The long term load forecasting: Manisa city

Electrical energy used in all areas of daily life stands out as an important energy source. The use of electrical energy has important role residential and industrial areas and forms the basis of social and economic development. For this reason, realistic electric load forecasting is very important for the developed and developing countries.Electric load forecasting is examined in short term, medium term, and long term bases. Long-term load forecasting covers a period of more than one year. That plays a crucial role in carrying out electricity network planning of new residential and industrial areas of the existing system.In this study, long-term electric load forecasting for the province of Manisa is performed with linear, curve linear, exponential, multiple regression and, expert systems by using ANN, fuzzy logic and ANFIS. In the prediction methods, month, temperature and population data used for the entire year and electricity consumption of the previous years are used for the output parameters. The consumption of electricity is investigated for three groups including province as a whole, city center town in total. And the best converging methods are defined which using different methods for different groups.Key Words: ANFIS, Artificial neural network, Fuzzy logic, Long term load forecasting, Regression analysis.