Artificial neural network and fuzzy logic control for HVDC systems

Abstract

This thesis comprises the studies and results from the application of artificial neural networks (ANN) and fuzzy logic theory to the control of the high voltage direct current (HVdc) systems. The studies considered their implementation in both low and high level control systems in HVdc systems. The study is verified using the electromagnetic transients simulation software. The results demonstrated successful performance for single mode control (either constant extinction angle or constant current) using an ANN based on-line controller. The results for the fuzzy logic based controller showed many improvements compared to the conventional HVdc control scheme. The fuzzy logic based controller concept was further successfully extended to high level control problems such as the control of SSR and power swings. Finally in order to facilitate further application of new control techniques such as ANN or fuzzy logic, a MATLAB to transient simulation soft-ware interface was developed. Using this interface, all the MATLAB commands and Tool-boxes may be used within the transient simulation software.