Calculating production by using short term demand forecasting models: a case study of fuel supply system
- 142 Downloads
In today’s world, energy is one of the key elements of growth and economic development. Because of the critical role of energy in production costs and social security and environmental issues, it is very important to forecast and to optimize energy consumption. This study investigates the influential factors in urban energy (petrol) consumption. Artificial neural network was used to forecast fuel consumption in one of fuel stations in Tehran. A neural network was trained by Levenberg method and genetic algorithm. Results obtained by this method were compared with results obtained by neural network and regression. This comparison showed that neural network and training by genetic algorithm was more efficient than Levenberg and regression. All of the data used in this study were collected from fuel distribution system in Tehran.
KeywordsFuel consumption Forecasting Neural network Genetic algorithm
- Aghaian H (2013) Statistics of petroleum consumption products, 1392Google Scholar
- Guyon I, Elisseeff A (2003) An introduction to variable and feature selection. J Mach Learn Res 3:1157–1182Google Scholar
- Kermanshahi B (1998) Recurrent neural network for forecasting next 10 years loads of nine Japanese utilities. Neuro comput 23(1–3):125–133Google Scholar
- Lee KY, Choi TI, Ku CC, Park JH (1993) Short-term load forecasting Using diagonal recurrent neural network. In Neural Networks to Power Systems, 1993. ANNPS’93, Proceedings of the Second International Forum on Applications Of. pp 227–232Google Scholar