A hybrid identification scheme combining singular entropy with ERA for wind turbine systems

Abstract

This paper presents an approach for identification of wind power conversion systems based on related real data from the wind farm supervisory control and data acquisition (SCADA) module. The proposed method is able to determine the order of the system more accurately and identify the system parameters more effectively, and such improvement is accomplished in two steps. The first step is to use a “hybrid” algorithm based on singular entropy increment (SEI) and weighted criterion to overcome the uncertainty and arbitrariness of traditional data based order determination methods. The second step is to blend SEI analysis and eigensystem realization algorithm (ERA) with Kalman filter to improve the parameter estimation accuracy in the presence of white noise inevitable in wind turbine systems. As a validation of the developed method, the SKYSTREAM 3.7 is used as a benchmark for simulation experiments, which demonstrate that the proposed hybrid scheme leads to good performance in wind power system order determination and parameter identification.