Power Control by Reinforcement Learning
摘要
The previous chapter deals with the Simulated Annealing (SA) method for optimal power limitation to the VRF air-conditioners in order to adapt to the Real-Time Pricing (RTP) as shown in Fig. 11.1. It required a neural network model that predicts power consumption and room temperature to decide the next power limitation. This chapter dealt with a new method that does not require a neural network.