| Summary: | 博士 === 國立成功大學 === 化學工程學系碩博士班 === 101 === Every chemical process can be evaluated according to more than one performance criterion. A good design should be not only cost optimal but also operable in a realistic environment. For the latter purpose, various operational properties of a given process, e.g., flexibility, controllability, reliability, and safety, have been taken into consideration in the past. The present study is performed mainly to address critical issues in implementing flexibility analysis.
A motivational example is first presented in this thesis to justify the need to analyze the impacts of uncertainties in process designs. By addressing both flowsheeting and scheduling issues, a deterministic approach is adopted in this example to design batch azeotropic distillation processes for the homogeneous ternary and quaternary systems. Since the variations in feed quality and operating conditions (e.g., pressure, temperature, etc.) are not considered in the synthesis strategies, the resulting networks may not always be operable if the design parameters deviate from their nominal values. These concerns can be addressed in rigorous flexibility analyses detailed later in the thesis.
Flexibility index (FI) is a quantitative measure of the feasible region in the space of the uncertain parameters. More specifically, FI corresponds to the maximum allowable deviation of the uncertain parameters from their nominal values, by which feasible operation can be guaranteed with the proper manipulation of the control variables. Two previously unsolved problems in flexibility analysis are discussed in this thesis: (1) identification of the optimal nominal parameter values, and (2) formulation of a generic model for computing the temporal flexibility index.
A solution strategy is developed in this work to optimally stipulate the nominal operating conditions of an existing process for maximum flexibility. The conventional flexibility index model is used to evaluate system resiliency based on fixed nominal conditions, while a direct search method (differential evolution) is performed accordingly to identify the best candidates. The impacts of nominal settings and the effectiveness of the proposed optimization approach are demonstrated in the given examples.
On the other hand, the cumulative effects of temporary disturbances in finite time intervals are also analyzed in this work so as to avoid serious consequences in realistic operations. In particular, the temporal flexibility concept is defined to address this issue mathematically. The optimization program used for evaluating the corresponding performance measure is built on the basis of a dynamic system model, which usually consists of a set of differential-algebraic equations (DAEs). To simplify numerical calculations, the differential quadrature (DQ) is utilized to approximate these DAEs with equality constraints. It can be observed from the examples provided in the thesis that this approach is convenient and effective.
Finally, a realistic solar driven membrane distillation desalination system (SMDDS) is presented as a further example to show the usefulness of temporal flexibility index. By assessing operational flexibilities of alternative candidates, the most appropriate design can be identified systematically and it is clearly demonstrated that the proposed approach is suitable for addressing various operational issues in SMDDS design.
|
|---|
