C. McGreavy

Characterization of transport properties for fixed bed in terms of local bed structure and flow distribution

Abstract Transport processes can have a dramatic influence on the intrinsic chemical kinetics of catalytic processes and so are one of the important factors which must be allowed for in the design of fixed bed reactors. Unfortunately, the data and correlations currently available are often not appropriately matched to the models used in predicting behaviour. In particular, the design parameters are intimately linked to the flow distribution, which in turn depends on the structure of the bed. Failure to account for this level of detail has meant that correlations of the data are not suitable for use in analyzing reactors intended to operate at high performance, especially when stability is an issue. Data are presented which shows that earlier attempts to identify velocity distribution without relating it to bed structure have resulted in wrong characterization of the system and explains the difficulties arising from attempts to provide a rational basis for correlations of film heat transfer coefficients at the wall of packed tubes.

Application of wavelets and neural networks to diagnostic system development, 1, feature extraction

An integrated framework for process monitoring and diagnosis is presented which combines wavelets for feature extraction from dynamic transient signals and an unsupervised neural network for identification of operational states. Multiscale wavelet analysis is used to determine the singularities of transient signals which represent the features characterising the transients. This simultaneously reduces the dimensionality of the data and removes noise components. A modified version of the adaptive resonance theory is developed, which is designated ARTnet and uses wavelet feature extraction as the substitute of the data pre-processing unit. ARTnet is proved to be more effective in dealing with noise contained in the transient signals while retains being an unsupervised and recursive clustering approach. The work is reported in two parts. The first part is focused on feature extraction using wavelets. The second part describes ARTnet and its application to a case study of a refinery fluid catalytic cracking process.

Application of wavelets and neural networks to diagnostic system development, 2, an integrated framework and its application

Abstract A method for feature extraction from process dynamic transient signals using wavelet multiscale analysis was introduced in part 1 of this paper. In part 2 we describe an integrated framework combining wavelet feature extraction and an unsupervised neural network for identification of operational states. Application of the system to a refinery residual fluid catalytic cracking process is also presented.

Consistent evaluation of effective diffusion and reaction in pore networks

Abstract A model is developed based on the equivalent pore network conceptualize to account for diffusion and reaction processes in catalytic pore structures. The chromatographic reactor is adopted as a reference system so that the transit time moments of the response curve can be used to assess the effectiveness of the intraparticle diffusional transport and overall catalyst activity. As a consequence of the detailed description of the phenomenology at the microscopic pore scale, the results can then be expressed in terms of a Thiele modulus (o) which is a purely physico-chemical parameter, not dependent on the structural aspect of the pore space. At particular ranges of o values, a strong influence of the pore structure is detected onthe effective diffusion and effective reaction coefficients. It is then appropriate to say that the modelling strategy devised here not only sets a framework for the selection of a suitable catalyst pore structure. It also indicates that since the structure is known and consistently modelled, there is scope for design in terms of the reactor operational conditions to increase the catalyst performance.

Consistent measurement of diffusion coefficients for effectiveness factors

Abstract An examination has been made of how structural factors influence the measured values of effective diffusivity using the steady(Wicke-Kallenbach) and unsteady (chromatographic) state methods. For a solid with a typical bidisperse pore size distribution it is shown that the steady state method tends to give longer values. This suggests that the relative dominating characteristics are associated with the pore size distribution rather than the influence of dead-end pores. Such an explanation is internally consistent with tortuosity effects. The development of a reliable method of determining diffusivities for use in effectiveness factor calculations is dependent on the more precise identification of the structural characteristics along the lines presented here.

A multivariable coordinated control system based on predictive control strategy for FCC reactor-regenerator system

Abstract Fluid catalytic cracking unit (FCCU) involves complex interactive processes with constrained operating variables with are difficult to control. This paper describes a methodology for control which takes account of different situations where several manipulated variables encounter constraints derived from an on-line closed loop dynamic process model. A priority matrix used to capture the experience of skilled operators and to select suitable manipulated variables. A single step predicted control algorithm is integrated with the priority matrix into a coordinated control system which is easy to understand and is comparable in performance with dynamic matrix control (DMC).

Generalized criteria for parametric sensitivity and temperature runaway in catalytic reactors

Abstract Instability in the steady state operation of fixed bed catalytic reactors may arise from various sources, the most probable being parametric sensitivity as a consequence of the heterogeneity of the system, leading to temperature runaway, outside of the region of multiple steady states. A technique has been developed for predicting regions of temperature runaway in heterogeneous fixed bed reactors, which is not dependent on temperature gradients. The relationship between this form of instability and that due to multiple steady states is examined.

Flow of information through an artificial neural network

Abstract The patterns of information flow through an artificial neural network are examined in terms of how and why a network characterises input/output relationships, and what insight these patterns give as to the characteristics of the network that could be changed to improve its description of a system. The prediction of vapour-liquid equilibrium in terms of bubble-point conditions is used as a case study and shows that the network is capable of identifying the intrinsic characteristics of the system. However the accuracy of the prediction depends on the region of the input/output data space considered, drawing attention to the difficulties encountered by the empirical structuring of a network. The ability to identify the type and strength of the relationships between process variables indicates that a priori knowledge of the system could be used to relate parts of the network to dominant elements of the intrinsic model. This implies there could be advantages to be gained by exploiting knowledge of the system to maximise the information content captured by the network and establish a systematic way of designing its structure.

A Concurrent Engineering Environment for Chemical Manufacturing

Design and operation of process plants are now dominated by constraints imposed by safety, environmental impact, waste minimiza tion, cost effectiveness, controllability, and operability over the complete product and process life cycle including market analysis, conceptual and detailed design through to commissioning, operation, and disposal of products and decommissioning of the plant This means it inevitably in volves a large number of activities carried out by teams of engineers which draw on a variety of technologies that rely on diverse types of informa tion having complex data structures and relationships which must be shared by team members and integrated through computer based software packages Conflicting goals, uncertainty, and multiple solutions are crucial elements of these activities which are essentially about decision mak ing This paper describes the architectural and functional characteristics of a computer integrated concurrent engineering environment for life cy cle chemical manufacturing which allows project teams to work over a heterogeneous computer network Emphasis is placed on the key imple mentation issues with specific attention being paid to STEP-based chemical process data modelling, information sharing and communication, and distributed agent cooperation. The benefits this approach can bring are illustrated by considering the revamp design of a refinery fluid cata lytic cracking process, particularly in respect to the way the environment supports the project team

Prediction of the effectiveness factor and selectivity for highly exothermic reactions

Abstract The need to produce a simple but accurate model for determining the effectiveness factors and selectivities for complex reaction schemes has led many authors to make convenient but arbitrary simplifying assumptions in the formulation of their models. No systematic evaluation of these assumptions has been made, so no general conclusions are available as to the reliability of the various approximations. The most common simplifications are examined in detail, by applying them to a system of highly exothermic consecutive and parallel reactions. It is shown that, for the practical range of system parameters, a lumped thermal resistance model with external film resistance gives a very good approximation to the fully distributed model, enabling the effectiveness factor and selectivity to be obtained by the solution of an algebraic equation. Considerable savings in computation time result, and diffiiculties in the numerical solution, usual with problems of this kind, are avoided.