Accurate forecasting is important for cost-effective and efficient monitoringand control of the renewable energy based power generation. Wind based power isone of the most difficult energy to predict accurately, due to the widelyvarying and unpredictable nature of wind energy. Although Autoregressive (AR)techniques have been widely used to create wind power models, they have shownlimited accuracy in forecasting, as well as difficulty in determining thecorrect parameters for an optimized AR model. In this paper, ConstrictionFactor Particle Swarm Optimization (CF-PSO) is employed to optimally determinethe parameters of an Autoregressive (AR) model for accurate prediction of thewind power output behaviour. Appropriate lag order of the proposed model isselected based on Akaike information criterion. The performance of the proposedPSO based AR model is compared with four well-established approaches;Forward-backward approach, Geometric lattice approach, Least-squares approachand Yule-Walker approach, that are widely used for error minimization of the ARmodel. To validate the proposed approach, real-life wind power data of\textit{Capital Wind Farm} was obtained from Australian Energy Market Operator.Experimental evaluation based on a number of different datasets demonstratethat the performance of the AR model is significantly improved compared withbenchmark methods.
展开▼
