Artificial Neural Network (ANN) Backpropagation for Forecasting 100% Renewable Energy in North Sumatera
DOI:
https://doi.org/10.5755/j01.erem.81.1.37767Keywords:
renewable energy, artificial neural network, optimisation, firefly algorithm, particle swarm optimisation, energy trasitionAbstract
Growing environmental awareness and the increasing need to reduce reliance on fossil fuels have driven the development of renewable energy (RE) technologies, such as micro-hydro, photovoltaics, biomass, geothermal, and biogas. However, the utilisation of RE in North Sumatera remains limited compared with fossil fuels, highlighting the need for optimisation strategies to accelerate the transition towards 100% RE. This study develops a predictive model using the artificial neural network (ANN) backpropagation algorithm to maximise RE contributions, minimise dependence on fossil fuels, and forecast the timeline for a full transition to RE. Data from Perusahaan Listrik Negara (North Sumatera), Independent Power Producers (IPP), and palm oil mills were used to model a hybrid generation system incorporating micro-hydro, hydropower, geothermal, biomass, biogas, and solar energy. Simulations were carried out using the firefly algorithm (FA) and particle swarm optimisation (PSO), with optimisation assessed through the renewable energy contribution ratio (RECR). The results indicate that FA outperforms PSO in meeting RE targets, with an average RECR of −51.514 for FA compared with −911.054 for PSO. Predictions using ANN backpropagation suggest that the transition to 100% RE could be realised by 2064 (FA) and 2065 (PSO). This research offers valuable insights for accelerating the transition to sustainable energy, enhancing energy resilience, and reducing environmental impacts.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Environmental Research, Engineering and Management

This work is licensed under a Creative Commons Attribution 4.0 International License.
The copyright for the articles in EREM is retained by the author(s) with the first publication right granted to the journal. The authors agree to the Creative Commons Attribution License 4.0 agreement under which the paper in the Journal is licensed.