traverses the two sides of each blade of the inspected fan in sequence. For example, a common fan is generally provided with three blades, so that the automatic inspection route is in a Y shape as a whole, and the rear edges of all three blades are traversed in sequence.

According to the intelligent inspection system for the fan set blades and the related methods, factors such as task scheduling, plan execution, automatic inspection, intelligent analysis, report auditing and the like in the process of inspecting the fan blades are comprehensively considered, inspection task arrangement of each wind field is uniformly planned, the wind field can submit task applications according to self needs, inspection results are audited layer by layer, and error probability is reduced; the automation and intelligent degree of the inspection process and the data processing are improved, the inspection efficiency of the blade is improved, and the inspection cost of the unmanned aerial vehicle is reduced.

The implementation method related to the application is only a preferred scheme and is not used for limiting the protection scope of the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application, that is, although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims (11)

1. An intelligent inspection system for fan set blades, comprising: a task management system, a task processing system, a data analysis system and a data acquisition system,

Wherein,

The task management system is in signal connection with the task processing system and is used for carrying out unified planning management on the inspection task and issuing the inspection task to the task processing system;

The task processing system is in signal connection with the data analysis system and the data acquisition system, and is used for distributing the inspection task to the data acquisition system and receiving, managing and auditing the analysis result of the data analysis system;

the data acquisition system is used for arranging and executing blade inspection work according to the inspection task and uploading stored blade image data to the data analysis system;

The data analysis system is in signal connection with the data acquisition system and is used for automatically analyzing according to the inspection image data of the data acquisition system and uploading analysis results;

The number of the task processing systems is more than or equal to 1, and each task processing system is in signal connection with the corresponding data analysis system and data acquisition system;

Wherein the data acquisition system comprises a fan information acquisition module, a ground control module and an intelligent inspection unmanned aerial vehicle,

The fan information acquisition module is used for acquiring the direction angle and the phase angle of the fan impeller;

the ground control module is used for automatically generating a routing inspection route and monitoring a flight process of the unmanned aerial vehicle;

the fan information acquisition module comprises a handheld direction finding unit and an angle calculating unit, and the handheld direction finding unit is in signal connection with the angle calculating unit;

The handheld direction-finding unit is used for selecting each point of the blade tips of two blades on the tested fan, and measuring to obtain the geomagnetic angle of the connecting line between the current observation point and the tested point of the blade tips;

the angle calculating unit is used for calculating the impeller orientation angle through the geomagnetic angle;

The ground control module comprises an automatic route generating unit and a flight process monitoring unit;

The automatic route generating unit is in signal connection with the angle calculating unit;

The automatic route generating unit is used for directly generating a routing inspection route according to the direction of the impeller of the tested fan, the phase of the impeller, the height of the hub, the length of the blade and the coordinate information of the fan, and transmitting the routing inspection route to the intelligent routing inspection unmanned aerial vehicle for automatic routing inspection;

The flight process monitoring unit is used for receiving the shooting picture of the unmanned aerial vehicle in real time and monitoring in real time.

2. The intelligent inspection system of fan set blades according to claim 1, wherein the task management system comprises a wind field information management module, an inspection task management module and an inspection report management module,

Wherein,

The wind field information management module is used for creating or importing wind field information, wherein the wind field information comprises a wind field code, a wind field name, a wind field geographic position, a unit model and blade information;

the inspection task management module is used for issuing all the inspection tasks and auditing the inspection task information reported by all the wind farms according to the unified planning of the area, and checking the in-process inspection task and historical task information;

The patrol report management module is used for checking the detail of the patrol report uploaded by the task processing system, auditing and uniformly managing the patrol report;

The wind field information management module comprises a manual wind field information introduction functional unit, a template wind field information introduction functional unit and a wind field information editing functional unit, wherein the manual wind field information introduction functional unit, the template wind field information introduction functional unit and the wind field information editing functional unit are in signal connection with the inspection task management module;

The manual wind field information importing functional unit is provided with a wind field creating interface, and new wind field information is manually created by manually inputting corresponding information into the wind field creating interface;

the template importing wind field information function unit is used for filling in template data, selecting importing template information and creating wind field information;

The wind field information editing functional unit is used for inputting and modifying the wind field information;

The inspection task management module comprises an inspection task issuing functional unit, an inspection task auditing functional unit and an inspection task list functional unit;

the patrol task issuing functional unit and the patrol task auditing functional unit are both in signal connection with the patrol task list functional unit and the task processing system;

the patrol task list functional unit is in signal connection with the patrol report management module;

The inspection task issuing functional unit is used for an administrator to issue the inspection task aiming at a certain wind field;

The inspection task auditing function unit is used for auditing the inspection task submitted by the wind farm by a manager;

the patrol task list functional unit is used for displaying all patrol plans;

and the patrol report management module is in signal connection with the task processing system.

3. The intelligent inspection system for fan blades according to claim 1, wherein the task processing system comprises an inspection plan management module, an inspection workbench module and an inspection information summarization module,

Wherein,

The inspection plan management module is in signal connection with the inspection workbench module, the task management system and the data acquisition system;

the inspection workbench module is in signal connection with the inspection information summarizing module, the data acquisition system and the data analysis system;

The patrol information summarizing module is in signal connection with the task management system;

the inspection plan management module is used for making and filling an inspection plan according to the received inspection task and distributing the inspection plan to each data acquisition system;

the inspection workbench module is used for inputting the inspection information of the blade, has temporary storage and preview functions, and inquires, modifies and generates an inspection report;

the inspection information summarizing module is used for checking and managing the inspection reports of all fans in the wind field range uploaded by the inspection workbench module;

The inspection plan management module comprises an inspection calendar functional unit, an inspection plan editing functional unit and an inspection plan inquiring functional unit;

the inspection calendar functional unit and the inspection plan editing functional unit are connected with the inspection plan inquiring functional unit through signals;

the inspection calendar functional unit is used for displaying an inspection plan of the latest date;

The inspection plan editing functional unit is used for filling an inspection plan, and submitting the inspection plan to archiving after filling;

the inspection plan inquiring functional unit is used for receiving and managing the inspection plan information uploaded by the inspection calendar functional unit and the inspection plan editing functional unit;

the inspection workbench module comprises an inspection site information input functional unit and a query, modification and report generation functional unit;

The inspection site information input functional unit is used for inputting the blade inspection information uploaded by the data acquisition system;

The receiving and managing inspection calendar function unit is used for enabling a manager to inspect inspection analysis results uploaded by the data analysis system and generating the inspection report.

4. The intelligent inspection system of fan set blades according to claim 1, wherein the data analysis system comprises a data import/export module a, an image preprocessing module, a defect marking module, a defect classifying module, a defect diagnosing module and a data import/export module B;

Wherein,

The data importing/exporting module A, the image preprocessing module, the defect marking module, the defect classifying module, the defect diagnosing module and the data importing/exporting module B are connected in sequence through signals;

The data import/export module A is in signal connection with the task processing system and the data acquisition system;

The data import/export module B is in signal connection with the task processing system;

the data import/export module A is used for uploading the image data of the data acquisition system;

the image preprocessing module is used for preprocessing the image data and determining a monitored area;

The defect marking module is used for marking the defects of the image;

the defect classification module is used for classifying the marked defects;

The defect diagnosis module is used for carrying out grading diagnosis on the identified defects;

The data import/export module B is used for uploading the analysis result of the completed image to the task processing system.

5. The intelligent inspection system for fan blades according to claim 1 or 4, wherein the data analysis system is provided with an inspection end and a data center end,

Wherein,

The inspection end is operated on each inspection device and is used for analyzing the image data and giving out a preliminary analysis result including whether the inspected blade has defects and the defect positions;

the data center end is used for reading the blade image obtained by inspection uploaded by the inspection end, carrying out recognition analysis, and giving out detailed analysis results including whether the inspected blade has defects and defect positions, defect types and defect classification.

6. The intelligent inspection system for fan set blades according to claim 1, wherein,

The intelligent inspection unmanned aerial vehicle comprises an unmanned aerial vehicle flight platform, a positioning unit, an image acquisition unit, a data storage unit, an information matching unit and an obstacle avoidance unit;

the unmanned aerial vehicle flight platform is in signal connection with the route automatic generation unit, the positioning unit, the image acquisition unit and the data storage unit;

the positioning unit is in signal connection with the image acquisition unit, the data storage unit and the information matching unit;

The image acquisition unit is in signal connection with the data storage unit and the information matching unit;

The information matching unit is in signal connection with the data storage unit;

The obstacle avoidance unit is in signal connection with the unmanned aerial vehicle flight platform;

the unmanned aerial vehicle flight platform is used for carrying all the function inspection function modules;

the positioning unit comprises an airborne positioning unit and a ground base station and is used for positioning the unmanned aerial vehicle and transmitting position information obtained by positioning to the information matching unit;

the image acquisition unit comprises a high-power camera, a self-stabilizing cradle head and a visual guiding unit, wherein the high-power camera is used for taking the pictures of the blades; the self-stabilizing cradle head is used for connecting the high-power camera signal to the unmanned aerial vehicle and counteracting high-power camera shake caused by the flight of the unmanned aerial vehicle; the vision guiding unit is used for assisting the high-power camera to aim at the blade in real time, and guaranteeing that the blade is positioned in the center of a view finding frame of the high-power camera;

The data storage unit is used for receiving and storing the image data of the information matching unit, and transmitting the image data to the data analysis system after the unmanned aerial vehicle finishes flying;

The information matching unit is used for reading and matching the picture information acquired by the image acquisition unit and the position information of the shooting point where the unmanned aerial vehicle is located so as to locate a defect position and transmit the defect position to the data storage unit;

The obstacle avoidance unit is used for scanning the surrounding environment of the unmanned aerial vehicle, and when an obstacle appears in a set distance, the obstacle avoidance unit gives an alarm, and the unmanned aerial vehicle suspends a route and hovers;

The inspection route sequentially traverses the two sides of each blade of the fan to be inspected.

7. A fan set blade inspection method based on the fan set blade intelligent inspection system of any one of claims 1-6, the method comprising:

Step 1, creating a wind field: creating wind field information in the task management system, filling in wind field codes, wind field names, wind field geographic positions, unit models and blade information, and synchronizing the wind field codes, the wind field names, the wind field geographic positions, the unit models and the blade information to the task processing system;

Step 2, creating a task: generating a new inspection task in the task management system, and issuing the inspection task to the task processing system of a corresponding wind field or auditing a task application submitted by the task processing system, and allowing execution;

Step3, task planning: the task processing system receives the inspection task, makes an inspection plan, and plans the inspection time, the inspection personnel, the inspection equipment number and the construction period of each fan in a field;

Step 4, executing the job: the data acquisition system executes blade inspection work according to the task plan;

Step5, data analysis: the data analysis system analyzes the inspection data uploaded by the intelligent inspection unmanned aerial vehicle and uploads the result to the task processing system;

step 6, completing the task: the task processing system examines the data analysis result, records the execution condition of the inspection operation, generates an inspection report and uploads the inspection report to the task management system;

Step 7, task closing: and the task management system examines the task processing system to upload the inspection report, and the task is closed after no objection is confirmed.

8. The fan set blade inspection method of claim 7, wherein in step 5 the data analysis system performs the steps of:

Step 1a0, reading image data: uploading fan blade image data shot by the unmanned aerial vehicle to a data analysis system;

Step 2a0, image preprocessing: removing the background of the fan blade image, and removing the influence of illumination on the fan blade image;

step 3a0, defect identification: screening the fan blade images, removing the non-defective blade images, and marking the defects of the blades by utilizing boxes with different colors;

step 4a0, synchronizing the data analysis results obtained through the steps 1a0-3a0 to the task processing system, giving a patrol report,

Or (b)

The following steps are performed:

step 1a, reading image data: uploading fan blade image data shot by the unmanned aerial vehicle to a data analysis system;

Step 2a, image preprocessing: removing the background of the fan blade image, and removing the influence of illumination on the fan blade image;

Step 3a, defect identification: screening the fan blade images, removing the non-defective blade images, and marking the defects of the blades by utilizing boxes with different colors;

step 4a, defect classification: extracting and identifying image features of the target area obtained by marking, and determining the category of the defect;

Step 5a, defect diagnosis: performing grading diagnosis on the identified defects to give the severity of the defects;

Step 6a, data uploading: and (3) synchronizing the data analysis results obtained in the steps 1a-5a to the task processing system for auditing, and ending the data analysis process.

9. The fan set blade inspection method of claim 7, wherein in step4, the following steps are performed:

step 1b, determining fan information: acquiring fan impeller orientation, impeller phase angle, hub height, blade length and fan coordinate information;

step 2b, generating an air route: inputting fan information into the ground control module, automatically generating route information data, and transmitting the route information data to the intelligent inspection unmanned aerial vehicle;

Step 3b, blade inspection: the unmanned aerial vehicle automatically flies along the route, and images of the blades are shot and stored;

step 4b, task end: the blade image data is imported into the data analysis system.

10. The method for inspecting fan blade of claim 9,

The step 3b includes:

Matching the stored image data with GPS position information of the shooting points one by one;

In the shooting process, a high-power camera in the intelligent inspection unmanned aerial vehicle is aligned to the blade in real time by utilizing a vision auxiliary lens, so that the blade is ensured to be positioned in the center of a view finding frame of the high-power camera;

The obstacle avoidance unit on the unmanned aerial vehicle emits laser in real time to scan the surrounding environment of the unmanned aerial vehicle, so that safety is ensured.

11. The method for inspecting fan blade of claim 9,

The step 1b further includes:

step 1c, turning a fan impeller, wherein the impeller is locked after rotating to a Y-shaped position;

Step 2c, measuring a measured point a of the tip of one blade above the impeller by using a direction-finding device, and recording a geomagnetic angle aa and a distance La of a connecting line between the current position of the direction-finding device and the measured point a;

step 3c, keeping the current position of the direction-finding device unchanged, measuring a measured point b of the tip of the other blade above the impeller, and recording a geomagnetic angle bb and a distance Lb of a connecting line between the current position and the measured point b;

step 4c, calculating the orientation angle of the impeller according to the geomagnetic angle 1, the geomagnetic angle 2, the distance La and the distance Lb,

Or (b)

The step 1b further includes:

step 1d, locking the impeller after the impeller of the fan rotates to a Y-shaped position;

Step 2d, vertically projecting the line between the observation point and a measured point a1 of a blade tip above the impeller to the ground to obtain a direction b1, and obtaining an included angle between a fixed direction and the direction b1 to obtain a geomagnetic angle c1 and a projection line length L1;

Step 3d, vertically projecting the line between the observation point and a measured point a2 of the tip of the other blade above the impeller to the ground to obtain a direction b2, and obtaining an included angle between the fixed direction and the direction b2 to obtain a geomagnetic angle c2 and a projection line length L2;

And 4d, calculating an included angle c3 between the fixed direction and the direction 3 through the geomagnetic angle c1, the geomagnetic angle c2, the projection line length L1 and the projection line length L2, and subtracting 90 degrees from the included angle c3 to obtain the orientation angle of the impeller, wherein the direction 3 is the direction of the ground projection connecting line of the measured points a1 and a 2.