Wind power prediction method based on IWOA-ELM

In wind energy storage microgrids, accurately predicting the actual output power of wind farms in advance can effectively improve the stability of grid integration regulation. To address the low prediction accuracy of existing models for wind power characteristic parameters, a wind power characteristic parameter prediction method based on improved whale optimization algorithm of extreme learning machine (IWOA-ELM) was proposed. By optimizing the parameters of the extreme learning machine using an improved whale optimization algorithm, a IWOA-ELM wind power characteristic parameter prediction model based on time series was established to predict the characteristic parameters of future wind power. Model performance was evaluated using metrics such as root mean square error and mean absolute error. Experimental results show that the proposed prediction method has a root mean square error of 5.488% and a mean absolute error of 3.72% for wind speed prediction, and a root mean square error of 19.354% and a mean absolute error of 12.46% for wind direction prediction. The prediction accuracy is significantly higher than that of other wind power prediction models such as WOA-ELM, PSO-ELM, BP, and ELM.