Multi-objective optimization of coal-fired power plants using differential evolution

•Multi-objective optimization of large-scale coal-fired power plants using differential evolution.•A newly-proposed algorithm for searching the fronts of decision space in a single run.•A reduction of cost of electricity by 2–4% with an optimal efficiency increase up to 2% points.•The uncertainty comes mainly from temperature- and reheat-related cost factors of steam generator.•An exergoeconomic analysis and comparison between optimal designs and one real industrial design. The design trade-offs between thermodynamics and economics for thermal systems can be studied with the aid of multi-objective optimization techniques. The investment costs usually increase with increasing thermodynamic performance of a system. In this paper, an enhanced differential evolution with diversity-preserving and density-adjusting mechanisms, and a newly-proposed algorithm for searching the decision space frontier in a single run were used, to conduct the multi-objective optimization of large-scale, supercritical coal-fired plants. The uncertainties associated with cost functions were discussed by analyzing the sensitivity of the decision space frontier to some significant parameters involved in cost functions. Comparisons made with the aid of an exergoeconomic analysis between the cost minimum designs and a real industrial design demonstrated how the plant improvement was achieved. It is concluded that the cost of electricity could be reduced by a 2–4%, whereas the efficiency could be increased by up to two percentage points. The largest uncertainty is introduced by the temperature-related and reheat-related cost coefficients of the steam generator. More reliable data on the price prediction of future advanced materials should be used to obtain more accurate fronts of the objective space.