Claim Missing Document
Check
Articles

Found 14 Documents
Search

Perancangan Program Pengestimasi Probabilitas Kegagalan Peralatan Penukar Panas Akibat Korosi Seragam Berbasis Deep Neural Network Fatriansyah, Jaka Fajar; Dhaneswara, Donanta; Hanifa, Muthia; Hartoyo, Fernanda; Pradana, Agrin Febrian; Anis, Muhammad; Fauzi, Andrian
Syntax Literate Jurnal Ilmiah Indonesia
Publisher : Syntax Corporation

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (345.943 KB) | DOI: 10.36418/syntax-literate.v8i3.11486

Abstract

Meningkatnya standar keamanan dan ketatnya persaingan antar perusahaan meningkatkan kebutuhan bagi suatu perusahan untuk mengendalikan kegagalan pada peralatan. Inspeksi secara teratur dilakukan sebagai bagian dari rangkaian pemeliharaan dan manajemen integritas peralatan. Dalam merencanakan dan melakukan inspeksi, diperlukan strategi yang tepat agar inspeksi yang dilakukan tepat sasaran dan sesuai dengan kebutuhan. Risk-based inspection merupakan teknik pengambilan keputusan dalam perencanaan pemeliharaan yang berdasar pada risiko. Pada saat ini, penggunaan metode-metode kecerdasan buatan untuk kegiatan penilaian risiko, pemodelan konsekuensi, dan perencanaan pemeliharaan telah dilakukan. Penelitian ini bertujuan untuk mengembangkan suatu program yang memanfaatkan pembelajaran mesin dan kecerdasan buatan untuk melakukan penilaian salah satu komponen risiko yaitu probabilitas kegagalan (Probability of Failure, PoF) pada bagian cangkang dalam peralatan penukar panas menggunakan deep learning. Model ini dapat membantu operator yang bekerja di bidang minyak dan gas untuk menentukan tingkatan risiko sehingga inspeksi dapat dilakukan dengan lebih efisien dan terarah. Penelitian ini menghasilkan sebuah program dan disain program pembelajaran mesin berbasis deep learning yang digunakan untuk memprediksi risiko kegagalan akibat korosi seragam pada peralatan sisi dalam cangkang peralatan penukar panas cangkang dan buluh (shell-and-tube heat exchanger) berdasarkan standar API 581 dengan akurasi sebesar 89% yang didapatkan dengan parameter-parameter diantaranya learning rate sebesar 0.001, epoch sebesar 150, random state sebesar 60, tiga hidden layer, dan test size sebesar 0.2.
Remaining Life Assessment and Fitness for Service Evaluation of Aging Chemical Reactors in Polyethylene Terephthalate Resin Industry Munthe, Aditya Pahlawan; Dhaneswara, Donanta; Putra, Wahyuaji Narottama; Widyaputra, Gama
Journal of Materials Exploration and Findings Vol. 4, No. 2
Publisher : UI Scholars Hub

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Aging chemical reactors in the polyethylene terephthalate (PET) resin industry require comprehensive evaluation to ensure safe continued operation. This study conducts a remaining life assessment (RLA) and fitness-for-service (FFS) evaluation on five 30-year-old reactors, based on API 510, API 579/ASME FFS-1, and ASME BPVC Section VIII Div. 1 standards. The analysis involves corrosion rate measurement, future corrosion allowance (FCA) projection, and minimum thickness verification. Among the reactors, R-120 was found to have the shortest remaining life less than 15 years. FFS assessments using three criteria Average Measured Thickness, MAWP from Point Thickness Readings, and Minimum Measured Thickness confirm that R-120 meets all safety requirements. The head and shell thicknesses exceed the minimum allowable values; calculated MAWPr values are above the design MAWP; and thicknesses adjusted for FCA remain above 50% of the minimum required. These results indicate that R-120 remains fit for continued service. This study underscores the critical role of standardized assessment and routine inspection in extending the safe operating life of aging process equipment
Spin Coater Design with PID Algorithm Using Polynomial Regression Approach and Bias Tuning for TiO2 Deposition Process Andika, Geo Surya; Sofyan, Nofrijon; Dhaneswara, Donanta; Yuwono, Akhmad Herman
Journal of Materials Exploration and Findings Vol. 4, No. 1
Publisher : UI Scholars Hub

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The thin-film deposition technique using spin coating offers a cost-effective alternative to Chemical Vapor Deposition (CVD) and Physical Vapor Deposition (PVD). The spin-coating process requires precise control of the motor drive system to ensure that the rotational speed, measured in rotations per minute (RPM), aligns with the set point and remains stable. This study presents the design and development of a spin coater prototype to achieve uniform thin-film deposition. The control method employed utilizes a Proportional-Integral-Derivative (PID) algorithm, incorporating a polynomial approach with bias tuning. The PID control was chosen to achieve stable operation in a non-linear system. The performance of the non-linear PID control system is compared with an open-loop control system by evaluating the overshoot behavior. In the first experiment, a proximity sensor was tested to measure the spin coater motor's speed in an open-loop control configuration. The performance was evaluated using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) metrics, with results showing an MAE of 1358.6 RPM and a MAPE of 23.13% compared to a tachometer. In the second experiment, step-response testing was conducted using a closed-loop PID control system with a polynomial approach and bias tuning. Compared to the open-loop system, the closed-loop PID controller reduced overshoot to less than 3%. The RPM deviation between the spin coater and the tachometer was limited to range, approaching ideal conditions. The closed-loop control was tested within the 5000–9000 RPM range, where stable RPM regulation resulted in more uniform TiO2 thin-film distribution on glass substrates. This study highlights the effectiveness of closed-loop PID control in achieving precise rotational control, which is essential for enhancing the quality of thin-film deposition.
Characteristics of Carbonaceous Materials Synthesized from Palm Oil Empty Fruit Bunch Waste Using Ferrocene Catalyst Shahab, Ahmad Nabil; Islam, Adinda Izzatul; Wardana, Afif; Yahya, Ilham Nur Dimas; Amalia, Ary Yanuar Tri; Sofyan, Nofrijon; Dhaneswara, Donanta
Journal of Materials Exploration and Findings Vol. 4, No. 1
Publisher : UI Scholars Hub

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The synthesis of carbonaceous materials for reduced graphene oxide (rGO) precursors using oil palm empty fruit bunches waste as a carbon source offers a sustainable solution for waste management in the palm oil industry while delivering high-performance materials. The oil palm empty fruit bunches were carbonized, followed by acid washing, pyrolysis with ferrocene (8%, 12%, and 16% variations), and ultrasonication to produce rGO. The structural, electronic, and morphological properties of the rGO were analyzed using various characterization techniques. The band gap values decreased with increasing ferrocene concentration, from 1.14 eV (8%) to 1.06 (16%), indicating enhanced electronic conductivity. XRD analysis revealed a crystal size increase from 11.3 nm (8%) to 181 nm (16%), while Raman spectroscopy showed a consistent D to G intensity ratio of 0.85, indicating reduced structural defects. SEM-EDS results demonstrated a carbon to oxygen atomic ratio of 4.38 (8%), 3.79 (12%), and 3.77 (16%), confirming successful reduction and improved carbon content. These finding highlight the potential of rGO synthesized from oil palm empty fruit bunches for applications in semiconductors, energy storage, and gas sensing, offering an innovative approach to sustainable materials development.