Research On Fault Diagnosis System Of Fan Blade Based On Inspection Image

As a clean and renewable energy source,wind energy has attracted the attention of the whole society and has become one of the major substitutions of traditional fossil energy.In recent years,the wind field construction has been deepening,the fan installed capacity keeps rising.Fan blade is an important part of fan energy conversion,which transforms wind energy into mechanical energy and is the main bearing part of the fan,and it plays a key role in the safe operation of the whole fan.In the process of continuous operation of fan blades,there will be failures.For instance,the sand dust on the sur face of the blade causes the sand holes appear,and it is inevitable that the blade of the blades can be twisted,bent,and vibrated to cause cracks or faults in the blades,and the blade was struck by lightning because of the high position of the fan blad e.During the service of the fan blade to the state of the fan blade can timely monitoring,so as to timely diagnose fault,to prevent the whole power system fault due to the blade failure and reduce the economic losses.For this purpose,we have designed a set of detection system for fan blades based on inspection image to conduct the fault diagnosis of fan blade.This paper mainly includes the following aspects:1)The fault types and existing detection methods of fan blades are analyzed after reading a large number of domestic and foreign literatures and field investigation of wind farms.The existing detection methods mainly include: infrared detection,ultrasonic testing,acoustic emission detection technology,etc.In addition,the causes and repair met hods of the blade fault are studied.2)In this paper,the fault detection system of fan blade based on inspection image i s designed from two parts: hardware and software.The design of the hardware system is to collect the clear image of the fan blade,the key point and difficulty of which is to select the appropriate image acquisition equipment.The software system is to improve the efficiency of the system and the accuracy of fault detection.The key point and difficulty is to select the appropriate image processing algorithm to improve the accuracy of the system.3)The identification,location and fault detection algorithm of fan blades are studied in order to realize the accurate location and fault detection of fan blade.Algorithm proposed in this article,firstly improve the efficiency of operation by simplifying image through gray-scale algorithm,and then obtain the optimal pretreatment effect using OTSU method.After that,a fault location algorithm based on morphological and connected domain markers is studied,and the fault location is determined by using th e open and closed operation and image difference operation.Finally,the fault type and structure information are determined based on the target identification method.