Development of P-graph approach for designing polygeneration systems

• المؤلف الرئيسي: Cayamanda, Christina D.
• التنسيق: text
• اللغة: English
• منشور في: Animo Repository 2014
• الوصول للمادة أونلاين: https://animorepository.dlsu.edu.ph/etd_masteral/4626
• المؤسسة: De La Salle University

Polygeneration systems reduce the use of resources while at the same time produces power and other products such as heat and cooling services. The traditional way of designing process systems such as polygeneration plants uses mixed integer linear programming (MILP). Another way of designing polygeneration systems besides MILP is to use graph-theory based approach such as P-graph. P-graph is used in designing network systems such as chemical and manufacturing plants, reaction kinetics, transportation, work allocation, and supply chains. This thesis developed the P-graph methodology for designing polygeneration systems starting from a simple trigeneration system progressing to a polygeneration system with biochar production. The progression of complexity of the design problem was done incrementally. The objective function of each design was to maximize the profit of the polygeneration system. The result of the P-graph design of each case study resulted in two solution structures where the optimal design of each polygeneration design was based in economic potential. For the simple trigeneration system the optimal design has an annual profit of 48,680.32 €, for polygeneration system with purified water production the optimal design has an annual profit of 306,838.90 €, for the polygeneration systems with biomass as part of feed the optimal design has an annual profit of 84,418.61 €, and for polygeneration systems with biochar production has an annual profit of 695,980.60 €. Therefore, it is possible to design polygeneration systems with more than two main products using P-graph where the objective function of each design was based on maximizing the profit. It is also possible for P-graph to generate more than one solution for each design compared with conventional methods such as MILP. However, P-graph is limited to designs with linear models compared with other optimizing programs such as LINGO where it can be used for non-linear models.