The inherently intermittent and highly variable nature of wind necessitates the use of wind power forecasting tools in order to facilitate the integration of wind turbines in microgrids, among others. In this direction, the present paper describes the development of a short-term wind power forecasting model based on artificial neural network (ANN) clustering, which uses statistical feature parameters in the input vector, as well as an enhanced version of this approach that adjusts the ANN output with the probability of lower misclassification (PLM) method. Moreover, it employs the Monte Carlo simulation to represent the stochastic variation of wind power production and assess the impact of energy management decisions in a residential wind-battery microgrid using the proposed wind power forecasting models. The results indicate that there are significant benefits for the microgrid when compared to the naïve approach that is used for benchmarking purposes, while the PLM adjustment method provides further improvements in terms of forecasting accuracy.Appl. Sci. 2017, 7, 1142 2 of 15 buy/sell electricity from/to the grid, forecasting the power output from microRES within a certain time frame enables more informed decisions in order to maximize the profit for the prosumer [7].A typical approach to estimate the power output of a wind turbine consists in matching forecasted wind speed values on the wind turbine power curve. The relevant literature in the field classifies the wind speed forecasting methods into two broad categories, namely physics-based numerical weather prediction models and data-driven approaches [8]. More specifically, the former rely on the physics of the lower atmospheric boundary layer to produce wind flow information, and are thus characterized by high computational complexity [9]. The latter are based on historical wind speed data and employ either statistics-based methods (e.g., time series [10], Kalman filtering [11], Markov chain models [12], and Bayesian methods [13]) or artificial intelligence-based (e.g., artificial neural networks (ANNs) [14], fuzzy systems [15], and support vector machines [16]), or hybrid approaches that combine both techniques to produce wind speed forecasts [17][18][19]. Depending on the intended application, the time scale of wind forecasting can be divided into ultra-short-term (minutes to 1 h ahead), short-term (1 h to several hours ahead), medium-term (several hours to one week ahead), and long-term (one week to more than six months ahead), with respect to the time horizon of the forecasts [20,21].In general, microgrids can operate either in grid-connected or islanded mode (e.g., in autonomous applications or due to faults in the upstream network), incorporating appropriate control strategies to ensure that the techno-economic requirements for an enhanced energy utilization rate and reduced operating costs are met [22,23]. Given that wind energy is one of the most attractive renewable energy sources (RES) in terms of efficiency and cost-competitiveness, yet it is in...