Lars Nyström

Fuzzy Neural Network as Instance Generator for Case-Based Reasoning System: An Example of Selection of Heat Exchange Equipment in Mixing Tanks

The early stages of design are characterised by imprecise and incomplete information about the features and properties of targeted product or process. Equipment selection is one of the decisions to be taken in an initial phase of design. Mixing is an example of process where the selection of the equipment is practically realised ‘by hand’. An application of rule-based or object-oriented tools in the selection of mixing equipment encounters many problems. The main reasons are the lack of well-structured knowledge and the great diversity of mixing problems. Case-based reasoning constitutes a good tool to handle the problems of mixing equipment selection. An instance generator for case-based reasoning system is described in this work. A fuzzy neural network system has been used to produce the cases. The following types of the equipment for heat exchange in stirred vessels have been considered: plain jacket, plain jacket with agitation nozzles, plain jacket with spiral baffles, dimple jacket, halfpipe jacket, helical coil, vertical coil and vertical plate coil.

Design and synthesis of multicomponent thermally coupled distillation flowsheets

Abstract The design and synthesis of thermally coupled distillation flowsheets for separations of five-component mixtures are studied. Four types of possible configurations are identified when simple and complex columns are both considered in a flowsheet. A universal design procedure is developed for design of any types of the identified configurations based on the abstraction of the three basic units in the flowsheets. Two examples demonstrated that this shortcut design method can be used in design of any types of the identified multicomponent thermally coupled distillation flowsheets, as well as give very good initializations for rigorous simulation of such configurations. Moreover, with a proposed computer representation of all the types of the feasible configurations, a synthesis algorithm is developed for synthesizing of multicomponent complex distillation flowsheets with both simple and complex columns. It is practical by the proposed methods for optimal design of multicomponent distillation systems in an extended search space to include the complex distillation flowsheets for industrial problems.

The synthesis of thermally coupled distillation flowsheets for separations of five-component mixtures

Abstract The thermally coupled distillation flowsheets for separation of five-component mixtures are studied based on the total annual costs. Fourteen simple column sequences and some feasible thermally coupled configurations with side strippers and/or side rectifiers are considered. The costs of the flowsheets are calculated and compared for several real five-component mixtures with various relative volatility distributions and a wide range of feed compositions. Several heuristics are proposed for the synthesis of multicomponent thermally coupled distillation flowsheets.

Case-based reasoning system for mixing equipment selection

Abstract The use of case-based reasoning in the mixing equipment selection is promising. Its main advantage is use of high quality, verified knowledge and easiness of updating. The main functions of the system are a systematic documentation of the cases and an accumulation of experience. Case-based reasoning (CBR) is an alternative approach to rule-based and object-oriented systems. CBR is based on the retrieval and adaptation of the old solutions to the new problems. There are five steps in CBR: presentation of a new problem, retrieval of the most similar cases, adaptation of the most similar old solutions, validation of the current solution and updating of the system by adding the verified solution to the database of cases. The main advantages of CBR are the easy knowledge acquisition and simple explanatory system. Other important features are easy maintenance and learning capabilities. Determination of the most important features in mixing equipment selection and their importance is studied in this work. The main features discussed are liquid volume and rheological parameters, content of particles and/or gas bubbles as well as the type of application and the aim of the process. The retrieval search has been organized by using flat memory and serial search. The following data are given as the output values of the system: impeller and entering type, scale of agitation, mixing power, number of impellers and geometry of tank. An example of CBR applications to the mixing equipment selection in wastewater industry is given.

Selection of internals for reactive distillation column¿case-based reasoning approach

The design of reactive distillation (RD) systems is considerably more complex than design of the individual, conventional reactors and distillation columns. One of the main issues in the design of RD column is a definition of type and specification of geometric features of the internal devices. In this work, there is described a design supporting tool for pre-selection of packing type for RD column. Case-based reasoning has been applied as a basic method for decision support. The computer system for pre-design of reactive column internals has been built as an implementation of developed methodology. The acid catalysed reaction of 2-methylpropylacetate synthesis was used to test the system. The CBR system can significantly reduce the development time for design of reactive distillation column and suggests a good start points for further design activity.

Selection of Internals for Reactive Distillation Column—Case-based Reasoning Approach

The design of reactive distillation (RD) systems is considerably more complex than design of the individual, conventional reactors and distillation columns. One of the main issues in the design of RD column is a definition of type and specification of geometric features of the internal devices. In this work, there is described a design supporting tool for pre-selection of packing type for RD column. Case-based reasoning has been applied as a basic method for decision support. The computer system for pre-design of reactive column internals has been built as an implementation of developed methodology. The acid catalysed reaction of 2-methylpropylacetate synthesis was used to test the system. The CBR system significantly reduces the development time for design of reactive distillation column and suggests a good start points for further design activity.

Adaptation phase in case-based reasoning system for process equipment selection

Abstract Case-based reasoning (CBR) is an alternative approach to rule case-based and object-oriented systems. CBR is based on the retrieval and adaptation of the old solutions to the new problems. There are five steps in CBR: introduction of a new problem, retrieval of the most similar cases, adaptation of the most similar solutions, validation of the current solution and updating of the system by adding the verified solution to the database of cases. Adaptation phase is a crucial element in obtaining a good quality solutions. Up to now, there were no proposals for the automatization of the adaptation phase in the systems for the equipment selection. In the presented system the retrieval search has been organized by using flat memory and serial search. The adaptation is realized by the fuzzy calculation of the membership function built on the basis of the most similar retrieved cases. The main advantage of this system is use of high quality, verified knowledge and easiness of updating.

Concurrent engineering: Robust design in fuzzy environment

Abstract The question of robust designis formulated and solved as fuzzy optimization problem. The process equations are equality constraints. The nonequality constraints are the criteria resulting from the “house of quality” method. A numerical example is presented to illustrate the basic idea of the robust design with fuzzy parameters.

A conflict-based approach for process synthesis with wastes minimization

Abstract This paper presents a conflict-based analysis approach for process synthesis with wastes minimization for dealing with the conflicts of multi-objective nature in the early design stage. A three-level hierarchical procedure is proposed for carrying out the identification of waste sources, the generation of pollution prevention alternatives and mathematical optimization. It has been shown that the superstructure aimed at the wastes minimization can be systematically formulated on the basis of the proposed strategy. The proposed approach has an important potential for achieving a more efficient solution in the stage of mathematical optimization. The approach is illustrated by a case study: air-based direct oxidation process for the production of ethylene oxide.

Case-based reasoning for equipment selection using rough sets analysis in adaptation phase

Abstract Case-based reasoning (CBR) is an approach based on the reuse of the past experiences in finding the solutions to the new, similar problems. Adaptation is the most difficult and the most important step in the succesful use of CBR system, especially if in the certain applications e.g. equipment design the non-numerical parameters have to be adapted. The objective of the paper is a presentation of the adaptation method of CBR that is based on rough sets. Two examples are given to illustrate the applicability of the different adaptation methods