Adaptive prediction of power fluctuations from a wind turbine at Kalpitiya area in Sri Lanka

Hydro power is the major renewable energy contributor to the national grid in Sri Lanka amounting to 48% of the total installed capacity. Further expansion of hydropower however is limited due to environmental and resource constraints. Meanwhile the demand for electricity is estimated to rise at an annual rate of 8% - 10% prompting the need to find alternative power options. The wind energy has been identified as a promising candidate to generate electricity in Sri Lanka. However for a reliable integration of wind energy the volatile nature of wind has to be understood. Wind speed-time series data typically exhibit autocorrelation, which can be defined as the degree of dependence on preceding values. Generally, statistical models and neural network techniques are being used for time series analysis. Present study shows how an adaptive digital filter can serve as a modelling, forecasting and monitoring technique, and, how they contribute to a successful integration of wind power into the national grid. The north-western region of Kalpitiya has been identified as one of the potential location for wind power generation in the country. This study also predicts power generation and investigates power fluctuations for grid integrations of a commercially available wind turbine installed in Kalpitiya area by using measured wind speeds and performance of the wind turbine.