[1M4-GS-10] AI application

[1M4-GS-10-03] Estimation of Power Saving Potential for Peak Demand Suppression of Electricity

〇Shunsuke Mitsuhashi¹, Shunichi Hattori², Takashi Onoda¹ (1. Department of Industrial and Systems Engineering, College of Science and Engineering, Aoyama Gakuin University, 2. Central Research Institute of Electric Power Industry)

Keywords:AI, SVM, Power Saving, Occupancy Detection, Power Crunch

As of 2023, the supply-demand balance of electric power in Japan is getting tighter, and maintaining the supply-demand balance is becoming more serious. In response to the tight power supply and demand, research is being conducted on how to reduce demand during peak demand periods. KEPCO conducted a field experiment in which coupons were distributed to encourage people to go out during peak demand hours in order to suppress electricity demand in general households. In this experiment, coupons were distributed randomly, so the power-saving effect was weak. Therefore, this study proposes a method for estimating the power consumption potential of each household by using SVM to determine whether or not a household member is at home. Experiments were conducted to estimate and compare the power-saving potentials of the case with the presence/absence judgment and the case where coupons were randomly distributed.