Power system network splitting and load frequency control optimization using ABC based algorithms / Kanendra Naidu a/l Vijyakumar

Vijyakumar, Kanendra Naidu (2015) Power system network splitting and load frequency control optimization using ABC based algorithms / Kanendra Naidu a/l Vijyakumar. PhD thesis, University of Malaya.

Preview

PDF (Full Text) Download (4Mb) | Preview

Abstract

Blackout is the worst case scenario that could occur in the power system network. The cascading failure from one part of the system can propagate throughout the network and cause total system blackout. In order to mitigate the effect of blackout which causes the system to split into infeasible islands, the intentional islanding method is considered. It is applied to split the system into operational smaller island of network until system restoration is carried out. With regards to this, an optimization technique is required to identify the optimum system splitting solution and at the same time ensure the post- islanding frequency stability in the system. This research presents a modified optimization program for the system splitting problem in large scale power system based on Artificial Bee Colony algorithm and graph theory. The post-islanding frequency stability is maintained by optimizing the Load Frequency Control‘s parameters. The graph theory is used to represent the power system based on interconnection between the buses through the adjacency matrix. The edge reduction algorithm is initially applied to reduce the search space to obtain the initial solution. Subsequently, the modified optimization is implemented in order to find the optimum system splitting solution which ensures minimal power flow disruption. The power balance is checked and transmission line power flow analysis is carried out in each island to ensure stability is maintained. In the event the power balance criteria are violated, load shedding scheme is further applied to meet generation-load equilibrium. In the post-islanding phase, the load frequency control for each islanded area is optimally tuned using multi objective ABC optimization technique to maintain nominal system frequency at all times. The online wavelet filter is further implemented to filter out the noise in the LFC model. The IEEE 30-bus, 39-bus and 118-bus test system are chosen to validate the proposed method in MATLAB.

Actions (For repository staff only : Login required)