Optimasi Alokasi Daya Pada Sistem Penyimpanan Energi Hibrida Berbasis Low Pass Filter Dan Artificial Neural Network

Amin, Richi Mohammad (2025) Optimasi Alokasi Daya Pada Sistem Penyimpanan Energi Hibrida Berbasis Low Pass Filter Dan Artificial Neural Network. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022241124-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Penelitian ini menyajikan pengoptimalan alokasi daya dalam sistem penyimpanan energi hibrida (HESS) dengan menggabungkan teknik Low Pass Filter (LPF) dan Artificial Neural Network (ANN). Strategi berbasis LPF konvensional menggunakan frekuensi cut-off tetap yang tidak dapat beradaptasi dengan kondisi pengoperasian dinamis, menghasilkan kinerja yang tidak optimal dan masa pakai baterai yang lebih rendah. Penelitian ini mengusulkan kerangka kerja optimasi yang menentukan frekuensi cut-off optimal melalui tiga pendekatan: optimasi berbasis biaya, berbasis stres, dan hybrid. Analisis spektral daya yang diminta menegaskan bahwa kandungan energi sebagian besar terkonsentrasi dalam rentang frekuensi rendah (10-6 hingga 10-3 Hz), membenarkan strategi alokasi daya berbasis frekuensi. Metode pengoptimalan hibrida menghasilkan frekuensi cut-off optimal 0,0013 Hz, menyeimbangkan pertimbangan ekonomi dan teknis dengan pengurangan tekanan baterai sebesar 15% dibandingkan dengan pendekatan berbasis biaya sambil meningkatkan biaya sistem hanya sebesar 8%. Model ANN yang diterapkan mencapai kemampuan prediktif (R² = 0,9728) dalam menentukan frekuensi cut-off optimal di berbagai kondisi operasi. Evaluasi kinerja menunjukkan bahwa pendekatan berbasis ANN menghasilkan pola alokasi daya yang hampir identik dibandingkan dengan metode LPF konvensional sambil mengeksekusi sekitar 120 kali lebih cepat, memungkinkan adaptasi real-time terhadap perubahan kondisi sistem. Metodologi gabungan secara efektif memisahkan komponen daya frekuensi rendah (87%) ke baterai dan komponen frekuensi tinggi (13%) ke superkapasitor, mengoptimalkan pemanfaatan komponen berdasarkan karakteristik intrinsiknya sambil mempertahankan efisiensi sistem di atas 88%.
==================================================================================================================================
This paper presents optimizing power allocation in hybrid energy storage systems (HESS) by combining Low Pass Filter (LPF) and Artificial Neural Network (ANN) techniques. Conventional LPF-based strategies employ fixed cut-off frequencies that cannot adapt to dynamic operating conditions, resulting in suboptimal performance and reduced battery lifespan. This research proposes optimization framework that determines optimal cut-off frequencies through three approaches: cost-based, stress-based, and hybrid optimization. Spectral analysis of requested power confirmed that energy content is predominantly concentrated in low-frequency ranges (10-6 to 10-3 Hz), justifying the frequency-based power allocation strategy. The hybrid optimization method yielded an optimal cut-off frequency of 0.0013 Hz, balancing economic and technical considerations with a 15% reduction in battery stress compared to cost-based approaches while increasing system costs by only 8%. The implemented ANN model achieved predictive capability (R² = 0.9728) in determining optimal cut-off frequencies across various operating conditions. Performance evaluations demonstrated that the ANN-based approach produces nearly identical power allocation patterns compared to conventional LPF methods while executing approximately 120 times faster, enabling real-time adaptation to changing system conditions. The combined methodology effectively separates low-frequency power components (87%) to batteries and high-frequency components (13%) to supercapacitors, optimizing component utilization based on their intrinsic characteristics while maintaining system efficiency above 88%.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Artificial Neural Network, Penyimpanan Energi Hibrida, Low Pass Filter, Alokasi Daya, Hybrid Energy Storage System, Power Allocation
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2921 Lithium cells. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2941 Storage batteries T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7872.C65 Supercapacitors. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7872.F5 Filters (Electric)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Richi Mohammad Amin
Date Deposited:	18 Jul 2025 08:52
Last Modified:	18 Jul 2025 08:52
URI:	http://repository.its.ac.id/id/eprint/120098

Actions (login required)

View Item