NSF/EPA: A Process Analysis Approach to Pollution Prevention

摘要：随着人们对环境的日益关注，有必要提高过程模拟能力，并开发设计工具，在设计和合成大型化学过程中考虑环境因素。设计和发展纳入环境控制战略的过程的一个主要障碍涉及某些技术和经济参数的不确定性，这通常导致设计、工厂性能和费用估计方面的不确定性。将环境问题纳入设计是一个多目标问题，需要满足许多措施和约束，如可控性、安全性、法规遵从性、成本和风险。任何化工过程的概念设计都涉及到最佳流程图结构的确定，通常称为最佳设计构型，通常由构成“上层结构”的大量备选方案获得。事实上，涉及不确定性的大规模过程的综合、设计和优化通常需要大量的计算工作。本文对随机建模、随机优化与规划、多目标优化和不确定综合等方面的有效算法进行了研究和验证。总体目标是通过(a)提供一个综合评估框架工具，以确保从合成到设计再到操作的过程工程活动的所有阶段都能充分解决环境问题，从而提高工艺设计领域的最先进水平；(b)开发改进的随机建模、随机优化、多目标优化算法，提高该框架对大规模复杂问题的适用性。与参与行业相结合的现实世界问题将被用作这些技术的试验台。所开发的方法将应用于分析复杂的前工艺流程图，并有望基于环境考虑和优化设计、工厂性能和不确定性问题提供工艺可行性的见解。

ABSTRACT - Diwekar - 9613561 With the growing concern for the environment, it is necessary to enhance process simulation capabilities and to develop design tools that account for environmental factors in the design and synthesis of large-scale chemical processes. A major obstacle in the design and development of processes incorporating environmental control strategies involve uncertainties in some of the technical and economic parameters, which commonly lead to uncertainties in design, plant performance and cost estimates. Incorporating environmental concerns in the design is a multi-objective problem satisfying a number of measures and constraints such as controllability, safety, compliance with regulations, cost and risk. A conceptual design of any chemical process involves the identification of an optimal flowsheet structure, commonly called optimal design configuration, obtained typically form a large number of alternatives constituting the "superstructure". In fact, synthesis, design, and optimization of large-scale processes involving uncertainties often requires considerable computational effort. This work and validation of efficient algorithms for stochastic modeling, stochastic optimization and programming, multi-objective optimization, and synthesis under uncertainty. The overall objective is to advance the state-of-the-art in the4 area of process design for environmental considerations by (a) providing an integrated assessment framework tool which can be used to insure that environmental issues are fully addressed in all phases of process engineering activities from synthesis to design to operations; and (b) developing improved algorithms for stochastic modeling, stochastic optimization, multi-objective optimization which will enhance the applicability of the framework for large scale complex problems. Real world problems in conjunction with participating industries will be used as test beds for these techniques. The methodology developed will be applied to analyze compl ex process flowsheets and is expected to provide insights into process feasibility based on environmental considerations and optimal design, plant performance and uncertainty issues.