Wind Turbine Performance Monitoring Using Advanced Adaptive Filters

Abstract

This paper analyze the role of adaptive filters for monitoring the performance of the wind turbine. Advanced adaptive filters are implemented using various algorithms like standard least mean square, normalized least mean square, generalized normalized gradient descent, weighted least mean square, recursive leastsquare, and affine projection algorithms. A Comparative analysis is done on the basis of parameters like mean absolute error, root mean square error, R-squared (R 2 ) score to determine which adaptive filter is suitable for estimating wind-power generation. Additional parameters such as convergence rate, computational complexity, and stability of the system are also analyzed to monitor the performance of each filter. Factual power data sets of two different sites are taken from resource file of National Renewable Energy Laboratory. The comparison is done under similar assumptions on both the data sets in order to extract accurate filter performance.