Neural network and fuzzy logic control of unbalanced operation of Static Var Compensators | |
| Author | Anawat Puangpairoj |
| Call Number | AIT Thesis no. ET-01-14 |
| Subject(s) | Neural networks (Computer science) Fuzzy logic Voltage regulators |
| Note | A thesis submitted in a partial fulfillment of the requirements for the degree of Master of Engineering, School of Environment Resources and Development |
| Publisher | Asian Institute of Technology |
| Abstract | Static Var Compensators (SVCs) are being increasingly employed in modern power systems. They improve power system stability and reduce losses by compensating reactive power of fluctuating heavy industrial loads and regulating terminal. As SVC essentially consists of a Thyristors Switched Capacitor (TSC) to supply reactive power to the system and a Thyristor Controlled Reactor (TCR) to absorb reactive power from the system. However, operation of SVCs introduces harmonic currents into the power system. These harmonics are generally removed using external filters. But in recent years, several approaches have been proposed for minimizilig the harmonic generation by appropriate controlling of SVC This thesis examines the optimal control of SVCs using Fuzzy Logic and Artificial Neural Nenvork (ANN) techniques. The main goal of the controller is to balance and minimize the reactive power drawn from source while lowering the harmonic generation by controlling the firing angles of the Thyristor Controlled Reactors (TCR) of the SVC An algorithm for computing reactive power compensation requirements at a given load condition and the corresponding TSC step and the TCR firing angles is presented. The algorithm also can compute the harmonic current injection at that particular SVC operating points. It was shown that there are certain operating points which results in lower harmonic level. The Fuzzy logic based optimal controller proposed in this study evaluates various possible operating points in term of the amount of reactive power drawn from the supply and the harmonic injection level. It then selects the best operating point. Since the Fuzzy logic based optimal controller need considerable processing time, it is proposed to improve the speed of calculation by using a set of neural network to evaluate the function of Fuzzy logic based controller. The proposed ANN based controller takes three phase reactive power demand as an input and estimates the required TSC step and TCR firing angles to achieve the optimal operation. The ANNs was trained using simulated data obtained with the Fuzzy logic based controller. Pe1formance of the Fuzzy logic based and ANN based SVC controllers were tested using three different load profiles; highly unbalanced, slightly unbalanced and actual arc furnace. The comparative test results are presented. |
| Year | 2001 |
| Type | Thesis |
| School | School of Environment, Resources, and Development (SERD) |
| Department | Department of Energy and Climate Change (Former title: Department of Energy, Environment, and Climate Change (DEECC)) |
| Academic Program/FoS | Energy Technology (ET) |
| Chairperson(s) | Surapong Chirarattananon ;Rajapakse, Athula |
| Examination Committee(s) | Weerakorn Ongsakul ; Singh, Sri Niwas |
| Scholarship Donor(s) | H.M. Queen's scholarship |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2001 |