This dissertation introduces a novel approach to improve the reliability in the short-term prediction of power generation for wind farms. The wind forecasting models need to be improved, and appropriate methods of wind measurement are also essential to yield accurate data for wind power prediction. Therefore, in this work, the study is divided into two parts, which are the study of wind measurement and the study of wind power forecasting. In wind measurement, the sampling rate is a crucial factor in wind data acquisition, for good accuracy in wind analysis. A high sampling rate is preferable for wind speed measurement. However, when a measurement at a high sampling rate is performed, a large amount of data is obtained for storage and computation. The Nyquist-based adaptive sampling rate method adapts the sampling rate to be the Nyquist frequency, according to actual wind conditions. In this study, the wind data at a high sampling rate of 10 Hz is used as a benchmark. The proposed Nyquist-based methodology is capable of providing high accuracy of analytical results, with percentage relative differences of less than 1% in wind analysis. In addition, the amount of wind data is significantly decreased (by 4000 times) from the benchmark. In forecasting wind power, the predictions of the autoregressive moving average model, the artificial neural network model, and the grey prediction model are comparatively studied for wind power generation. In this study, the weighting method systematically combines the predicted values of those three predictive models over time, based on their forecasting performance by the root mean square errors (RMSEs) between the actual values and the predicted values. The multiple forecasting models are applied to predict the wind power generation of a wind farm 1 h, 3 h, and 6 h ahead. The RMSEs of the multiple forecasting models are significantly the lowest values among those three predictive models and the benchmark by the persistence model. Furthermore, the prediction interval around the predicted value is statistically determined, to indicate the feasible range of wind power generation with a prescribed percentage of confidence under uncertainty. Uncertainty caused the historical prediction errors.