IPB University Logo

SCIENTIFIC REPOSITORY

IPB University Scientific Repository collects, disseminates, and provides persistent and reliable access to the research and scholarship of faculty, staff, and students at IPB University

AI Repository
 
Building and Categories


      View Item 
      •   IPB Repository
      • Final Assignments
      • Undergraduate Final Assignments
      • UF - Vocational School
      • UF - Computer Engineering Tehcnology
      • View Item
      •   IPB Repository
      • Final Assignments
      • Undergraduate Final Assignments
      • UF - Vocational School
      • UF - Computer Engineering Tehcnology
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Pembangunan Sistem Kendali Lingkungan Prediktif Melalui Implementasi Long Short Term Memory pada Gudang Alat Kesehatan

      Thumbnail
      View/Open
      Cover (514.4Kb)
      Fulltext (2.453Mb)
      Lampiran (510.2Kb)
      Date
      2026
      Jenis/Type
      Tugas Akhir
      Subtype
      Undergraduate Theses
      Author
      Kharina, Aura
      Marcelita, Faldiena
      Metadata
      Show full item record
      Abstract
      Medical device storage requires proper environmental control to maintain device quality and functionality. Conventional reactive control systems are limited by the fact that actuators only operate after critical conditions occur. This research aims to develop a predictive environmental control system based on the Internet of Things (IoT) and Long Short Term Memory (LSTM) that can monitor temperature, humidity, and CO2 in real time and predict conditions 15 minutes in advance to proactively activate actuators. The IoT system is built using an ESP32 microcontroller with SHT31 and MQ135 sensors that show a reading accuracy of 97–99%. The LSTM model was trained using 138.892 historical data for 27 days with a sliding window technique and evaluated using RMSE and MAE metrics which resulted in low prediction error rates for all three environmental parameters, namely RMSE of 0,2318°C and MAE of 0,1865°C for temperature, RMSE of 1,2362% and MAE of 1,0514% for humidity, and RMSE of 22,1802 ppm and MAE of 12,5938 ppm for CO2. Testing of the 3-level control system showed 100% success in 25 scenarios with an average response time of 1,24 seconds. After the system was operated, the maximum humidity value decreased from 83,20% to 76,51% and the standard deviation of CO2 decreased from 64,48 to 3,35 which proved that the predictive control system was successful in maintaining the environmental stability of the medical equipment warehouse.
       
      Penyimpanan alat kesehatan memerlukan pengendalian kondisi lingkungan gudang alat kesehatan yang tepat untuk menjaga kualitas dan fungsionalitas alat. Sistem kendali konvensional yang bersifat reaktif memiliki keterbatasan karena aktuator baru bekerja setelah kondisi kritis terjadi. Penelitian ini bertujuan membangun sistem kendali lingkungan prediktif berbasis Internet of Things (IoT) dan Long Short Term Memory (LSTM) yang mampu memantau suhu, kelembapan, dan CO2 secara real-time serta memprediksi kondisi 15 menit ke depan untuk mengaktifkan aktuator secara proaktif. Sistem IoT dibangun menggunakan mikrokontroler ESP32 dengan sensor SHT31 dan MQ135 yang menunjukkan akurasi pembacaan 97–99%. Model LSTM dilatih menggunakan 138.892 data historis selama 27 hari dengan teknik sliding window dan dievaluasi menggunakan metrik RMSE dan MAE yang menghasilkan tingkat kesalahan prediksi yang rendah untuk ketiga parameter lingkungan, yaitu RMSE sebesar 0,2318°C dan MAE sebesar 0,1865°C untuk suhu, RMSE sebesar 1,2362% dan MAE sebesar 1,0514% untuk kelembapan, dan RMSE sebesar 22,1802 ppm dan MAE sebesar 12,5938 ppm untuk CO2. Pengujian sistem kendali 3-level menunjukkan keberhasilan 100% pada 25 skenario dengan waktu respons rata-rata 1,24 detik. Setelah sistem dioperasikan, nilai maksimum kelembapan turun dari 83,20% menjadi 76,51% dan standar deviasi CO2 turun dari 64,48 menjadi 3,35 yang membuktikan sistem kendali prediktif mampu menjaga kestabilan lingkungan gudang alat kesehatan.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/178521
      Collections
      • UF - Computer Engineering Tehcnology [239]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository