The simulation model of API oil separator based on artificial neural network at Pt Saka Indonesia Pangkah limited

I. Putu Eka Widya Pratama*, Totok Soehartanto, Abdussalam Darmaatmaja, Safira Firdaus Mujiyanti

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

The research was designed an artificial neural network (ANN) to predict the output of API oil Separator at PT Saka Indonesia Pangkah Limited. This ANN has a feedforward structure and uses a Levenberg-Marquardt algorithm. It has three input nodes in the input layer of pH, temperature, and total suspended solids (TSS), one output node (TSS) in the output layer, and a hidden node that varies until twenty hidden layers. The simulation result shows that the best value of root mean square error (RMSE) lies in a neural network of 3-9-1 structure (Input layer-Hidden layer-Output layer) with a value of 32.65. Also, our ANN structure has a good result in estimating TSS output of the API Separator with a value of 14.16 - 169.12 mg/l compared to the output data of the API Separator in the range of 18 - 136 mg/l.

Original languageEnglish
Title of host publicationEngineering Physics International Conference 2021, EPIC 2021
EditorsAyodya Pradhipta Tenggara, Nur Abdillah Siddiq, Sita Gandes Pinasti, Rizki Insyani, Jundika Candra Kurnia, Geetali Saha, Rasool Moradi-Dastjerdi
PublisherAmerican Institute of Physics Inc.
ISBN (Electronic)9780735444843
DOIs
Publication statusPublished - 22 May 2023
Event3rd Engineering Physics International Conference, EPIC 2021 - Yogyakarta, Virtual, Indonesia
Duration: 24 Aug 202125 Aug 2021

Publication series

NameAIP Conference Proceedings
Volume2580
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference3rd Engineering Physics International Conference, EPIC 2021
Country/TerritoryIndonesia
CityYogyakarta, Virtual
Period24/08/2125/08/21

Keywords

  • API Separator
  • Artificial Neural Network (ANN)
  • Total Suspended Solids (TSS)

Fingerprint

Dive into the research topics of 'The simulation model of API oil separator based on artificial neural network at Pt Saka Indonesia Pangkah limited'. Together they form a unique fingerprint.

Cite this