Research On Fault Diagnosis And State Comprehensive Evaluation Method For Large-scale Wind Turbine

With the shortage of the world’s energy,wind energy has played an important role in various occasions and wind energy technology has been widely used in the world.However,due to remote location,poor maintenance technology and so on,which is prone to accident,maintenance difficulties,higher operating and maintenance costs in wind power generation process,making the operation of wind power generation system assessment work is particularly important.Wind power generation system is very complicated,it is difficult to carry out fault diagnosis and maintenance engineering,so it’s hardly to complete the equipment and diagnosis accurately rely on a single technology In summary,the wind power generation fault diagnosis and evaluation technology requires a reasonable evaluation system to solve the theses problems.Therefore,the main contents of this paper are as follows:(1)This paper first describes the basic structure and working principle of the current wind turbines,for the operating characteristics of wind turbines,the failure mechanism of the crew is analyzed,and the common methods of failure-prone parts or fault diagnosis are studied.Finally,based on the measured operating data of wind turbine SCADA system,two kinds of fault diagnosis and state comprehensive evaluation methods are designed.(2)Based on the normal and fault data of a wind farm SCADA system,a comprehensive evaluation model is established by using entropy weight fuzzy comprehensive evaluation method.According to the model,the running state of the unit is divided,in this paper,the operation status of the unit is divided into four grades: "excellent","good","medium","and "poor" to evaluate the health status of wind turbines.When calculating the running state as "poor" subsystem,it will be regarded as a fault state,and need to stop the machine immediately to check,avoiding the failure of serious and resulting in greater losses.(3)When the fault occurs,it is necessary to carry out a comprehensive diagnosis of the various parts and components.Then,The fault tree model of wind turbine is established the fault diagnosis knowledge of wind turbine based on fault tree is build.At the same time,this paper Introduces three kinds of fault reasoning machine.By using the method of forward and backward hybrid reasoning,the flow chart of fault diagnosis and the process of condition evaluation reasoning is designed,which realizes the simplification and modularization of complex system.(4)At the end of this paper,combined with fault tree model,by analyzing the normal and fault condition of the gearbox system and the generator system,which found that even a small accident can also affect the normal operation.Therefore,using the above method,when an abnormality phenomenon appears in a component of the unit,the system can timely and effectively identify the cause of the failure and can handle the failure as soon as possible,so it can provide some reference for fault diagnosis and treatment of the wind turbine.