Simultaneous Optimization for Organic Rankine Cycle Design and Heat Integration

Abstract

Organic rankine cycle (ORC) has been regarded as the most promising measure for converting low-grade waste heat into electricity during the past decades. To take better advantage of ORC, the parameters of ORC and its heat integration with the background process streams should be optimized simultaneously. In this work, the decision variables of ORC not only include the operating temperature, pressure, and flowrate but also contain the type of working fluids. A set of 43 pure fluids and 36 binary mixtures are adopted from the literature as candidates for working fluids. In addition, both the supercritical and subcritical conditions of ORCs are involved. Background process streams and ORC streams with variable temperatures and flowrates are synthesized for heat integration. Such a complicated heat integration problem cannot be solved by traditional methods, where the temperature and flowrates of streams are fixed. We develop a two-level optimization strategy. The outer level identifies the promising working fluids and their compositions, temperatures, and pressures inside the ORC. The inner level deals with the heat integration problem, where the temperatures and flowrates of streams are the variables. The flowrate of ORC, the utility consumption, as well as area of heat exchangers are optimized in the inner level based on the vertical heat transfer assumption of pinch technology. The outer level is solved by a genetic algorithm, while the inner level is solved by a deterministic algorithm. Case studies show that the inner level model for heat integration is superior in both efficiency and quality of the solution. We also provide an overview of the best ORC selection for a single waste heat source between 100 and 300 °C and show applicability for the integration of ORC and background processes with multiple streams in case studies.

From Molecules to Heat‐Integrated Processes: Computer‐Aided Design of Solvents and Processes Using Quantum Chemistry

“…However, current methods focus either on (1) energy and mass integration or (2) solvent selection. Energy and mass integration: In the literature, several solutions are presented to integrate process optimization with the design of the heat recovery network 9–14. These methods simultaneously design the (reaction‐) separation process and heat exchange by solving large superstructure optimization problems.…”