Design of heat and power recovery systems for energy conservation in chemical plants

This thesis presents new strategies for the design of power cycles to satisfy the heating and power demands of chemical processes. Methods for the design of efficient heat exchanger networks (HEN) have been extended to include the power cycles of utility systems and refrigeration systems. Because of the high energy and capital costs of these systems, their inclusion in the heat and power integration problem further reduces the energy consumption of the process by allowing heat integration between the power cycles and the heat recovery network of the process, and among the power cycles themselves. The heat and power integration problem is formulated as a nonlinear programming (NLP) model using an optimality criterion as a design objective that includes the energy-capital trade-offs for the power cycles. The strategies utilize the Temperature Interval method and permit the efficient use of negative heat deficits to obtain lower cost designs. Also, a technique for the lumping of temperature intervals is developed, which often increases the efficiency of optimization without adversely affecting the solution. Initially, a NLP model is formulated for the integration of the heat engines and heat pumps with the process. The model permits an easy assessment of the alternatives for integration and the potential working fluids. Then, an algorithm is presented for the synthesis of cascade refrigeration systems, integrated with the process, with alternate working fluids, and optimized over a continuous range of operating temperatures. Finally, a nonlinear programming strategy is presented for the synthesis of utility systems that satisfy the heating and power demands of the process at 100 percent efficiency. The utility system is modelled as a cascade of heat engines that exchange heat with the process and among themselves. Mass exchange between the adjacent heat engines produces lower pressure designs and provides heat for the process over a broader range of temperatures for a given power demand. Through mass integration, it was discovered that simple Rankine cycles are adequate to generate the complex utility systems that characterize industrial practice. Process examples are presented to illustrate the utility of these models