| Wind energy has drawn more and more attention around the world as a clean, renewable resource. Wind turbine is the main means of wind power generation and in recent years, with the rapid development of aerodynamics, mechanical engineering, electrical engineering, control engineering, structural mechanics, materials science, and other disciplines and fields, good theoretical basis is provided for study and design of wind turbines, which promotes modern wind turbine technology to lightweight, high efficiency, high reliability and large-scale development. To meet the challenge of wind turbine safe, stable and reliable operation, reduction of load change due to the increase of its unit size, improvement of the overall power efficiency, providing quality electricity, and extending the life of the unit, the kinetic model of the machine will also become more complex, and the role of the establishment of an accurate model of wind turbine technology development is also becoming increasingly prominent.In this paper, we base on the National Program on Key Basic Research Project (973Program) project "wind power generation system identification and adaptive control mechanism research (2012CB724404)", which consists of Zhejiang Windey Co., Ltd and Network Sensing and Control Research Group (Institute of Industrial Control, Control Science and Engineering, Zhejiang University) in cooperation. In system identification aspects of wind turbines, the main work of this paper can be listed as:Firstly, the history of the world’s wind power development and future trends is analyzed, and the opportunities and challenges which the world’s wind power industry is facing are given. And the significance of domestic and foreign research of wind turbine modeling techniques, existing problems and direction of future development is described.Secondly, the mechanism modeling technology of wind turbines is studied, and based on the operating characteristics the whole system can be divided into four subsystems as: aerodynamics subsystem, mechanical subsystem, electrical subsystem and actuator subsystem. Also, explanation of each operation’s mechanism and characteristics is given.Thirdly, the basic theory and methods of system identification is studied, and basic control program is explained based on the operation mechanism of wind turbine explaining. Then a control-oriented description of the wind turbine system identification problem is discussed, which divides the identification problem into torque loop and pitch loop, and the corresponding identification trial is designed.Fourthly, the basic features and principles of wind turbine simulation software Bladed is studied, and the utility of Bladed is grasped. Then, based on Bladed a series of work has been done: building of the whole model of wind turbines, export and analysis of linear models, test design of system identification, collect and export of test data, identification based on test data, and verification of the identification result.Fifthly, the LPV(Linear Parameter Varying) basic theory and method is studied, and an ADALINE(ADAptive LINear Element) technology based LPV system identification framework is proposed. Simulation examples are given to verify the validity of this method. Combined with analysis of LPV method and wind turbine characteristic, the Bladed based identification test is designed and furthermore the LPV model is obtained.Sixthly, the basic principles and functions based on Hardware Test is studied, and test experiment is projected through application programming design and development, with the employed of semi-physical simulation platform. System identification based on test data is presented and verification is shown. Then, based on the C#programming and development technology, the foundation of System Identification Toolbox software framework is constructed, and test and validation is demonstrated using Bladed test data. |
PDF Full Text Request