Development of new neurone structure for short term load forecasting

摘要

There are various methods available in literature like state space, ARMA, stochastic time series etc. to model the load forecasting problem. To improve the model accuracy artificial intelligence (AI) techniques like artificial neural network (ANN), knowledge based expert system (KBES) are being used. In this paper the short term load forecasting problem has been formulated using artificial neural networks model. But the existing neural networks have various drawbacks like large training time, huge data requirement to train for a non linear complex load forecasting problem, the relatively larger number of hidden nodes required etc. Hence, an attempt has been made to develop a Neuro-Fuzzy Approach for load forecasting problem overcoming above-mentioned problems of ANN by incorporating the features of Fuzzy Systems. A synergetic approach has been used to develop a fozzyfied neuron model by fuzzyfing the neuron structure, which incorporates the features of simple neuron as well as high order neuron. The proposed model has the capability of representing complex mathematical variations.