CD Skripsi

Identifikasi Tandan Buah Segar Kelapa Sawit Varietas Dura Dan Tenera Menggunakan Sensor Ultrasonik Dan Jaringan Syaraf Tiruan (Jst)

---

Oil palm is a major commodity in Indonesia’s plantation industry. However, the identification process of fresh fruit bunch (FFB) varieties such as Dura and Tenera is still carried out manually and destructively, making it inefficient and tiring. This research aims to develop a non-destructive identification system based on ultrasonic sensors and Artificial Neural Networks (ANN) to distinguish between Dura and Tenera varieties based on output voltage responses. The samples consisted of 40 FFBs classified by ripeness level, and their output voltages were measured using the LV-MaxSonar-EZ MB1010 ultrasonic sensor. The voltage data were used as input for an ANN with a backpropagation algorithm, implemented using Python programming language. The results showed that the voltage values for ripe Dura ranged from 10–14 mV, unripe Dura from 16–20 mV, ripe Tenera from 30–33 mV, and unripe Tenera from 34–38 mV. The ANN architecture consisted of one input layer, two hidden layers, and one output layer, which were used to classify each variety. The accuracy graph increased from epoch 5 to epoch 20, increasing from approximately 0.5 to 1.0. Meanwhile, the loss graph showed a consistent decrease in loss values from the beginning until around epoch 50, from approximately 0.7 to near 0. Using a confusion matrix, the model achieved a prediction accuracy of 100%.

Keywords: Sorting and Grading, Oil Palm Fresh Fruit Bunch, Dura and Tenera Variety, Ultrasonic Sensor, Artificial Neural Network

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2103110734

Penerbit

Pekanbaru : Universitas Riau FMIPA Fisika., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2103110734

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

FISIKA

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar