Design a Markov Decision Process-Based Dynamic Deadband Threshold Strategy for Primary Frequency Response Control

Abstract

Given that the static fixed deadband threshold strategy is difficult to coordinate the operational economy and security flexibly in primary frequency response control, then a dynamic deadband threshold strategy for primary frequency response control is proposed. Based on the aggregated system primary frequency response model, a Markov Decision Process (MDP) model is established with the sequential decision theory. Then, the value iteration algorithm is introduced to derive the dynamic deadband threshold strategy of primary frequency response control by solving the Bellman equation. Case studies are presented to verify the performance of the proposed strategy. It shows that the dynamic deadband threshold strategy is superior to the static fixed deadband threshold strategy on the aspects of economy and security. The research can provide further reference for finely management and control of primary frequency response.