Condition Monitoring And Evaluation Of Wind Turbines Based On SCADA Data

As an important form of clean energy power generation,wind power has been greatly developed by all countries in the world in recent years.Due to the complicated structure of the wind turbine itself and the harsh environment,the fast and effective parameter deterioration threshold is obtained by monitoring the state parameters of the wind turbines,which can effectively improve the safety and reliability of the operation of the wind turbine.Establish a unified model for state assessment to master the performance and efficiency status of wind turbines,which can provide data support for the formulation of production and scheduling strategies for wind farms.In this paper,the following research work is carried out around the status parameter monitoring and operational status assessment of wind turbines:Firstly,starting from the composition of the wind turbine system,the wind turbines are classified according to the different structural types of three wind turbine systems.The working principle and working characteristics of direct drive and doubly fed units are studied.Three levels of structures and core functions of the Supervisory Control and Data Acquisition(SCADA)are introduced,as well as the classification of the svstem monitoring state parameters and the division of operating conditions of wind turbines.Secondly,on the basis of analyzing the correlation between the state parameters of wind turbines,the box line diagram is used for data cleaning to obtain the normal data set of the state parameters.Further,the Copula function is used to analyze the wind speed value of the unit to obtain the wind speed related unit group:by comparing and analyzing the variation characteristics of the same parameter between multiple units and multiple parameters between single unit and the degradation state threshold of the parameter is obtained.Finally,the application state parameter data cloud model is used to describe the ambiguity of the wind turbines performance and the randomness of the wind turbines efficiency.The eigenvalues of the cloud model are used to estimate the degree of deviation and stability of the wind turbines state.Because the model has nothing to do with the capacity of the unit,the state evaluation of the same type of wind turbines and the comparative analysis of different types of wind turbines are realized.Selecting the characteristic data of nine different types and capacity units of five wind farms in different distribution areas for one year as model samples and verification examples proves that the model is feasible and the conclusion is true and reliable.