Speaker
Description
Wind power central to the energy transition, yet its variability challenges accurate forecasting. This work introduces an adaptive nowcasting method for probabilistic forecasting method combining the generalised logit transformation with a Bayesian framework. The transformation maps double-bounded wind power data to an unbounded domain, enabling Bayesian inference, while an adaptive mechanism updates the shape parameter using representative samples. Four adaptive methods are compared in a case study of over 100 wind farms in Great Britain over four years, focusing on one-step-ahead 30-minute forecasts. Performance is assessed with the Continuous Ranked Probability Score and functional reliability diagrams. Results show the proposed Bayesian method consistently improves forecast accuracy and reliability, supporting robust grid integration and decision-making.
