Prediksi Band Gap Material Semikonduktor Silikon dengan Menggunakan Machine Learning

MUHAMMAD RIZKY ANUGRAH, . (2024) Prediksi Band Gap Material Semikonduktor Silikon dengan Menggunakan Machine Learning. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

[img] Text
1_Cover.pdf

Download (909kB)
[img] Text
2_Bab I.pdf

Download (354kB)
[img] Text
3_Bab II.pdf
Restricted to Registered users only

Download (575kB) | Request a copy
[img] Text
4_Bab III.pdf
Restricted to Registered users only

Download (321kB) | Request a copy
[img] Text
5_Bab IV.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[img] Text
6_Bab V.pdf
Restricted to Registered users only

Download (192kB) | Request a copy
[img] Text
7_Daftar Pustaka.pdf

Download (395kB)
[img] Text
8_Lampiran.pdf
Restricted to Registered users only

Download (2MB) | Request a copy

Abstract

Material semikonduktor seperti silikon sangat penting dalam teknologi modern, termasuk elektronik, fotovoltaik, dan optoelektronik. Sifat utama semikonduktor adalah band gap, yaitu energi yang diperlukan agar elektron dapat berpindah dari pita valensi ke pita konduksi. Band gap ini mempengaruhi sifat elektronik dan optik material. Dalam penelitian ini, machine learning digunakan untuk memprediksi band gap silikon. Dataset berasal dari Materials Project (MP), yang menyediakan data sifat-sifat material, struktur kristal, dan lainnya. Penelitian ini bertujuan untuk mengembangkan dan mengevaluasi model machine learning untuk memprediksi band gap silikon. Prosesnya melibatkan data extraction, exploratory data analysis (EDA), data cleaning, dan data transformation. XGBoost Regressor digunakan untuk menentukan feature importance, kemudian dilakukan feature selection, dan principal component analysis (PCA) untuk mendapatkan pola pada features yang telah didapatkan. Hasil evaluasi menunjukkan model dapat memprediksi band gap dengan optimal menggunakan semua features dengan metrik seperti R-Squared sebesar 0,770789, MSE sebesar 0,33945, RMSE sebesar 0,582624, dan MAPE sebesar 19,166604. Pengujian dengan data baru menunjukkan model mampu memprediksi band gap dengan cukup optimal. Penelitian ini menunjukkan potensi penerapan machine learning dalam pengembangan material semikonduktor baru. Dengan model yang optimal, proses penemuan dan pengembangan material baru dapat dipercepat, mengurangi waktu dan biaya eksperimen laboratorium, serta mendorong inovasi dalam teknologi material. ***** Silicon semiconductor materials are crucial in modern technology, including electronics, photovoltaics, and optoelectronics. The primary property of semiconductors is the band gap, which is the energy required for electrons to move from the valence band to the conduction band. This band gap influences the electronic and optical properties of the material. In this study, machine learning is used to predict the band gap of silicon. The dataset comes from the Materials project (MP), which provides data on material properties, crystal structures, and more. This research aims to develop and evaluate a machine learning model to predict the band gap of silicon. The process involves data extraction, exploratory data analysis (EDA), data cleaning, and data transformation. The XGBoost Regressor is used to determine feature importance, followed by feature selection and principal component analysis (PCA) to identify patterns in the obtained features. Evaluation results show that the model can optimally predict the band gap using all features, with metrics such as R-Squared of 0.770789, MSE of 0.33945, RMSE of 0.582624, and MAPE of 19.166604. Testing with new data indicates that the model can predict the band gap quite accurately. This research demonstrates the potential of applying machine learning in the development of new semiconductor materials. With an accurate model, the discovery and development process of new materials can be accelerated, reducing the time and cost of laboratory experiments, and fostering innovation in material technology.

Item Type: Thesis (Sarjana)
Additional Information: 1). Dr. Teguh Budi Prayitno, M.Si. ; 2). Haris Suhendar S.Si., M.Sc.
Subjects: Sains > Fisika
Divisions: FMIPA > S1 Fisika
Depositing User: Users 24937 not found.
Date Deposited: 15 Aug 2024 06:00
Last Modified: 15 Aug 2024 06:00
URI: http://repository.unj.ac.id/id/eprint/49381

Actions (login required)

View Item View Item