Embedded Meta evolutionary-firefly algorithm-ANN for multi dg planning in distribution system / Siti Rafidah Abdul Rahim

The depletion of fossil fuel and climate change challenge has gathered worldwide effort to develop sustainable energy systems. Several issues such as energy efficiency, environmental impact and security of supply are the major concerns when dealing with the DG installation. As a result, the penetration of DG in the electricity network will increase and may affect the system. In light of this, various forms of Distributed Generation (DG) technologies have been connected to the system, either to the transmission or distribution system. The installation of DG requires optimisation process to identify the correct location and sizing. Improper sizing and location of DG installation may result to overcompensation or under compensation. Most optimisation techniques are found to face inaccurate and stucked at local minimum phenomena with computationally burdensome. Thus, a reliable optimisation technique is crucial to address this issue. This thesis presents a novel Embedded Meta Evolutionary–Firefly Algorithm-Artificial Neural Network for Multi-DG planning in distribution system. In this study, Meta Evolutionary–Firefly Algorithm (EMEFA) was initially developed to expedite the computational time in multi-DG installation with improved accuracy. Optimal location and sizing are determined using the proposed EMEFA technique. Consequently, a new clustering technique was developed to categorise the DG placement in accordance with different DG types and DG models. Load in the distribution system is divided into three categories i.e. residential, commercial and industrial. These three load types are voltage dependent, and active and reactive power components respond differently to variations in voltage. The voltage dependent load has a main impact on distribution system planning studies. In achieving optimal allocation of DG, two techniques were proposed to study the DG planning which is the ranking identification for DG installation and the integrated clustering development and pre-developed EMEFA was employed.