Optimising distributed generation planning using hybrid artificial intelligence approach

Distributed Generation (DG) has become more recognised in the power sector due to its capability of reducing power loss, having low investments cost and most significantly, able to exploit renewable-energy resources. DG is a small scale source, which is not centrally planned or dispatched. It is gen...

Full description

Saved in:

Bibliographic Details
Main Author:	Tan, Wen Shan
Format:	Thesis
Published:	2013
Subjects:	Q Science
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-utm-ep.42180
record_format	uketd_dc
spelling	my-utm-ep.421802020-08-17T01:01:30Z Optimising distributed generation planning using hybrid artificial intelligence approach 2013 Tan, Wen Shan Q Science Distributed Generation (DG) has become more recognised in the power sector due to its capability of reducing power loss, having low investments cost and most significantly, able to exploit renewable-energy resources. DG is a small scale source, which is not centrally planned or dispatched. It is generally directly connected to the power system near the point of the end users. The optimal siting and sizing of DG are required for maximising the DG potential benefits in a power system. This thesis proposes a multi-objective index-based approach to determine the optimal placement and size of multiple DG in distribution systems, including the voltage rise phenomenon. The proposed approach considers multiple technical aspects, such as the total real power losses, voltage profile, Mega Volt Ampere (MVA) intake by the grid and greenhouse gases emission. Two novel hybrid population-based algorithms: Hybrid Particle Swarm Optimisation-Gravitational Search Algorithm (PSOGSA) and Hybrid Genetic-Firefly Algorithm (GAFA) are introduced in this thesis. An analysis was carried out by using Matlab programming to reveal the validity of the hybrid algorithms on 69 bus systems, for both time invariant and time variant load models, in which the results were obtained by using pre-combined methods. The results show that the proposed algorithms are robust, efficient and capable of solving mixed integer nonlinear optimisation problem. Moreover, the PSOGSA has parallel structure that gave better quality results with lowest multi-objective performance index (MPI) values of 0.4322 and shorter 25 seconds processing time as compared to the sequential structure of GAFA, with lowest MPI values of 0.4439 and 40 seconds processing time. This indicates that placement of multiple DG is more beneficial than the placement of single DG. 2013 Thesis http://eprints.utm.my/id/eprint/42180/ http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:77863 masters Universiti Teknologi Malaysia, Faculty of Electrical Engineering Faculty of Electrical Engineering
institution	Universiti Teknologi Malaysia
collection	UTM Institutional Repository
topic	Q Science
spellingShingle	Q Science Tan, Wen Shan Optimising distributed generation planning using hybrid artificial intelligence approach
description	Distributed Generation (DG) has become more recognised in the power sector due to its capability of reducing power loss, having low investments cost and most significantly, able to exploit renewable-energy resources. DG is a small scale source, which is not centrally planned or dispatched. It is generally directly connected to the power system near the point of the end users. The optimal siting and sizing of DG are required for maximising the DG potential benefits in a power system. This thesis proposes a multi-objective index-based approach to determine the optimal placement and size of multiple DG in distribution systems, including the voltage rise phenomenon. The proposed approach considers multiple technical aspects, such as the total real power losses, voltage profile, Mega Volt Ampere (MVA) intake by the grid and greenhouse gases emission. Two novel hybrid population-based algorithms: Hybrid Particle Swarm Optimisation-Gravitational Search Algorithm (PSOGSA) and Hybrid Genetic-Firefly Algorithm (GAFA) are introduced in this thesis. An analysis was carried out by using Matlab programming to reveal the validity of the hybrid algorithms on 69 bus systems, for both time invariant and time variant load models, in which the results were obtained by using pre-combined methods. The results show that the proposed algorithms are robust, efficient and capable of solving mixed integer nonlinear optimisation problem. Moreover, the PSOGSA has parallel structure that gave better quality results with lowest multi-objective performance index (MPI) values of 0.4322 and shorter 25 seconds processing time as compared to the sequential structure of GAFA, with lowest MPI values of 0.4439 and 40 seconds processing time. This indicates that placement of multiple DG is more beneficial than the placement of single DG.
format	Thesis
qualification_level	Master's degree
author	Tan, Wen Shan
author_facet	Tan, Wen Shan
author_sort	Tan, Wen Shan
title	Optimising distributed generation planning using hybrid artificial intelligence approach
title_short	Optimising distributed generation planning using hybrid artificial intelligence approach
title_full	Optimising distributed generation planning using hybrid artificial intelligence approach
title_fullStr	Optimising distributed generation planning using hybrid artificial intelligence approach
title_full_unstemmed	Optimising distributed generation planning using hybrid artificial intelligence approach
title_sort	optimising distributed generation planning using hybrid artificial intelligence approach
granting_institution	Universiti Teknologi Malaysia, Faculty of Electrical Engineering
granting_department	Faculty of Electrical Engineering
publishDate	2013
_version_	1747816709253758976

Optimising distributed generation planning using hybrid artificial intelligence approach

Similar Items