Soft computing investigation of stand-alone gas turbine and hybrid gas turbine–solid oxide fuel cell systems via artificial intelligence and multi-objective grey wolf optimizer

Despite of huge effort performed in the literature encompassing the concept of hybrid solid oxide fuel cell–gas turbine (SOFC-GT), the economic and environmental superiority of this hybrid system over a conventional gas turbine (GT) has not yet been clarified. In the present study, a proper scale of solid oxide fuel cell was integrated with a GT and a comprehensive comparison analysis based on energy, exergy, exergo-economic and environmental (4E) criteria were conducted between 10 MW GT and hybrid SOFC-GT systems. Three different scenarios were considered for the comparison analysis, encompassing baseline operation, parametric behavior and optimum conditions. An appropriate artificial neural network (ANN) was utilized to model the systems, and then, the obtained ANNs were introduced to Grey Wolf optimizer. Using grey wolf optimizer, a multi-objective optimization problem was solved considering exergy efficiency, CO2 emission and unit cost of product as the objectives. The hybrid SOFC-GT system had a better performance than the GT system in terms of environmental and exergo-economic performances as the results of all three scenarios showed. However, this integration slightly increased the investment cost of the system at the baseline scenario. After optimization, at best trade-off between the objectives, the SOFC-GT exergy efficiency, CO2 emission and unit product cost were obtained to be 68.5 %, 277.7 kg/MWh, and 22.7 $/GJ, however, these values for the GT were 32.38 %, 588.4 kg/MWh, and 26.9 $/GJ, respectively.

Перевод пока недоступен