Zorka Novak Pintarič

A strategy for MINLP synthesis of flexible and operable processes

Abstract This paper presents a sequential two-stage strategy for the stochastic synthesis of chemical processes in which flexibility and static operability (the ability to adjust manipulated variables) are taken into account. In the first stage, the optimal flexible structure and optimal oversizing of the process units are determined in order to assure feasibility of design for a fixed degree of flexibility. In the second stage, the structural alternatives and additional manipulative variables are included in the mathematical model in order to introduce additional degrees of freedom for efficient control. The expected value of the objective function is approximated in both stages by a novel method, which relies on optimization at the central basic point (CBP). The latter is determined by a simple set-up procedure based on calculations of the objective function’s conditional expectations for uncertain parameters. The feasibility is assured by simultaneous consideration of critical vertices. The important feature of the proposed stochastic model is that its size depends mainly on the number of design variables and not on the number of uncertain parameters. The strategy is illustrated by two examples for heat exchanger network synthesis.

An integrated sustainability performance assessment and benchmarking of breweries

Breweries are responding to some sustainability challenges but many of them find sustainability assessment and reporting to be very complex, difficult, and time-consuming tasks. Despite several existing frameworks for the sustainability assessment of companies, none of them specifically addresses breweries. They do not provide them with a transparent, comprehensive, and integrated approach to sustainability assessment, adjusted to the particular circumstances of traditional beer production. In view of these requirements by the brewing industry, this article aims to support breweries in sustainability assessment activities by proposing a methodology for integrated performance assessment. This methodology proposes environmental, societal, economic, and integrated indicators reflecting the characteristics of the brewing industry, compatible with those general indicators proposed by the Global Reporting Initiative (GRI). Although it is important to assess sustainability using several indicators, it may sometimes be difficult to make decisions based on a wide number of performance measurements. Thus, the proposed methodology gradually aggregates sustainable development indicators into sustainability sub-indices and, finally, to a composite sustainability index that tracks integrated information on the economic, environmental, and societal performances of a brewery over time. They can be used both internally, for the identification of “hot spots” and externally, for sustainability reporting and stakeholder engagement. Since breweries strive to outperform their competitors, the proposed methodology enables the benchmarking of a brewery against best performance practices, as a catalyst for improvement and innovation, by providing benchmark values for each indicator. The case study presented in this article illustrates how the proposed methodology could be easily applied in practice, and stimulates breweries to test their effectiveness themselves.

The importance of proper economic criteria and process modeling for single- and multi-objective optimizations

Abstract This paper provides an overview of the influences that different economic objectives have on the efficiencies of those optimal process designs obtained by using single- and multi-objective optimizations. Optimizations of monetary criteria, like the profit, lead to operationally and environmentally more efficient but economically less attractive designs than optimization of non-monetary economic objectives, like the internal rate of return. The net present value produces compromise designs with intermediate efficiencies and environmental impacts. These differences are significant only if the processes’ mathematical models are sufficiently accurate for establishing appropriate trade-offs between investment and cash flow. The Pareto curves obtained by different economic objectives vary regarding the maximum environmental impacts and in the intervals of the environmental indicators. The composed criteria that combine the economic and environmental indicators into one single objective produce smaller differences between optimum designs that are closer to those designs with minimum possible environmental impacts.

The two-level strategy for MINLP synthesis of process flowsheets under uncertainty

Abstract This paper presents a two-level strategy for stochastic synthesis of chemical processes under uncertainty with a fixed degree of flexibility by using the mixed-integer nonlinear programming (MINLP) approach. The objective is to develop an automated and robust strategy which could handle nontrivial optimization problems (about 1000 equations and variables with a considerable number of uncertain parameters — up to 30) that, at the present, cannot be solved in a reasonable period of time by using rigorous stochastic optimization methods. To accomplish the task, the nonlinear subproblems at fixed structures have been decomposed into design and operating optimization levels, the former being facilitated by using a direct search method and the latter by using the reduced dimensional stochastic procedure. Two examples are presented to illustrate the robustness and efficiency of the proposed strategy at solving medium- and large-scale problems.

MINLP Synthesis of Reactive Distillation Using a Disjunctive, Hybrid Model

Abstract In order to develop optimization models for process industries, detailed hybrid modelling providing accurate descriptions of processes is required, where the simplest part of the model is represented explicitly in an equation-oriented environment whilst detailed calculations, e.g. enthalpies, kinetics, equilibrium and efficiencies, are programmed by external implicit functions. The two main objectives of our work have been: i) the development of such hybrid modelling environment, and ii) implementation of efficient solution strategies in an MINLP process synthesizer MIPSYN, the successor of PROSYN (Kravanja and Grossmann, 1994), which could make solving such complex industrial problems possible. The present study focuses on the catalytic distillation processes. Since the use of a multilevel MINLP approach reduces the effects of nonconvexities, better solutions were obtained by multilevel than by single level MINLP strategies.

Towards Outcomes-Based Education of Computer-Aided Chemical Engineering

Abstract Chemical engineering education is nowadays increasingly supported by the use of various computational tools as the employers’ requirements for computing skills of graduates are growing too. However, students often acquire computational skills in an unsystematic manner due to a lack of defining and applying computer-based outcomes within the syllabuses suitable for the particular level of the Bologna three-cycle system. This paper bridges this gap by providing the review of the essential learning outcomes in the computer-aided chemical engineering education during all three cycles. The identified outcomes gradually progress from application-based competencies up to more advanced process modeling ones based on knowledge synthesis and creation. Accordingly, the educational strategies and curricula can be redesigned in order to integrate courses more efficiently both horizontally and vertically, and upgrade the use of computational tools.

Design of flexible process flow sheets with a large number of uncertain parameters

Abstract This work presents a solution strategy for design and synthesis of flexible process flow sheets with a large number of uncertain parameters, i.e. 50-100. The strategy relies on the standard two-stage stochastic formulation for fixed degree of flexibility, and combines the following solution techniques: a) prescreening analyses for reduction of uncertain parameters used in the approximate stochastic optimization, b) determination of the reduced set of critical points for feasibility, c) decomposition technique for individual stochastic suboptimizations of uncertain parameters and determination of basic point for approximate stochastic optimization, and d) approximate stochastic optimization for flexible design. These methods are merged into the strategy for process flow sheet optimization with a large number of uncertain parameters,which is illustrated with bioethanol case study.

Suitable Process Modelling for Proper Multi-Objective Optimization of Process Flow Sheets

Abstract This paper presents the use of different economic criteria during the generation of the Pareto optimal solutions within the multi-objective optimization of process flow sheets. It is shown that various economic criteria have a significant effect on the set of the Pareto solutions, which differ in the maximum values of the selected environmental criterion as well as in the ranges of their values. The reasons for variations are different stationary conditions of economic criteria. However, significant differences could be observed only if sufficiently accurate and precise process models were used, in which proper relations were established between the investment and the cash flow. In this case, the net present value would be the most appropriate economic measure for multi-objective optimization of process flow sheets.

Two-step multiobjective synthesis of sustainable chemical processes

This paper presents the development of a two-step multiobjective approach for the superstructural synthesis of sustainable chemical processes by means of mixed integer nonlinear programming methods. The first step is the usual economy-based synthesis. It is performed on a superstructure of technological and economic alternatives for identifying the best economic solution for use as a reference solution during the second step. The second step is sustainable multiobjective synthesis in which the superstructure is extended using sustainable alternatives. The economic and environmental criteria are identified for the multiobjective optimisation in order to generate sets of Pareto optimal solutions. The experiences gained by means of the two-step synthesis approach indicate that solutions can be obtained that have improved in both, often conflicting, criteria. This approach is illustrated using a well-known process of toluene hydrodealkylation.

Bi-objective MINLP optimization of an industrial water network via benchmarking

Abstract This paper presents an approach to water system retrofitting by estimating both the economic and environmental impacts of the water network design using bi-objective optimization. The environmental impact is evaluated via benchmarking. By using benchmarking, the decision maker could obtain insight not only into the environmental impact of a certain design belonging to the Pareto optimal solutions, but also into the competitiveness of the design within a particular production sector. The economic criterion used is the total cost of the water network and involves the freshwater cost, annual investment costs of the storage tank, piping and local treatment unit installation, and wastewater treatment cost. This approach uses a mixed-integer nonlinear programming (MINLP) model that enables water re-use and regeneration re-use in batch and semi-continuous processes. The Pareto front is generated using the Normal-Boundary Intersection (NBI) method. The proposed approach can be used for the separate integration of production sections, but also for joint integration of the sections via temporal decomposition. The proposed approach was applied to an industrial case study in a brewery.