Advanced Heat Integration Tool for Simulation-based Optimization Framework

Submission for Process Design

Authors: Yang Chen, John C. Eslick, Ignacio E. Grossmann and David C. Miller
Framework for Optimization and Quantification of Uncertainty and Sensitivity (FOQUS) developed by Carbon Capture and Simulation Initiative (CCSI) is powerful software for the optimal design and operation of energy and chemical processes. A heat integration tool has been incorporated into this framework so that simultaneous optimization and heat integration can be achieved for rigorous models developed in process simulators [1].
In this framework, process models are simulated via commercial flowsheet simulators (such as Aspen Plus, Aspen Custom Modeler and gPROMS), and are optimized via external derivative free optimizers (DFOs). Heat loads and temperatures of process streams are transferred from simulation results to the heat integration tool and are solved within each iteration of DFO to update the energy target. The heat integration tool is developed in GAMS by using the linear programming (LP) transshipment formulation [2], which predicts the minimum utility cost (or consumption). Both process simulation and heat integration results are returned to the DFO to minimize the total cost of the entire process. This optimization and heat integration framework has achieved a significant increase of net power efficiency in the case study of a power plant
with carbon capture [1].
However, the original heat integration tool is based on the assumption that constant heat capacity flow rates (FCps) hold between the supply and target temperatures for all process streams. This assumption may not be true for many real applications, especially for process streams with phase changes or non-ideal thermodynamic behavior. These streams usually show variable FCps and nonlinear composite curves. Hence, the simplification for constant FCps may overestimate (or underestimate) the real heat recovery and even give rise to infeasible heat exchanger network design. One way to deal with this issue is to apply piecewise linear approximation to the composite curves. By constructing a larger number of temperature intervals, we can still assume constant FCps for process streams in each small temperature interval. To implement this idea, we use a series of heaters/coolers with identical load for each cold/hot
stream in the process model so that the sum of enthalpy changes within all intermediate temperature intervals matches the total change of enthalpy predicted by simulation results. The heat integration tool for variable FCps is demonstrated in two case studies: the mono chloro benzene (MCB) separation process and the carbon capture and compression process. Both case study results reveal that more accurate utility consumptions can be obtained after considering variable FCps.
Area targeting model is added to the heat integration tool which is based on the LP formulation developed by Jezowski et al. [3]. The total cost of heat exchanger network is estimated by summation of utility cost (by energy target) and capital cost (by area target), which is then transferred to the DFO together with process simulation results to obtain more accurate total cost. The heat integration tool with both energy and area targets is tested for minimizing the
cost of electricity (COE) of a supercritical coal power plant with carbon capture and compression. A lower COE is achieved than the case without heat integration, and a more accurate estimation
of COE is obtained compared to the old heat integration tool with only energy target.

Reference

[1] Chen Y, Eslick JC, Grossmann IG, Miller DC. Simultaneous optimization and heat integration based on rigorous process simulations. Proceedings of the 8th International Conference on Foundations of Computer-Aided Process Design - FOCAPD 2014, Cle Elum, Washington, USA, July 13-17, 2014.

[2] Papoulias SA, Grossmann IE.A structural optimization approach to process synthesis â?? II.
Heat recovery networks. Computers and Chemical Engineering. 1983;7(6):707-721.
[3] Jezowski JM, Shethna HK, Castillo FJL. Area target for heat exchanger networks using linear programming. Industrial & Engineering Chemistry Research. 2003;42(8):1723-1730.