This study proposes a new method for direct generation of synthetic wind power time series for a wind farm. The method combines the random nature of wind with the operational information of the wind turbines (i.e., failure and repair rates). It uses chronological or sequential Monte Carlo Simulation instead of non-sequential one due to its usefulness and flexibility in preserving statistical characteristics of the chronological processes. The validity of the synthetic values generated by the proposed method and the conventional Markov Chain Monte Carlo methods is compared with the measured data in terms of average and variance values, Probability Distribution Function, and Auto-Correlation Function. Due to increasing interest in the use of the storage system in paralleling with wind power generation, a practical application of the proposed method is also included. Optimal sizing of various energy storage technologies is obtained through a cost-benefit analysis in a typical Micro-Grid.