Arini, Rahmadian Putri (2024) Sistem Pemantauan dan Prediksi Terhadap Suhu dan Kelembaban Pada Ruang Penyimpanan Tembakau Menggunakan Long Short-Term Memory. Diploma thesis, Institut Teknologi Sepuluh Nopember.
Text
2040201068-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (5MB) | Request a copy |
|
Text
2040201068-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (5MB) | Request a copy |
Abstract
Penelitian ini berfokus pada penyelesaian tantangan dalam penyimpanan tembakau, khususnya dalam memastikan suhu dan kelembaban yang ideal. Fluktuasi suhu dan kelembaban mampu mempengaruhi secara langsung kualitas dari daun tembakau. Untuk mengatasi masalah ini, pada penelitian ini mengusulkan penggunaan metode Long Short-Term Memory (LSTM) untuk memprediksi suhu dan kelembaban berdasarkan data historis dan tren saat ini. Dengan adanya prediksi memungkinkan operator untuk mendapatkan rekomendasi harus melakukan penyesuaian kondisi sebelum lingkungan berubah, sehingga suhu dan kelembaban tetap stabil dan kualitas tembakau terjaga. Dilakukan penelitian untuk memprediksi Suhu dan Kelemababan 1 Jam, 3 Jam, dan 5 Jam kedepan yang digunakan untuk menghasilkan rekomendasi yang tepat. Hasil penelitian menunjukkan bahwa hasil prediksi 1 jam memiliki hasil yang yang terbukti efektif dalam memprediksi suhu dan kelembaban dengan tingkat akurasi yang tinggi, ditunjukkan oleh nilai R-Squared 0.97733 untuk kelembaban dan 0.75751 untuk suhu. Sehingga penelitian ini memberikan hasil yang signifikan dalam pengembangan model prediksi deret waktu yang efektif dan mampu membantu industri tembakau dalam menghadapi tantangan masa depan.
===================================================================================================================================
This research focuses on solving challenges in tobacco storage, particularly in ensuring ideal temperature and relative humidity. Fluctuations in temperature and relative humidity can directly affect the quality of tobacco leaves. To address this issue, this research proposes the use of Long Short-Term Memory (LSTM) method to predict temperature and relative humidity based on historical data and current trends. The prediction allows the operator to get recommendations to adjust the conditions before the environment changes, so that the temperature and relative humidity remain stable and the quality of tobacco is maintained. Research was conducted to predict the Temperature and Relative Humidity 1 hour, 3 hours, and 5 hours ahead which is used to produce appropriate recommendations. The results showed that the 1-hour prediction results had results that proved effective in predicting temperature and relative humidity with a high level of accuracy, indicated by the R-Squared value of 00.97733 for relative humidity and 0.75751 for temperature. Thus, this research provides significant results in the development of an effective time series prediction model that can help the tobacco industry in facing future challenges.
Item Type: | Thesis (Diploma) |
---|---|
Uncontrolled Keywords: | tobacco, temperature and humidity, Long Short-Term Memory (LSTM), tembakau, suhu dan kelembaban |
Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control |
Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
Depositing User: | Rahmadian Putri Arini |
Date Deposited: | 17 Sep 2024 08:13 |
Last Modified: | 17 Sep 2024 08:13 |
URI: | http://repository.its.ac.id/id/eprint/115644 |
Actions (login required)
View Item |