Please use this identifier to cite or link to this item:
http://studentrepo.iium.edu.my/handle/123456789/11654
Title: | Application of coactive-anfis to predict Micro-EDM performances | Authors: | Wan Azhar Wan Ahmad | Supervisor: | Tanveer Saleh, Ph.D Mohd Asyraf Mohd Razib, Ph.D |
Subject: | Microelectromechanical systems -- Computer simulation | Year: | 2022 | Publisher: | Kuala Lumpur : Kulliyyah of Engineering, International Islamic University Malaysia, 2022 | Abstract in English: | Micro Electrical Discharge Machining (μEDM) is one of the most demanding manufacturing processes available today. The selection of μEDM parameters remains a challenge since it is frequently based on machinist intuition and heuristic approaches. In recent years, soft computing and artificial intelligence have been used to model and predict the μEDM machining process. However, artificial intelligence has not been established for predicting μEDM performances based on material properties. Therefore, this research proposed a model that considers the material properties, such as thermal conductivity, melting point, and electrical resistivity. Since μEDM is a non-linear and stochastic process, Coactive Neuro-Fuzzy Inference Systems (CANFIS) was proposed to model and predict the multiple μEDM performances on various materials. The material properties, feed rate, capacitance, and gap voltage are the input parameters in a three-level design based on a full factorial experiment. The CANFIS model can accurately predict the material removal rate (MRR), total discharge pulse, overcut, and taperness in a single model. The mean average percentage error (MAPE) from the model prediction for test dataset of various outputs such as MRR, total discharge pulse, overcut and taper angle were found to be 9.5% (90.5% accuracy), 8.9% (91.1% accuracy),16.9% (83.1% accuracy) and 15.7% (84.3% accuracy) respectively. This research proposes a novel approach in modelling and predicting μEDM performances by considering workpiece’s materials using artificial intelligence. Mikro Pemesinan Nyahcas Elektrik (μEDM) adalah merupakan salah satu proses pembuatan yang mendapat permintaan yang tinggi pada ketika ini. Pemilihan parameter μEDM masih lagi menjadi cabaran kerana ia sering ditentukan berdasarkan gerak hati juruteknik dan pendekatan heuristik yang mencabar untuk dimodelkan. Kebelakangan ini, pengkomputeran lembut dan kecerdasan buatan banyak digunakan untuk memodel dan meramal proses μEDM. Walau bagaimanapun, kecerdasan buatan belum lagi digunakan untuk meramal prestasi μEDM berdasarkan kepada sifat bahan. Oleh itu, penyelidikan ini mencadangkan suatu model yang mengambil kira sifat bahan, seperti kekonduksian terma, takat lebur dan daya tahan elektrik. Oleh kerana μEDM ialah proses yang tidak linear dan stokastik, sistem inferens neuro-kabur koaktif (CANFIS) telah diketengahkan untuk memodel dan meramal pelbagai prestasi μEDM untuk pelbagai jenis bahan. Sifat bahan, kadar suapan, kapasitansi dan jurang voltan adalah dipilih sebagai parameter input yang direka bentuk dalam tiga peringkat berdasarkan eksperimen berfaktorial penuh. Model CANFIS ini berjaya meramal dengan tepat prestasi μEDM yang terdiri daripada kadar penyingkiran bahan (MRR), jumlah pulsa nyahcas, lebihan potongan dan ketirusan hanya dengan satu model. Peratusan ralat min mutlak (MAPE) daripada ramalan model untuk data ujian bagi pelbagai prestasi seperti MRR, jumlah pulsa nyahcas, lebihab potongan dan ketirusan masing-masing mencatatkan 9.5% (90.5% ketepatan), 8.9% (91.1% ketepatan), 16.9% (83.1% ketepatan) and 15.7% (84.3% ketepatan). Penyelidikan ini mencadangkan pendekatan baharu dalam memodel dan meramal prestasi μEDM dengan mengambil kira sifat bahan mengunakan kecerdasan buatan. |
Call Number: | t TK 7875 W244A 2022 | Kullliyah: | Kulliyyah of Engineering | Programme: | Master of Science in Engineering | URI: | http://studentrepo.iium.edu.my/handle/123456789/11654 |
Appears in Collections: | KOE Thesis |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
t11100487747WanAhamdBinWanAzhar_24.pdf | 24 pages file | 3.32 MB | Adobe PDF | View/Open |
t11100487747WanAhamdBinWanAzhar_SEC.pdf Restricted Access | Full text secured file | 18.11 MB | Adobe PDF | View/Open Request a copy |
Google ScholarTM
Check
Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated. Please give due acknowledgement and credits to the original authors and IIUM where applicable. No items shall be used for commercialization purposes except with written consent from the author.