Optimization Techniques in Engineering

14Performance of P, PI, PID, and NARMA Controllers in the Load Frequency Control of a Single-Area Thermal Power Plant

Ranjit Singh^* and L. Ramesh

Dr. M.G.R. Educational and Research Institute, Chennai, India

Abstract

This chapter investigates the load frequency control (LFC) based on a neural network for improving power system dynamic performance of a single-area thermal power plant. The performance of P, PI, PID, and NARMA-L2 controllers is studied in terms of their capability to reduce the frequency deviations to zero. NARMA-L2 controller is tuned with different epochs to best tune the system and with the application of PID controller, an error has been minimized quickly. NARMA-L2 controller is found to overcome the control problems of P, PI, and PID controllers due to the input multiplicities. Necessary graphs are drawn in order to prove the superiority of the NARMA-L2 controller. MATLAB/SIMULINK software is used in this chapter.

Keywords: Frequency deviations, NARMA-L2 controller, artificial neural network, load frequency control, automatic generation control

14.1 Introduction

Small signal stability is defined as that property of the power system which maintains its synchronism under small disturbances. Due to continuous variations in the power generation and load demand, disturbances occur in the power system [1].

Due to these disturbances, oscillations occur in the power system, which must be damped to maintain the system stability. Instability can be of two forms: ...