Adaptive and Energy-Saving Smart Lighting Control Based on Deep Q-Network Algorithm

Abstract

Smart lighting systems play a more and more important role in people's work and life, since efficient adaptive and energy-saving lighting control is a necessary way to improve people's quality of life and create an economical, comfortable, and environmentally friendly lighting environment. To improve the energy consumption in an indoor environment and satisfy body comfort, this paper proposes an adaptive and energy-saving smart lighting control algorithm based on a Deep Q-Network algorithm (DQN), wherein the light number, illuminance and body comfort are considered. The proposed algorithm has some research in other fields, but no one has used it in smart lighting control yet, and few people consider body comfort in the existing energysaving control research of smart lighting. The simulation results show that the proposed DQN based control algorithm can save power up to 62% without body comfort loss, compared to the traditional lighting system. Further, there is no additional delay under the proposed control algorithm.