Abstract
In the recent years, chemical process modelling has dominated the research and
development from a process life-cycle perspective, with the need of providing
a flexible set of computer-aided tools for model development and maintenance
using an integrated environment. Such modelling tools should facilitate the
(re)use of modelling knowledge during the process life-cycle, since the information incorporated in the models can be easily translated for different target applications such as steady-state simulation, dynamic simulation, process optimisation or experimental design. Thus, it is useful to take advantage of the Computer-Aided Modelling Systems (CAMS), which must ensure the integration of existing tools and models into a software environment to support model definition, model evaluation, model analysis, model verification and model validation. An important modelling step is the model analysis that assures the model confidence, the input/parameter sensitivity, the model statistics, etc., using experimental data when it is possible. This step corresponds to a diagnosis based verification approach and may involve statistical analysis of the fitted (optimised) model parameters, which can help in the design of experiments with minimal effort. The main contribution of this thesis is the development of a modelling framework and its corresponding software for systematic chemical process modelling, called ICAS-MoT, which has all the aforementioned features. It was specifically designed with focus on the structure and reuse of models and to facilitate model implementation. From the application perspective, the model description, model analysis, model identification (parameter estimation) and static dynamic simulation using the proposed modelling framework are shown and highlighted through several case studies. A principal goal of this PhD-Thesis has been to use models for product and process design by testing and implementing them as fast as possible (in a reliable and efficient manner), writing the model equations without any programming effort, generating modules that can be used in another external software (via COM-object), and implementing several process/model configurations in the same environment. New features in the modelling language have been implemented to make it more powerful and easier to use, covering a wide range of applications.
development from a process life-cycle perspective, with the need of providing
a flexible set of computer-aided tools for model development and maintenance
using an integrated environment. Such modelling tools should facilitate the
(re)use of modelling knowledge during the process life-cycle, since the information incorporated in the models can be easily translated for different target applications such as steady-state simulation, dynamic simulation, process optimisation or experimental design. Thus, it is useful to take advantage of the Computer-Aided Modelling Systems (CAMS), which must ensure the integration of existing tools and models into a software environment to support model definition, model evaluation, model analysis, model verification and model validation. An important modelling step is the model analysis that assures the model confidence, the input/parameter sensitivity, the model statistics, etc., using experimental data when it is possible. This step corresponds to a diagnosis based verification approach and may involve statistical analysis of the fitted (optimised) model parameters, which can help in the design of experiments with minimal effort. The main contribution of this thesis is the development of a modelling framework and its corresponding software for systematic chemical process modelling, called ICAS-MoT, which has all the aforementioned features. It was specifically designed with focus on the structure and reuse of models and to facilitate model implementation. From the application perspective, the model description, model analysis, model identification (parameter estimation) and static dynamic simulation using the proposed modelling framework are shown and highlighted through several case studies. A principal goal of this PhD-Thesis has been to use models for product and process design by testing and implementing them as fast as possible (in a reliable and efficient manner), writing the model equations without any programming effort, generating modules that can be used in another external software (via COM-object), and implementing several process/model configurations in the same environment. New features in the modelling language have been implemented to make it more powerful and easier to use, covering a wide range of applications.
| Original language | English |
|---|
| Publisher | Technical University of Denmark |
|---|---|
| Number of pages | 304 |
| ISBN (Print) | 87-91435-37-4 |
| Publication status | Published - May 2006 |