Research On Wind Turbine Blade Fault Detection Method Based On Digital Image

The problem of environmental pollution due to industrial production is becoming more and more serious, improving energy structure and promoting energy reform is one of the methods to solve the problem.Wind energy is a kind of clean energy which is rich in reserves and has great development potential. It has been adapted to the requirement of energy reform in China. In the process of the development of wind power, some problems need to be solved gradually, and the fault detection of the wind turbine blade is one of the most urgent problems.At present, the domestic wind power station commonly used manual inspection methods, namely, in the parking of wind turbine, the use of the telescope to observe whether there is a failure of the blade, this method is time-consuming and inefficient.In this context, this paper studies and designs a digital image based on wind turbine blade vehicle fault detection system, mainly from the following aspectsCollect and organize the domestic and foreign research deeply blades fault detection technique at present: acoustic emission, infrared, ultrasonic, fiber grating detection technology,their principle, application types, advantages and disadvantages are summarized.The blade structure, manufacturing process and material are introduced, the Mechanism of the fault type such as porosity, crack is described.Design of blade on board fault detection system from two parts of hardware and software The key of hardware lies in the selection of a suitable for practical use and promotion of camera, mainly from the shooting distance, focus and zoom, frames of several parameters, such start, through on-the-spot investigation test, and ultimately determine the type of the cameraSoftware focus on improving the degree of automation of the system, improve the convenience of manual operation By cradle head camera operation can achieve fast positioning bladeThrough the regional invasion method can realize the automatic capture blade image and other functions At the same time, the results are given.To collect the blade image digital processing, in order to achieve the purpose of detection and location of the fault Firstly, the image is processed by Two-Value Transfrom. Using K-means clustering method to achieve the segmentation of the blade(target) and the sky(background), According to the target area, the connected domainmark is carried out to determine the position of the blade, on this basis To operate the image with morphology and corrode the point of failure of the image, Then, the image and the original image are carried out and the operation is carried out. Finally, the fault point is marked on the original image.