Wind Power Prediction and Dynamic Economic Dispatch Strategy Optimization Based on BST-RGOA and NDO-WOA

Abstract

In order to address the new challenges brought by large-scale wind power grid integration to the safe operation of the power system, a non-deterministic optimization whale swarm optimization algorithm is introduced to predict wind power. This method uses trend optimization function to optimize control coefficients and weight coefficients to ensure the algorithm randomness and avoid falling into local optima. On this basis, combined with the evolutionary strategy of bee species and the fast global search mechanism, a fast global optimization algorithm for bee species transition is proposed to achieve dynamic economic scheduling. The results showed that under clear weather conditions, the mean root mean square error of the prediction obtained by the non-deterministic optimization-whale optimization algorithm was only 2.68%, while the least squares support vector machine method was as high as 5.43%. Under cloudy conditions, the mean root mean square error of this algorithm was only 5.82%, which was 6.28% lower than the particle swarm optimization back propagation algorithm. Under rainy conditions, the mean root mean square error of the algorithm was only 6.75%. In addition, the average running time of the rapid global optimization algorithm for bee species transition on dynamic economic scheduling problems was only 28.9 seconds. The runtime of the artificial bee colony algorithm was as long as 198.6. Overall, the wind power prediction algorithm and dynamic economic dispatch algorithm have achieved significant advantages in prediction accuracy and solution effectiveness. This is conducive to achieving efficient optimization and scheduling of the power system, improving the stability and economy of the power system.