Skripsi

PREDIKSI DURASI PENYIRAMAN OTOMATIS PADA TANAMAN TOMAT BEEF (LYCOPERSICUM ESCULENTUM MILL) MENGGUNAKAN ALGORITMA LONG SHORT-TERM MEMORY DALAM SISTEM FARMBOT

---

Beef tomato (Lycopersicum esculentum Mill) is a crop that requires stable water availability to support growth and prevent water stress or diseases caused by excessive or insufficient watering. Manual irrigation is often inefficient and leads to uncontrolled soil moisture levels. The optimal soil moisture level for tomatoes is 60%, thus a system capable of predicting and regulating irrigation more precisely is needed. This study develops an automatic irrigation system for Farmbot based on soil moisture threshold control and duration prediction of moisture increase using the Long Short-Term Memory (LSTM) algorithm. The system incorporates a soil moisture sensor as the primary data source, supported by DHT22 and DS18B20 sensors for monitoring environmental conditions. The soil moisture data were processed and used to train the LSTM model to estimate the time required for moisture levels to reach 60%. The experimental results show that the LSTM model provides good predictive performance, with an absolute error of 1.15–2.27 seconds for morning data and 0.28–1.56 seconds for afternoon data. This indicates that the model is capable of improving the precision of automatic irrigation for beef tomato plants. Keywords: Automatic irrigation, Soil moisture, Beef tomato, LSTM, Farmbot, Duration prediction.

---

Availability

Detail Information

Series Title

-

Call Number

T1896962025

Publisher

Indralaya : Prodi Teknik Elektro, Fakultas Teknik Universitas Sriwijaya., 2025

Collation

xv, 123 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

635.642 07

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Elektro
Budidaya Tanaman Tomat

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment