An improved leader particle swarm optimisation algorithm for solving flexible ac transmission systems optimisation problem in power system
Line outage contingencies in power systems are likely to result in overloads in branches,voltage deviations in buses and excessive power losses. The most common approach for alleviating such consequences is using flexible AC transmission systems (FACTS) devices. Since overloads are the main concern...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/48250/1/FK%202014%2058R.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Line outage contingencies in power systems are likely to result in overloads in branches,voltage deviations in buses and excessive power losses. The most common approach for alleviating such consequences is using flexible AC transmission systems (FACTS) devices. Since overloads are the main concern in line outage contingencies and thyristor-controlled series compensator (TCSC) has proved to be effective in power flow control, it is used for mitigating consequences of line outage contingencies. When FACTS devices are utilised in a power system, they should be allocated optimally. From the optimisation perspective, optimal allocation of FACTS devices is a very complex optimisation problem. Heuristic approaches are the most common approaches for solving FACTS allocation problems. Among heuristics, particle swarm optimisation (PSO) has some advantages which make it popular in solving FACTS allocation roblems.Despite all PSO advantages, it suffers from premature convergence.Due to PSO’s premature convergence, in TCSC allocation problem during contingency, it is not able to find near-global solution. Therefore, considerable amounts of overloads, voltage deviations and power losses are obtained. Although existing approaches mitigatePSO’s premature convergence to some extent, they have some shortcomings that should be addressed carefully.Their explorative capability is not decreased during the run.In most cases, the mutated object is transferred to new position either it leads to lower objective value or not. In most of cases, the mutations are applied to positions or velocities of particles, while applying mutation to the leader may enhance the leader and push all particles toward better regions of search space.They also do not provide any mechanism for jumping out particles after stagnation. The aim of this research is to develop a new PSO variant called improved leader PSO (ILPSO) by addressing the mentioned shortcomings of existing premature convergence mitigation strategies. It should efficiently mitigate premature convergence problem and result in lower amounts of overloads, voltage deviations and power losses (than existing algorithms) in D-TCSC allocation problem during line outage contingencies. The proposed PSO variant features a five-staged successive mutation strategy. The security enhancement problem is formulated as a multi-objective optimisation problem while its objectives are minimising overpowerflows, voltage deviations, power losses and also maximising line utilisation factors. The results of applying improved leader PSO to IEEE 14 bus power system shows its significant outperformance over six other optimisation algorithms including conventional PSO, mutated PSO, enhanced PSO, harmony search,genetic algorithm and gravitational search algorithm. Improved leader PSO leads to lower amount of overpowerflows and power losses with respect to other algorithms, whileit also results in low values of voltage deviation.ILPSO has also been applied to Malaysian 87 bus power system (TNB), wherein overloads and power losses are reduced significantly. |
|---|