Reinforcement Learning solution for economic scheduling with stochastic cost function

Abstract

Reinforcement Learning (RL) is a machine learning paradigm in which learning system learns which action to take in different situations by using a scalar evaluation received from the environment on performing an action. One major feature of this learning method is that it can learn in a stochastic environment. RL has been successfully applied to many power system optimization problems. Economic Scheduling is an important optimization problem to decide the amount of generation to be allocated to each generating unit so that the total cost of generation is minimized without violating system constraints. One scheduling issue is to accommodate the stochastic cost behaviour of the different generating units. In this paper we demonstrate the capacity of RL algorithm to account the stochastic nature of fuel cost.