Optimal model predictive control of energy storage devices for frequency stability of modern power systems

Abstract

Load Frequency Control (LFC) has become a more challenging issue, especially with the increases in generation's unpredictability, inconsistency, and load variations leading to reduced system stability and reliability. This paper presents a novel application of the transient search optimization (TSO) upon Model Predictive Control (MPC) based regulators to solve the LFC problem for multiple zones power networks with hybrid renewable generators and storage devices. An exact model with a governor's dead band, generator's rate constraint, and measurement communication delay are employed. Renewable energy sources (RESs), such as wind and solar systems, in addition to fuel cell generators with different storage elements, such as superconducting magnetic energy storage (SMES) and battery energy storage (BES), are incorporated into the power system investigated in this study. To assess the effectiveness of the TSO-MPC controllers in nonlinear conditions due to renewable energy generation and storage ambiguities, actual wind speed measurements are incorporated into the wind generation model and real solar irradiance and temperature data for the photovoltaic system to achieve an extra credible analysis. Furthermore, the performance of the TSO-MPC regulator is compared with the performance of other optimization techniques based on proportional-integral-derivative (PID) controllers under different conditions, and the results of the proposed controller outperform other controllers by >10 % of the transient specifications. The transient response of the power system is significantly enhanced with the controller proposed in this study.