UPM Institutional Repository

Genetic algorithm-based optimal overcurrent relays coordination for standalone sustainable hydrokinetic renewable energy distribution network


Citation

Ahmad, Saiful Zuhaimi (2019) Genetic algorithm-based optimal overcurrent relays coordination for standalone sustainable hydrokinetic renewable energy distribution network. Masters thesis, Universiti Putra Malaysia.

Abstract

The Standalone Sustainable Hydrokinetic Renewable Energy Distribution Network (SHRE-DN) system is a very unique distribution system. SHRE-DN is an off grid system and use a hydrokinetic concept which turn the river stream into a power generation source. The development SHRE-DN is to develop rural electrification system for native long houses along the river. The research study is tested on a DigSILENT develop model of the SHRE-DN and in accordance with all respectively unique parameters and relevant standards such as IEC or IEEE standard which is compulsory to comply for the relay protection scheme. The most commodious protection devices are overcurrent relays (OCRs) which responsible to isolate and clear a fault occurred into the distribution network. Generally, OCRs work in pairs knows as primary relay and backup relay. The primary relay must trip first in order to clear and isolate the fault accordingly. In the event of the primary relay is fail to trip or malfunction, the backup up relay must then took a place to trip and clear the faults. These relay must be set precisely so that the fault clearance can be done as shortest possible time to avoid undesirable tipping of the relays. Since SHRE-DN is a new standalone distribution system, an efficient and properly coordinated overcurrent protection system must be provided and it poses a great challenge to protection coordination scheme setup, due to the unique network topology. Improper and miscoordination among OCRs can result in maloperation of the protection system that can lead to false tripping and an unnecessary outage and power system instability. Thus, the objective of this work is to employ Genetic Algorithm (GA) technique in Matlab/Simulink for optimal overcurrent coordination and settings among all OCRs in the SHRE-DN in order to improve the speed of OCR tripping operation. OCRs depend on its Time Dial Setting (TDS) values which effect the operating time of the relays. This research work propose the artificial intelligent (AI) solution on the objective function (OF) formulation, with the application of genetic algorithm (GA) optimization solver, to determine each relay best optimal operation for the TDS value and response time to fault accordingly and also eliminate miscoordination among the relays. GA are good at taking larger, potentially huge, search space and navigating them looking for optimal combinations of things and solution. Furthermore, the project is fast track and requiring the simplest method available. In this strategy, all TDS values belonging to the respective relays are given to the algorithm in order to get the optimized value of the TDS. The obtained optimized TDS Values from the GA Optimization Technique in Matlab/Simulink Toolbox produce about 20% and almost 52% improvement for each OCRs respectively. Overall improvement of the operating time for the protection scheme for the distribution network is about 36% of an improvement as compared to the conventional technique and approach in OCRs coordination and setting for the distribution network. Thus, the objective of this work to provide the most proper and efficient for the SHRE-DN is successfully achieved.


Download File

[img] Text
FK 2019 139 - ir.pdf

Download (1MB)

Additional Metadata

Item Type: Thesis (Masters)
Subject: Renewable energy sources
Subject: Smart power grids
Call Number: FK 2019 139
Chairman Supervisor: Mohammad Lutfi Othman, PhD, PEng
Divisions: Faculty of Engineering
Depositing User: Ms. Nur Faseha Mohd Kadim
Date Deposited: 27 Jan 2021 03:26
Last Modified: 04 Jan 2022 00:52
URI: http://psasir.upm.edu.my/id/eprint/84379
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item