Determination of industrial energy demand in Turkey using MLR, ANFIS and PSO-ANFIS

Energy is one of the most critical inputs in social and economic development, an essential factor increasing living standards creating sustainable development. Since energy indispensable input all sectors, dependence contributed to necessity countries' policy. It has importance predict demand determine policies. According Turkey's annual consumption, industry sector consumed last five years. The prediction analysis with data industrial indicator development relationship between industry. In this study, Multiple Linear Regression (MLR), Adaptive Neuro-Fuzzy Inference System (ANFIS), optimized ANFIS Particle Swarm Optimization (PSO) methods are employed forecast for Turkish sectors. indicators which affect consumption were determined estimate demand. 30-year dataset 1990 2019 was split as training test set.  MLR, PSO- compared according performance evaluation, proper model identified. coefficient determination (R2) PSO-ANFIS, models 0.9951, 0.9889, 0.9932 stage, 0.9423, 0.9181, 0.8776 testing respectively.  study results indicated that PSO-ANFIS showed superior capability least estimation error than MLR models. Consequently, parameters tuned able Turkey high accuracy.