Integrating Evolving Fuzzy Neural Networks and Tabu Search for short term load forecasting

An IntegratedEvolving Fuzzy Neural Network and Tabu Search I(EFNNTS) for short term load forcasting method is presented in this paper.In this paper, a shortterm load forecasting is presented firstusing Fuzzy Hyper-Rectangular Composite Neural Networks (FHRCNNs). Then, we use evolutionary programming (EP) and Tabu Search (TS) to find the optimal solution of the parameters of FHRRCNNs (that parameters include such as synaptic weights(w(jk)), biases (theta(jk)), membership function (m(j)((x) under bar (t))), sensitivity factor in membership function ((Sj)) and adjustable synaptic weight (M-ij and m(ij)). We know that the EP has a good capability at search globe optimal value, but has poor capability search local optimal. But the TS just has good capability at local optimal search. So, here, we combine this two methods advantages to improve the shortcoming of the tradition training that the weights and biases always trapped into-a 1 cal optimal. Finally, we use this (IEFNNTS) can improve the solution quality. Actually, we can reduce the error of load forecasting. The proposed IEFNNTS load forecasting scheme was test using data obtained from a sample study include one year, month and 24 hours. The result demonstrated the accuracy of the proposed load forecasting scheme.