Data-driven based financial analysis of concentrated solar power integrating biomass and thermal energy storage: A profitability perspective

This study utilizes decision tree algorithms to estimate the financial feasibility of concentrated solar power (CSP). The main focus of CSP is on solar tower (ST) technology combined with two backup systems, such as biomass boilers and thermal energy storage (TES). The main goal is to develop three decision tree algorithms to predict the power plant's profitability factor (P_F) for each of the following three operational scenarios: solar tower-base case-no biomass (ST-BC-NB), solar tower-operation strategy 1-medium biomass (ST-OS1-MB), and solar tower-operation strategy 2-full biomass (ST-OS2-FB). P_F was predicted according to main input parameters, including direct capital costs, biomass cost, annual escalation rate, hourly electricity price, annual escalation rate, and peaks and troughs for daily electricity prices. Thermal energy storage was in five different capacities: no-thermal energy storage (No-TES), 5 h, 10 h, 15 h, and 20 h. The decision tree models demonstrated accurate predictions with low errors, high confidence levels, and most data falling within the 95% confidence interval for the “No-TES” case. Solar power plants with biomass backup had a 30% reduction in generation costs compared to conventional plants. The configurations without thermal energy storage had the highest profitability, with a maximum P_F of −0.014 USD/kWh and a 25% chance of achieving profitability.