Integrating Reliability Models and Adaptive Algorithms for Wind Power Forecasting

Abstract The high proliferation of wind generators used in modern electrical grids determines several critical issues pushing power system operators to improve critical operation functions, such as security analysis and spinning reserve assessment, by taking into account the effects of intermittent and nonprogrammable power profiles. To address this challenging issue, a large number of frameworks for wind power forecasting have been proposed in the literature. Although these tools reliably allow prediction of wind speed and theoretical generated power, more complex phenomena need to be investigated to comprehensively model wind power uncertainty and its effect on power system operation. To address this issue, this chapter proposes a probabilistic model based on Markov chains, which predicts the injected power profiles considering wind speed forecasting uncertainty and generator operation states. Experimental results obtained on a real case study are presented and discussed to assess the performance of the proposed method.