Optimal network reconfiguration and intelligent service restoration prediction technique based on Cuckoo search spring algorithm / Mohamad Izwan Zainal
Modern distribution network is a complex network due to load demand varies and normally designed in radial circuit in which the circuit has ready to close switch at the end of the circuit when the situation needed. This switch will change the condition after distribution network triggers any abnorma...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://ir.uitm.edu.my/id/eprint/78551/1/78551.pdf https://ir.uitm.edu.my/id/eprint/78551/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Modern distribution network is a complex network due to load demand varies and normally designed in radial circuit in which the circuit has ready to close switch at the end of the circuit when the situation needed. This switch will change the condition after distribution network triggers any abnormal during operating system. The performance of the distribution network is very important, and it is characterized by some measurable item such as voltage profile and losses, to evaluate the actual value comply with the system needs. In this research, Cuckoo Search Spring Algorithm (CSSA) is proposed to enhance the robustness of algorithm by constructing the optimal network reconfiguration consist of reducing power losses and improve voltage profile with the various loadability factor as the constraint according to load profile, based on single and multiobjective model. The performance of single and multiobjective CSSA optimization were compared to initial conditions and the value of power losses, voltage profile, number of switching involved, and convergence curve were obtained. In addition, objective function using the same CSSA algorithm were applied i.e., Vmin and Ploss as the objective function, and multi-objective involves Vmin and Ploss as the objective function. From the analysis, the multiobjective CSSA (MOCSSA) had yielded better optimal solutions with faster convergence time as compared to the other two algorithms with improvement of voltage profile and losses minimisation. Later, this MOCSSA is applied on service restoration to ensure the proposed algorithm technique is suitable in selecting the optimal switches for supply recovery after the line section is isolated from the system either by forced outage or planned outage purposes. Furthermore, the isolation line is tested on different section and load factor to recognize the improvement of optimal distribution network performance. These output data are analysed to present the optimal prediction output of service restoration using Cuckoo Search Spring – Least Square Support Vector Machine (CSS-LSSVM). Finally, a novel hybrid CSS-LSSVM was presented, and the result showed better prediction performance compared to classical LSSVM |
---|