Infoscience

Thesis

Environomic modeling and optimization of advanced combined cycle cogeneration power plants including CO2 separation options

Considering the economic importance of the electricity generation sector as well as its contribution to local, regional and global pollution, the development of new cost-effective and environmentally benign electricity generation and cogeneration systems is an essential task. In this context, combined cycle cogeneration power plants are among the most promising systems. The integration of gas turbines with a steam turbine cycle leads to complex systems and to an increasing number of technological options for their improvement. Faced with growing concerns about greenhouse gas emission, systems for CO2 separation become an additional option and compete with measures for efficiency improvement. The simultaneous consideration of economic and environmental concerns as well as the increasing system complexity and the growing number of options represent an important challenge to the engineer and require the development of new approaches and design tools. The present work contributes to the development of such an approach and its application to combined cycle cogeneration power plants and their advanced options. These options, such sequential combustion gas turbines, compressor intercooling of CO2 separation can improve performance with respect to efficiency and emissions, but require increased investment costs. These competing factors are united in a so-called environomic model which takes into account the costs of resources and capital investment, as well as the costs of pollution, and calculates a single criterion – the total cost of electricity production. The costs of pollution used in this work are based on economic studies from the literature. For internalization, a penalty factor allows the degree of pollution already present in the environment to be taken into account. Benefits due to the substitution of domestic furnaces through cogeneration are also considered. A so-called "superconfiguration" model allows the modeling of simultaneous changes in configuration and design. Uncertainties linked to thermodynamic performance and investment costs model are addressed through calibration by means of reference combined cycle power plants, as well as through a proposed method for directly taking them into account in the model. The influence of varying requirements such as the power plant size, operating time and emission limits can also be modeled, providing valuable information such as thermodynamic states, component performance and investment costs. The resulting optimization problem is of the non-linear, mixed integer type. The complex task of finding optimum solutions is accomplished by means of powerful optimization methods that allow searching of the entire solution space. In this work, a genetic algorithm was adapted to the requirements of the superconfiguration optimization, permitting the simultaneous optimization of configuration and design. Genetic algorithms work with populations instead of a single data point and rely on biological mechanisms such as mating, mutation and replacement, and provide multiple promising solutions. Reference cases with different electricity and heat demand as well as different annual operating time are optimized and analyzed in detail with respect to the solutions obtained. Based on these reference cases, sensitivity studies on the cost of pollution, as well as emission limits, show the influence of these factors on the combined cycle configuration and design, as well as on the cost structure of electricity, and demonstrate an application of the method.

Fulltext

Related material