Techno-environmental assessment and machine learning-based optimization of a novel dual-source multi-generation energy system

Amirreza Javaherian, Nima Ghasemzadeh*, Nima Javanshir, Mortaza Yari*, Mohammad Vajdi, Hossein Nami

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review


The utilization of high-temperature hybrid energy systems has a vital and promising role in reducing environmental pollutants and coping with climate change. So, in the present research, a dual-source multigeneration energy system composed of a gas turbine, a supercritical carbon dioxide recompression Brayton cycle, an organic Rankine cycle, an absorption refrigeration system, and a reverse osmosis desalination unit is designed and analyzed from thermodynamic, environmental and economic perspectives. The system supplies power with a stable load to follow the changes in the demand side which is important for off-grid distributed energy systems. The dual-source operation of the system makes it possible to generate sustainable electricity leading to less utilization of fossil fuels in the gas turbine subsystem and reduction in environmental pollution, and furthermore, malfunctioning of a subsystem will not lead to the failure of the entire plant. Three multi-objective optimizations with different objective functions are accomplished using artificial neural network from data learning and genetic and Greywolf algorithms to obtain the best-operating conditions. Under the base conditions, for the total input energy of 699 MW to the entire system, the energy and exergy efficiencies, the unit exergy cost of products, the carbon dioxide emission index, and the payback period, respectively, were found to be 45%, 54%, 15.3 $/GJ, 112.2 kg/MWh, and 7.2 years. The net output power of the proposed system was calculated as 288.2 MW. A sensitivity analysis revealed that with a change in the pressure ratio of the supercritical carbon dioxide cycle, the net generated power and overall efficiency take maximum values of 293.9 MW while the unit exergy cost of products and carbon dioxide emission index take minimum values of 15.3 $/GJ and 110.1 kg/MWh, respectively. Furthermore, increasing the pressure ratio of the gas turbine leads to maximum values of 45 and 54% in overall energy and exergy efficiencies, respectively.
Original languageEnglish
JournalProcess Safety and Environmental Protection
Pages (from-to)537-559
Publication statusPublished - Aug 2023


  • Emission reduction
  • Multigeneration system
  • Gas turbine
  • Techno-environmental
  • Multi-objective optimization
  • Machine learning


Dive into the research topics of 'Techno-environmental assessment and machine learning-based optimization of a novel dual-source multi-generation energy system'. Together they form a unique fingerprint.

Cite this