Computer Aided Chemical Engineering

Abstract

Most renewable energy networks rely on wind and solar energy; known as variable renewable energy (VRE); and its generation process is heavily dependent on weather conditions; leading to fluctuations in the supply side. As a result; this study aims to: 1) develop an optimal forecasting model to predict the supply-demand balance; 2) provide different thresholds to generate potential scenarios; and 3) compare the scenarios using a techno-economic assessment. The optimal model in this study is a GRU; which has an R2 score of 0.994. The levelized cost of electricity (LCOE) ranges from 0.03 USD/kWh to 0.07 USD/kWh. The key conclusions of the study are as follows: 1) conversion factors are used to show that the processed data can be converted to match the feasible pattern of the target year's VRE data; 2) the sampling method accounts for uncertainties in future data and those caused by limited time-series data; 3) the optimal model can be identified by comparing various models using sample data as input; 4) the feasibility of scenarios consisting of a techno-economic component is validated by IRENA; and 5) the probability of LCOE can inform expected budget for energy policy. © 2023 Elsevier B.V.