FIRMANSYAH, ACHMAD DICKY
Unknown Affiliation

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Estimasi SOC Saat Discharging pada Baterai VRLA Berbasis Elman Backpropagation YANARATRI, DIAH SEPTI; SUTEDJO, SUTEDJO; FIRMANSYAH, ACHMAD DICKY; IRIANTO, IRIANTO; RAKHMAWATI, RENNY; ADILA, AHMAD FIRYAL
ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika Vol 12, No 4: Published October 2024
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/elkomika.v12i4.862

Abstract

ABSTRAKPenurunan performa baterai terjadi akibat siklus pengisian dan pengosongan berulang yang melebihi batas, mempercepat degradasi. Penelitian ini bertujuan untuk meningkatkan akurasi estimasi State of Charge (SOC) baterai menggunakan Artificial Neural Network (ANN) dengan algoritma Elman Backpropagation. Metode digunakan karena menambahkan lapisan context neuron yang mampu menangkap pola dinamis pada data baterai. Pengujian dilakukan dengan membandingkan hasil estimasi SOC dari metode ini dengan metode Coulomb Counting. SOC baterai diestimasi dari 100% hingga 60%, dan hasil menunjukkan bahwa meskipun Coulomb Counting awalnya memberikan SOC lebih tinggi, estimasi dari kedua metode menjadi lebih mirip seiring waktu. Error estimasi berkisar antara 0,1% hingga 14,7%. Algoritma Elman Backpropagation terbukti mampu memberikan estimasi SOC yang lebih akurat, namun masih memerlukan kalibrasi lebih lanjut untuk meningkatkan presisi.Kata kunci: Artificial Neural Network, Baterai, Coulumb Counting, Elman Backpropagation, State of Charge. ABSTRACTThe decline in battery performance occurs due to repeated charge and discharge cycles that exceed limits, accelerating degradation. This study aimed to improve the accuracy of State of Charge (SOC) estimation using an Artificial Neural Network (ANN) with the Elman Backpropagation algorithm. The method used was unique in adding a context neuron layer capable of capturing dynamic patterns in battery data. Testing was conducted by comparing SOC estimates from this method with the Coulomb Counting method. The battery's SOC was estimated from 100% to 60%, and the results showed that although Coulomb Counting initially provided higher SOC estimates, the estimates from both methods became more similar over time. Estimation errors ranged from 0.1% to 14.7%. The Elman Backpropagation algorithm proved to provide more accurate SOC estimates, although further calibration is needed to improve precision.Keywords: Artificial Neural Network, Battery, Coulumb Counting, Elman Backpropagation, State of Charge.