8th International Conference on Advanced Power System Automation and Protection (APAP2019)

In recent years, wind power generation is rapidly gaining popularity due to the major concerns about the excessive emissions and the worldwide electrical energy crisis. In addition, this kind of power systems have shown more security options than others. Due to the highly variable and intermittent nature of the wind energy, it is crucial to achieve higher accuracy of long-term wind speed prediction for improving the reliability and economic feasibility of power systems. Hence, this paper proposes a novel methodology for long-term wind speed prediction using fuzzy logic-based prediction network. It uses fuzzy logic and employs Mamdani product inference engine, singleton fuzzifier and center average de-fuzzifier. The algorithm is capable of predicting values in linear, non-linear and even chaotic sequences. Literature search indicates that many researchers have developed similar algorithms and used Mackey Glass time series for prediction. Our attempt has used data series of wind speed. The simulation results indicate that the algorithm works reasonably well. However, prediction accuracy, in occasion of data series, depends on extent of chaos. This algorithm works reasonably well even when the mathematical model of the system is not available.

Fuzzy Logic, Inference Engine, Mamdani, Chaotic, Prediction, Wind Speed, Power Generation

No Data