Deep learning optimization of a combined CCHP and greenhouse for CO2 capturing; case study of Tehran

Abstract

Although renewable-based energy systems have been increasing and attractive worldwide, the intermittent nature of renewable energy sources cannot cause fossil fuels to be completely removed from the energy systems. So, an attractive approach is to capture produced CO2 from existing power plants for several purposes for the status quo. One of these purposes is to use CO2 in greenhouses in the propinquity of power plants. In this context, a greenhouse combined with an absorption chiller and Organic Rankin Cycle was designed and analyzed in order to exert the high temperature exhausted gases enthalpy from the micro power plants. The produced CO2 is also separated using a catalytic converter and is employed for the greenhouse products to compensate for the lack of CO2 for the required standard for plant growth. Compared to similar systems, the present analysis is comprehensive enough in all energy, exergy, economic, and environmental perspectives. Besides, all functions are calculated using a deep artificial neural network algorithm, and two scenarios were considered for both summer and winter. Weather conditions data were collected during a 10-year interval to predict the system's behavior. After optimization, the CO2 production rate was declined by 56% and the maximum energy and exergy efficiencies were achieved by 47.3% and 36.6%, respectively. Also, a net annual interest of 23.4 M$ was achieved due to an increase in greenhouse harvest.