Long-Term Wind Speed and Power Forecasting Using Local Recurrent Neural Network Models

TLDR

Long‑term wind speed and power forecasting from meteorological data is the focus of this study. The study proposes two novel online learning schemes for updating recurrent network weights using a recursive prediction error algorithm. Hourly 72‑hour forecasts for a Crete wind park are generated using SKIRON meteorological inputs and three local recurrent neural networks (IIR‑MLP, LAF‑MLN, diagonal RNN) with internal IIR feedback, and their performance is compared to static FIR‑NN and MLP models. The proposed online learning schemes guarantee continuous stability and yield recurrent models that outperform both conventional dynamic back‑propagation and static FIR‑NN/MLP baselines, achieving significant gains over the persistent method.

Abstract

This paper deals with the problem of long-term wind speed and power forecasting based on meteorological information. Hourly forecasts up to 72-h ahead are produced for a wind park on the Greek island of Crete. As inputs our models use the numerical forecasts of wind speed and direction provided by atmospheric modeling system SKIRON for four nearby positions up to 30 km away from the wind turbine cluster. Three types of local recurrent neural networks are employed as forecasting models, namely, the infinite impulse response multilayer perceptron (IIR-MLP), the local activation feedback multilayer network (LAF-MLN), and the diagonal recurrent neural network (RNN). These networks contain internal feedback paths, with the neuron connections implemented by means of IIR synaptic filters. Two novel and optimal on-line learning schemes are suggested for the update of the recurrent network's weights based on the recursive prediction error algorithm. The methods assure continuous stability of the network during the learning phase and exhibit improved performance compared to the conventional dynamic back propagation. Extensive experimentation is carried out where the three recurrent networks are additionally compared to two static models, a finite-impulse response NN (FIR-NN) and a conventional static-MLP network. Simulation results demonstrate that the recurrent models, trained by the suggested methods, outperform the static ones while they exhibit significant improvement over the persistent method.