Solar photovoltaic system fault identification using artificial intelligence

Mohamad Nor, Mohamad Syahmi Damia (2019) Solar photovoltaic system fault identification using artificial intelligence. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Preview

Text (Full Text) Solar photovoltaic system fault identification using artificial intelligence.pdf - Submitted Version Download (5MB) | Preview

Abstract

Solar photovoltaic system is design to generate electricity and operate reliably over the entire life of the product. Despite this, there are still failures could occur that can affect the performance of the product. Power failures could be one of the causes. The causes of fault could be from lightning, natural disaster, animal etc. The most advanced solutions, such as expert systems are related to knowledge-based systems. However,the subject field is experiencing congestion as it is unable to learn or adapt to new situations. This thesis is dedicated to implement artificial neural networks (ANNs) for fault identification at the solar photovoltaic system. ANNs is a computing system that inspired from biological neural network such as human brain and has ability to extract significant links from data presented. In principle, ANNs can remove the limitation of expert system as it has adaptive structure. Back propagation neural network (BPNN)was used in this project as it has simplest form but effective. Supervised learning is the training method that been used. The results from this thesis demonstrate that BPNN have high accuracy and give good performance in fault identification at solar photovoltaic system.

Actions (login required)

View Item