PREDIKSI DINAMIKA SUHU SENSOR T-ISO1 PADA SISTEM TRANSFER PANAS REAKTOR NUKLIR MENGGUNAKAN RANDOM FOREST DAN DECISION TREE BERBASIS STREAMLIT

Hilda Zakiatun Nufus, . (2025) PREDIKSI DINAMIKA SUHU SENSOR T-ISO1 PADA SISTEM TRANSFER PANAS REAKTOR NUKLIR MENGGUNAKAN RANDOM FOREST DAN DECISION TREE BERBASIS STREAMLIT. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (238kB)
[img] Text
AWAL.pdf
Download (2MB)
[img] Text
BAB 1.pdf
Restricted to Repository UPNVJ Only
Download (256kB)
[img] Text
BAB 2.pdf
Restricted to Repository UPNVJ Only
Download (632kB)
[img] Text
BAB 3.pdf
Restricted to Repository UPNVJ Only
Download (481kB)
[img] Text
BAB 4.pdf
Restricted to Repository UPNVJ Only
Download (1MB)
[img] Text
BAB 5.pdf
Download (240kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (253kB)
[img] Text
RIWAYAT HIDUP.pdf
Restricted to Repository UPNVJ Only
Download (131kB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (13MB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository staff only
Download (1MB)
[img] Text
ARTIKEL KI.pdf
Restricted to Repository staff only
Download (1MB)

Abstract

This research is motivated by the limitations of current nuclear reactor temperature monitoring systems, which remain reactive and may cause delays in detecting abnormal temperature rises. This study aims to develop and evaluate a predictive model for the T-ISO1 sensor temperature in the heat transfer system of a nuclear reactor using Random Forest and Decision Tree algorithms. The planned method involves developing the model using a walk-forward validation approach with an expanding window on time-series data, using the T-ISO2 sensor temperature and time in seconds as input variables. The developed model is then integrated into a Streamlit-based dashboard capable of displaying real-time actual and predicted temperatures, complemented with color-based visual indicators and interactive graphs to facilitate the detection of temperature deviations. The evaluation results show that the Decision Tree model delivers the best performance, achieving an MAE of 0.0092, MSE of 0.00015, RMSE of 0.0122, and an R² of 0.9948, while the Random Forest model also demonstrates strong performance but slightly lower. The superior performance of the Decision Tree in this case is due to the relatively simple data patterns, allowing the model to recognize trends more quickly and accurately. This research successfully addresses the research problem and produces an accurate and efficient reactor temperature prediction system that supports predictive monitoring and early anomaly detection.

Item Type: Thesis (Skripsi)
Additional Information: [No. Panggil: 2110511146] [Pembimbing 1: Didit Widiyanto] [Pembimbing 2: Kharisma Wiati Gusti] [Penguji 1: Widya Cholil] [Penguji 2: Nurhuda Maulana]
Uncontrolled Keywords: Time-Series, Random Forest, Decision Tree, Sistem Transfer Panas Reaktor Nuklir
Subjects: T Technology > T Technology (General)
Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science
Divisions: Fakultas Ilmu Komputer > Program Studi Informatika (S1)
Depositing User: HILDA ZAKIATUN NUFUS
Date Deposited: 06 Aug 2025 07:28
Last Modified: 06 Aug 2025 07:28
URI: http://repository.upnvj.ac.id/id/eprint/37452

Actions (login required)

View Item View Item