Text

SKRIPSI IMPLEMENTASI METODE LONG SHORT-TIME MEMORY (LSTM) UNTUK PREDIKSI SUHU PADA SISTEM MONITORING KANDANG AYAM BROILER

---

This study developed a broiler cage temperature prediction model using the Long Short-Term Memory (LSTM) algorithm to help farmers maintain cage temperature in the optimal range (20°C-32°C). The problem of unstable cage temperature can cause thermal stress, reduce productivity, and even increase the risk of broiler death. The temperature monitoring system is designed using a DHT11 sensor connected to an Arduino Uno microcontroller, an RTC, and a microSD-based storage module. The collected temperature data is processed using the LSTM algorithm which excels in analyzing time series data. The model was trained using 100 neurons with 20 epochs and evaluated with RMSE metrics of 1.75 and MAPE of 3.8%, indicating a high level of accuracy. Calibration of the DHT11 sensor against an HTC-2 digital thermometer resulted in an average temperature difference of 0.19°C with 1% error rate and 99% accuracy, proving the reliability of the device. The results show that this prediction system is able to rovide benefits in supporting real-time and preventive cage temperature management. Thus, this research not only improves the efficiency of cage management but also supports broiler productivity.

Keywords : Broiler Chicken, Cage Monitoring, Long Short-Term Memory (LSTM), Temperature Prediction

---

Ketersediaan

125-UN57.U1-STE-II-2025

ELEKTRO SAK I 2025

Ruang Skripsi (TEKNIK ELEKTRO)

Tersedia namun tidak untuk dipinjamkan - No Loan

Informasi Detail

Judul Seri

-

No. Panggil

ELEKTRO SAK I 2025

Penerbit

MAGELANG : UNIVERSITAS TIDAR., 2025

Deskripsi Fisik

XVI;116HAL;29.5CM

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

621.38

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

PRODI_S1_TEKNIK_ELEKTRO
FAKULTAS_FT

Info Detail Spesifik

-

Pernyataan Tanggungjawab

-

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar