Demand side management using collective intelligence and accounting for user comfort

This project was designed to integrate human health and user comfort (with a special focus on elderlies and vulnerable groups while considering climate change and extreme weather events) through developing an AI-based early-warning system integrated into an innovative demand side management (DSM) system based on collective intelligence (CI), which is called CI-DSM. CI is a form of distributed intelligence that emerges in collaborative problem solving and decision making. CI-DSM enables the simple and secure communication and control of buildings and energy system with a minimum need for ICT upgrade. During the course of this project, a novel decision making approach based on Reinforcement Learning (called RL-based approach) was developed to enhance decision making in multi-variant complex environment while accounting for several objectives (e.g. decreased energy demand, increased user comfort, and increased resilience). The climate resilience of the enhanced CI-DSM and its effectiveness to increase the human comfort and energy sustainability in urban areas was assessed. This project was conducted in close collaboration with Hiroshima University in Japan and NODA Intelligent Systems AB in Sweden. The project was financed by VINNOVA.