The IDAES framework utilizes the Pyomo ( Hart et al., 2017) algebraic modeling language (AML), an open source framework for formulating large scale optimization problems based on the Python programming language, which can interface with a wide range of optimization solvers. Department of Energy’s Institute for the Design of Advanced Energy Systems (IDAES) is addressing this need by developing a next-generation process systems engineering framework ( Miller at al., 2018) that is built for optimization from the ground up, enabling the use of modern optimization solvers with a framework for advanced process modeling.
ANALYTIC SOLVER PLATFORM FOR INCREASING SERIES SOFTWARE
Thus, there exists a gap in the capabilities of existing software for optimizing process flowsheets. Developing and applying these models to chemical processes also requires expert knowledge in modeling and optimization not available to the typical process engineer. In these cases, models must be laboriously assembled for specific systems, often requiring specialized initialization procedures. In contrast, there exist several platforms and modeling languages designed specifically for solving large scale optimization problems however, these platforms lack the specialized infrastructure and model libraries necessary for easily simulating chemical processes and energy systems. However, most process simulation packages focus primarily on solving well defined simulation problems, and often have only limited capabilities for advanced optimization, such as conceptual design or optimization under uncertainty. These tools allow engineers to solve very large process flowsheets under both steady-state and dynamic conditions. To facilitate the development of process models, a number of simulation packages, such as Aspen Plus®, gPROMS, ProSim, PRO/II®, have been developed which provide robust and easy to use tools, including libraries of common process unit operations, thermo-physical property models, and efficient algorithms to solve large, sparse systems of nonlinear differential algebraic equations. Miller, in Computer Aided Chemical Engineering, 2018 1 IntroductionĪdvances in computational power and numerical optimization routines have enabled the possibility of applying rigorous simulation and optimization techniques to large scale problems such as those associated with the design, optimization and control of integrated chemical processes and energy systems.