1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Defect detection on solar cell using morphology and fuzzy logic techniques
Authors:	Aghamohammadi Amir Hossein (P53661)
Supervisor:	Assoc. Prof. Dr. Anton Satria Prabuwono
Keywords:	Photovoltaic (PV) Solar cells Detection Fuzzy Inference System (FIS) Computer vision.
Issue Date:	12-Apr-2012
Description:	Recently, photovoltaic (PV) systems or solar cells have been widely used as a main renewable energy resource to generate electricity. Generally, electricity energy needs a good quality of solar cell. Physically defects such as crack, broken lines and fragmentation can lead to reduce performance of the PV. Therefore, finding any defects on the solar cell is significantly important. Non-contact, non-destructive and real time inspection system should be developed in order to detect the defects on solar cell. Automated visual inspection system (AVIS) provides those characteristics of the intended system. The general aim of this research is to develop an automated visual inspection system in real time environment specifically for crystalline solar cell. The specific objectives are, to develop image processing technique based on combination of Mathematical Morphology and edge detection operators, to classify the defect types on solar cell using Fuzzy Inference System (FIS), and to evaluate the production rule technique, Mamdani and Sugeno Fuzzy Inference System in classification phase. The proposed AVIS consists of hardware and software frameworks. The hardware framework includes a web-camera, four fluorescent lamps and conveyor belt, and the software framework involves pre-processing and post-processing stages. The captured images of solar cell represented by Region of Interest (ROI) under the sufficient light and fixed speed of conveyor belt. Thresholding, Mathematical Morphology (MM) and edge detection operators are accomplished to segment and highlight the defects on solar cell. Feature extraction is performed to extract the features for classification phase. Finally, production rule, Mamdani and Sugeno Fuzzy Inference System (FIS) are implemented to classify the defects. Experimental result shows that the accuracy rate of the proposed system is 96.54%.,Kini, sistem fotovolta (PV) atau sel suria digunakan secara meluas sebagai sumber tenaga terbaharut utama bagi menjana elektrik. Secara amnya tenaga elektrik memerlukan sel suria yang berkualiti baik. Kecacatan fizikal seperti retak, aliran terputus dan pecah dapat menurunkan prestasi PV. Oleh itu, pemeriksaan sebarang kecacatan pada sel suria adalah sangat penting. Sistem pemeriksaan masa nyata, bukan sentuh, dan bukan memusnah dibangunkan bagi mengesan kecacatan pada sel suria. Sistem pemeriksaan visual automatik menyediakan ciri-ciri sistem berkenaan. Secara amnya objektif penyelidikan ini adalah membangunkan sistem pemeriksaan visual automatik dalam persekitaran masa nyata khasnya bagi sel suria silikon kristal. Objektif spesifik ialah: membangunkan teknik pemprosesan imej berasaskan kombinasi morfologi matematik dan operator pengesan tepi; mengelaskan jenis kecacatan pada sel suria menggunakan sistem inferens kabur (FIS) Mamdani; dan menilai kabur Mamdani, sistem inferens kabur Sugeno dan teknik petua pengeluaran dalam fasa pengelasan. AVIS yang dicadangkan terdiri daripada kerangka kerja perisian dan perkakasan. Kerangka kerja perkakasan termasuk satu kamera-web, empat lampu kalimantang, tali sawat penyampai, dan kerangka kerja perisian termasuk tahap pra-pemprosesan dan pasca-pemprosesan. Imej sel suria yang ditangkap diwakilkan dalam bentuk “Region of Interest (ROI)” di bawah keadaan cahaya yang mencukupi dan tali sawat dengan kelajuan terlaras. Ambang, morfologi matematik (MM) dan operator pengesanan tepi dicapai untuk menembereng dan menyerlahkan kecacatan pada sel suria. Pengekstrakan ciri dijalankan bagi mengekstrak ciri untuk fasa pengelasan. Akhir sekali, petua pengeluaran, Mamdani dan sistem inferens kabur (FIS) Sugeno dilaksanakan untuk mengelas kecacatan. Keputusan ujikaji menunjukkan bahawa kadar ketepatan sistem yang dicadangkan adalah 96.54%,Master
Pages:	164
Call Number:	TA1632 .A365 2012 3
Publisher:	UKM, Bangi
URI:	https://ptsldigital.ukm.my/jspui/handle/123456789/476288
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.