Agent-based Modeling and Simulation of Household Electricity Demand Profiles

Abstract

In Pakistan, there is shortage of electricity generation. As per Pakistan National Transmission & Dispatch Company (NTDC); Pakistan will be able to meet the projected demand of peak hours in 2019. In Pakistan, the major consumer class is domestic which constitutes approximately more than 40% of the electricity sale share, therefore the importance of forecasted domestic electricity demand profiles cannot be denied. The forecasting of household's profiles is not only an important factor for the establishment of sustainable energy systems but also helps in the development of demand-side management policies, future energy generation mix, planning of electricity transmission and distribution network and end-user tariff designs. Due to lack of efficient electricity infrastructure in Pakistan, domestic electricity consumption is not being accurately monitored, thus we are unable to predict our future domestic electricity demands and always faces electricity shortage of 10 to 12 hours in summer peak season. Smart Grid Infrastructure deployment and Home Energy Management Solutions currently available are very expensive and require a lot of long-term planning for a developing country like Pakistan. To forecast the household’s electricity load profiles, we propose a bottom-up agentbased modeling and simulation framework. Our approach provides households per minute electricity consumption by using the behavioral modeling of electrical appliances uses and suitable for replicating at real-world urban infrastructure scenarios. The proposed ABMS framework will support in: (i) Estimation of the future energy demands; (ii) Analysis of the complex dynamic behavior of the population; (iii) Promote responsible use of energy by incorporating necessary policies; and (iv) Effective production planning using mix strategy electricity generation.