Long-term industrial load forecasting and planning using neural networks technique and fuzzy inference method

Load forecasting plays a dominant part in the economic optimization and secure operation of electric power systems. The plans of the electric power sector have been done and developed with the aid of statistical prediction methods. Electric utility companies need monthly peak and yearly load forecasting for budget planning, maintenance scheduling and fuel management. This paper presents a new approach based on a hybrid fuzzy neural technique which combines artificial neural network and fuzzy logic modeling for long term industrial load forecasting in electrical power systems. An extensive study is carried out to find the accurate forecasting model through an application on an industrial 10th of Ramadan city in Egypt. Actual record data is used to test the proposed method. A large number of influencing factors have been examined and tested. This paper presents a fully developed system for the prediction of electric maximum demand and consumption for the future 24 months. Also very long-term load forecasting was carried out. The strength of this technique lies in its ability to reduce appreciably computational time and its comparable accuracy with other modeling techniques. The outcome of the study clearly indicates that the proposed composite model of neural network technique and fuzzy inference method can be used as an attractive and effective means for industrial monthly and yearly peak load forecasting. The test results showed very accurate forecasting with the average percentage relative error of 1.98%.