Cost-Effective Processes of Solar District Heating System Based on Optimal Artificial Neural Network

Aligning with the EU 2030 climate and energy package to achieve a share of at least 27% of renewable energies, and to improve the energy efficiency by at least 27%, the future solar district heating systems (SDHS) may enable the transition to a complete renewable society. Even though this promising tendency of the SDHS, a range of potential barriers are obstructing the wide deployment of SDHS and promoting high variation in quantifying the SDHS benefits over its lifetime. In this context, the optimization approaches are a viable option for determining the optimal structure, sizing, and operation of the SDHS. However, Meta-heuristics optimization models are computationally very expensive and have many limitations regarding the optimization process. Aligning with these challenges, this work tends to develop a robust Artificial Neural Network model based on Bayesian Optimization to solve the computational obstacle associated with heuristics optimization models for SDHS.