Thyristor Controlled Reactor with Different Topologies Based on Fuzzy Logic Controller

Abstract

This paper studies improving the power factor by using Thyristor Controlled Reactor (TCR) shunted with a fixed capacitor to get a controlled source of reactive power. This technique has three topologies of TCR connection like Star, Delta, and Hybrid. The fuzzy controller is used to control the thyristor firing angle according to the reactive power measured at the PCC to improve the power factor. Also, the paper discusses the using of the hybrid technique to solve the problem of Total Harmonic Distortion (THD) as one side effect of the using the TCR system. The data from the statistics refer that the most plant loads are inductive loads like motors, transformers, and arc furnaces which require a magnetic field to operate. The magnetic field is necessary for these load types, the nature of these loads cannot be altered but the problems arising from these load types can be overcome. Increasing the reactive power will decrease the power factor and consequently will decrease the capacity of the electrical network. Improving power factor can be achieved by using conventional techniques or by using advanced techniques. This paper introduces the TCR system as advanced technique to compensate the reactive power smoothly. Actually, the TCR system is not exist as a standalone compensator but combined with other configuration such as FC, TSC, MSC, and the filter. The TCR branch is responsible for a positive reactive power while the other integrated branches with the TCR system are used for producing the negative reactive power (1-4) as shown in the Fig 1. As for controllers, they can be implemented by conventional controller like PI and PID or by using advanced controller like a fuzzy logic controller. The comparison between the two previous types is shown in many literatures. The question attends why the fuzzy logic controller?, the answer mainly comes from that the fuzzy controller is preferred upon the conventional controllers for the following reasons; there is no need for mathematical model, its rule are based on the human experience so it is easy to be understood, flexibility to be changeable, dealing with linear and nonlinear systems, and behaves like traditional controllers (5).