A forecasting model approach of sustainable electricity management by developing adaptive neuro-fuzzy inference system.
Aamir Nawaz Khan, Muhammad Asif Nadeem, Muhammad Shahid Hussain, Muhammad Aslam, Aqeel Ahmed Bazmi
Author Information
Aamir Nawaz Khan: Department of Management Sciences, COMSATS University Islamabad, Lahore Campus, Defense Road, Off Raiwind Road, Lahore, Pakistan.
Muhammad Asif Nadeem: Department of Management Sciences, COMSATS University Islamabad, Lahore Campus, Defense Road, Off Raiwind Road, Lahore, Pakistan.
Muhammad Shahid Hussain: Department of Chemistry, University of Sargodha Sub Campus Bhakkar, Punjab, Pakistan.
Muhammad Aslam: Department of Chemical Engineering, COMSATS University Islamabad, Lahore Campus, Defense Road, Off Raiwind Road, Lahore, Pakistan.
Aqeel Ahmed Bazmi: Department of Chemical Engineering, COMSATS University Islamabad, Lahore Campus, Defense Road, Off Raiwind Road, Lahore, Pakistan. abazmi@cuilahore.edu.pk.
With an exponential industrial growth, an accurate demand forecasting of energy is of prime importance for strategic decision-making and new power policies regarding generation and distribution in the power sector. This is a great impediment in economic development as well as shattering people's daily life. Hence, forecasting of energy demand in emerging markets is one of the most important policy tool used by decision-makers all over the world. This study focused on the forecasting approach of electricity consumption in Pakistan by developing a model that is called ANFIS (Adaptive neuro-fuzzy inference system). A framework was developed comprising economic and demographic variables as input. Previous historical data of GDP, population, industry efficiency, and weather (annual average temperature) was collected as input to the model and electricity consumption as output of the model. By developing ANFIS model, forecasting was done up to 2045. The increasing trends with respect to predictors showed significant association with electricity consumption. The overall least error proved this model best for forecasting and planning electricity demand to achieve sustainability in the power sector.
