EFFECTS OF STATISTICAL PROPERTIES OF DATASET IN PREDICTING PERFORMANCE OF VARIOUS ARTIFICIAL INTELLIGENCE TECHNIQUES FOR URBAN WATER CONSUMPTION TIME SERIES

摘要: Water Consumption forecasting is very essential for any development program in an urban area and also proper planning management of water resources. Both variability uncertainty determining consumption includes several concepts which depends on issues related to vague incomplete information. In this context, Artificial intelligence (AI) techniques such as fuzzy logic Adaptive Neuro Fuzzy Inference system (ANFIS) method integrates ANN methods shown the potential benefits a single framework. study,ANFIS methodology proposed self organize model structure adapt parameters short term, medium long term prediction. addition this, results various AI were compared with Logic statistical multiple linear regression (MLR) .The time series data from mixed growth under Mangalore city corporation, Karnataka, India used analysis. The performances evaluated using criteria Mean square error (MSE) relative (MRE).From results,it was found that ANFIS Takaki-Sugeno inference performed better than based Mamdani system. majority cases,MLR but distinct down model. can be successfully employed estimate daily, weekly monthly accuracy.