Topic: Price-based Demand Response-A Deep Learning Approach

Lecturer: Prof. Dirk Neumann, Dean of the School of Economics at Albert-Ludwigs-Universität Freiburg

Time: 10:00-11:30 a.m. May 24th, 2023

Venue: B321 Zhixin Building, Central Campus

Abstract: This paper presents a novel Demand Response (DR) program to mitigate grid congestion. The program stands out by adjusting electricity prices online and solely based on observed aggregate electricity consumption. It can cope with non-elastic and time-interdependent demand. Additionally, the solution approach based on Deep Reinforcement Learning reaches a cost saving of 40% to 60%after 25 simulation days. Regarding load flexibility, this paper find that the responses of the four load types strongly differ depending on prices and notification interval, which is an important design parameter. The result simply that system operators should flexibilize their DR programs with the help of learning algorithms and tailor their program to the local load composition, congestion frequency, and forecasting quality.