A data-driven analytical roadmap to a sustainable 2030 in South Korea based on optimal renewable microgrids

Combined technological, economic, sociological, and environmental (TESE) models can play a unique role in leveraging renewable energies and supporting sustainable development. Yet, a multi-aspect TESE model has never been used in Korea to adapt to climate change through sustainable energy policies. This comprehensive TESE study addresses several sustainability challenges in the Korean energy sector. First, demand electricity is predicted using four deep and stacked neural networks to develop a smart demand-based model. Second, optimal hybrid renewable microgrids (HRMGs) are simulated at 17 sites to evaluate five renewable energy sources in three scenarios. Third, hybrid assessment results are clustered using a K-means algorithm to generate hybrid renewable energy maps for South Korea. Fourth, the TESE model is analyzed with more than 13 variables using a cascading multi-criteria decision-making approach to prorate a budget and develop a prioritized roadmap for a sustainable 2030 in Korea. Fifth, a stochastic linear mathematical model is developed to propose a sustainable energy policy that considers the water-energy-carbon nexus. The results show that a convolutional neural network can efficiently predict sequential demand electricity (R-2 = 98.79%), with respective biogas, solar, hy-drostatic, wind, and hydrokinetic energy fractions of 45.7%, 34.5%, 14%, 5.78%, and 0.01% under optimal conditions in Korea. The present free market-based policies are recommended to be revised in favor of domestic production of renewable energy facilities if new jobs can be created for more than $7500 each, and the carbon penalty cost can be kept below $83/ton CO2-eq in Korea.

Automated summary

Combined technological, economic, sociological, and environmental (TESE) models are crucial for leveraging renewable energies and supporting sustainable development in Korea. This comprehensive TESE study addresses multiple sustainability challenges in the Korean energy sector. The study utilizes deep neural networks, simulation models, clustering algorithms, and decision-making approaches to predict electricity demand, evaluate hybrid renewable microgrids, generate renewable energy maps, develop a prioritized roadmap, and propose a sustainable energy policy.