Distributed load frequency control of deregulated Nigerian power system

Abstract

In this study, load frequency control (LFC) of the 330kV, 50Hz Nigerian Power System (NPS) is studied. The aim of the work is to develop a control scheme that maintains near-zero steady state errors for deviation in the system frequency and that of net inter-area power flow within an acceptable limit in a deregulated environment. The entire NPS network, comprising of 11 distribution companies (DISCOs) and 8 generation companies (GENCOs) is partitioned into seven control areas (CAs). The control scheme is developed with distributed control architecture. Each of the seven CAs (CA₁, CA₂… CA₇) is equipped with a proportional integral (PI) controllers. These local PI controllers (slave controllers) compute the optimal control signal for appropriate valve positioning of all the generators in its CAs, which in regulate the frequency and tie-line powers. In addition, each of the slave controllers exchange the information about their control action with the neighbouring controllers as well as a central master controller located at the National Control Center (NCC), Osogbo, Osun State. The local PI controllers are optimally tuned using moth flame optimization (MFO) algorithm by minimizing the Integral Square Error (ISE) of the state errors. A model predictive controller (MPC) applied as the master controller is used to establish the optimal set-points of the slave PI controllers. The effectiveness of the developed scheme is verified by implementing it on the seven-CA Nigerian deregulated power system perturbed with a step load demand. From the stimulations carried out in MATLAB environment, it is established that the developed scheme is not capable of maintaining near-zero steady state errors for deviation in system frequency and tie-line powers but outperformed the present conventional control scheme in optimality and stability.