Neural Network-Based Intelligent Computing Algorithms for Discrete-Time Optimal Control with the Application to a Cyberphysical Power System

Algorithm 2

Model-based VI.

	Step 1: (Initialization)
	Let the iteration index .
	Select an initial value function .
	Choose a small enough computation precision .
	Step 2: (Policy Improvement)
	With , compute the iterative control policy by
	Step 3: (Policy Evaluation)
	With , calculate the iterative value function by
	Step 4: if , stop and the optimal control policy is acquired;
	Else, let and go back to Step 2.