A Master of Science thesis in Chemical Engineering by Ibrahim Tamer Masoud entitled, "Heat Integration and Controllability Analysis of Heat Exchanger Networks," submitted in May 2014. Thesis advisor is Dr. Nabil Abdel Jabbar and co-advisor is Dr. Rachid Chebbi. Available are both soft and hard copies of the thesis.
The objective of this research is to present a methodological framework for designing heat exchanger networks, which best addresses heat recovery, economics and controllability. The proposed framework formulates a systematic approach consisting of a series of simple design steps. The steps include heat integration techniques such that the design can achieve its energy recovery targets: a detailed cost analysis to minimize both capital and utility costs, and steady-state controllability measures to keep the design controllable. A heat exchanger network case study was used to test the proposed framework, and the results were compared with previous works in the literature. Pinch and Superstructure heat integration methods were applied to the case study; both designs achieved the system's required heat recovery, however, the Superstructure design showed lower costs than the Pinch design. Both heat integration methods were also compared in terms of their inner loop interactions by performing Relative Gain Array and Singular Value Decomposition analyses. The results showed that the Superstructure design had less inter-loop interactions than the Pinch design. Control of the heat exchanger networks was achieved by placing bypasses around some of the heat exchangers and manipulating the bypass fractions. All bypass fractions were set at 10%. It was found that bypass fractions marginally increase the capital cost of the HEN of about 2-4%. However, the increase in the bypass fractions did not affect the steady state controllability of the HEN system. The design with the proposed framework was further verified by a dynamic analysis and compared with a benchmark case from the literature. The closed-loop dynamic simulation was performed via ASPEN-HYSYS for different HEN the design that was obtained from the proposed framework in this study and the ones obtained in the literature. Dynamic simulation results revealed that our design exhibited better control characteristics in terms of disturbance rejection and set point tracking. Furthermore, it was found that the best control performance which was achieved in this study with the highest bypass fraction of 10%, had the highest capital cost for HEN design. This finding confirmed that there is a trade-off between the design and controllability of HENs.