Juraj Labovský

Impact of mathematical model selection on prediction of steady state and dynamic behaviour of a reactive distillation column

The objective of this paper was to compare the prediction of the equilibrium (EQ) and nonequilibrium (NEQ) models during safety analysis of a reactive distillation column focusing on the identification of hazardous situations or particular operability problems. The safety and operability analyses are based on application of the HAZOP procedure integrated with a mathematical model with the aim to determine the column response to deviations from normal operation conditions or during a nonstandard procedure, e.g. the start-up of the reactive distillation column. A significant part of the safety and operability problems analysis is the identification of multiple steady states and their stability. A reactive distillation column can in general exhibit multiple steady states which reduce the column operability and controllability during perturbations of the manipulated variables and particularly during the start-up and shut down procedures. The EQ and NEQ models were compared focusing on prediction of the multiple steady states phenomenon and of the consequences which can result from this phenomenon.

Safety analysis and risk identification for a tubular reactor using the HAZOP methodology

A model approach to Hazard and Operability (HAZOP) analysis is presented based on the mathematical modeling of a process unit where both the steady-state analysis, including the analysis of the steady states multiplicity and stability, and the dynamic simulation are used. Heterogeneous tubular reactor for the ethylene oxide production from ethylene and oxygen was chosen to identify potential hazards for real system. The computer code DYNHAZ was developed consisting of a process simulator and a generator of the HAZOP algorithm.

HAZOP study of a fixed bed reactor for MTBE synthesis using a dynamic approach

A methodology for hazard investigation based on the integration of a mathematical model approach into hazard and operability analysis is presented. This approach is based on mathematical modelling of a process unit where both steady-state analysis, including analysis of the steady states multiplicity and stability, and dynamic simulation are used. The dynamic simulation serves for the investigation of consequences of failures of the main controlled parameters, i.e. inlet temperature, feed temperature and feed composition. This simulation is also very useful for the determination of the influence of failure duration on the reactor behaviour. On the other hand, the steady state simulation can predict the reactor behaviour in a wide range of failure magnitude and determine the parametric zones, where shifting from one steady state to another one may occur. A fixed bed reactor for methyl tertiary-butyl ether synthesis was chosen to identify potential hazard and operational problems of a real process.

CFD-based atmospheric dispersion modeling in real urban environments

The process of CFD model application for atmospheric dispersion modeling is presented. Increasing the CPU power opens new possibilities of the CFD approach application for consequence analysis in real complex urban environments. As successful CFD simulation is directly dependent on the quality and complexity of the computational mesh, a new methodology of transferring the Geographic Information System (GIS) data to the computational mesh can be utilized. A user software for importing and manipulation with the GIS data and their subsequent transfer to an instructional file for the generation of the computational mesh was prepared and tested. The introduced methodology is relatively simple and it requires only a small amount of input data. The process of creating a computational mesh is very straightforward and fast, which enables the application of CFD modeling in urban environments in all fields of engineering applications in safety analysis. Several recommendations concerning proper definition of boundary conditions for atmospheric dispersion modeling were summarized. The presented approach was tested on a realistic case study of liquefied chlorine release in a real town. Results obtained by the CFD approach were compared with those obtained by a simpler but standard integral model.

Utilization of parallel computing in chemical engineering

Abstract The main objective of the presented work was to explore the possibilities of parallel computing utilization in chemical engineering. Parallel computers and principles of parallel computing are in brief described in Introduction. The next part exposes the possibilities of parallel programming in Matlab and C# programming language environment. The next three parts provide case studies of parallel computing in chemical engineering. Each example of the benefits of HPC involves a comparison with its serial equivalents.

Ammonia synthesis fundamentals for a model-based HAZOP study

Abstract Hazard and operability (HAZOP) analysis is a highly disciplined process hazard analysis (PHA) technique based on the exploration of the effects of process variables deviations. Inconveniences of a conventional HAZOP study are its time-consuming character and high cost. The principal objective of this paper is to present a new methodology for hazard identification of a selected chemical production process. Model-based HAZOP study is a very robust tool for predicting a systems response to deviations from design or operation conditions. An approach based on the mathematical modelling of a process can help to identify sources of hazard that could be overlooked by conventional PHA techniques. A case study focused on the multiple steady states phenomenon in an ammonia synthesis reactor is presented. The process simulation was performed using the Aspen HYSYS v8.4 process modelling environment. Nonlinear behaviour of the investigated fixed-bed reactor system was confirmed by an accident in an industrial ammonia synthesis reactor. The analysed system exhibited the feed temperature and pressure dependence of various operation parameters. This fact indicates the presence of multiple steady states. From the safety analysis point of view, switching between steady states can lead to process hazards.

Design, optimization and safety analysis of a heterogeneous tubular reactor by using the HAZOP methodology

Abstract In this contribution, a model approach to Hazard and Operability (HAZOP) analysis is presented. This analysis is based on the mathematical modelling of a process unit, where both the steady-state analysis, (including the analysis of the steady states multiplicity and stability) and the dynamic simulation are used. A heterogeneous tubular reactor for the production of ethylene oxide and a reactor for the production of MTBE were chosen to identify potential hazards for a real system. The computer code DYNHAZ consisting of a process simulator and a generator of the HAZOP algorithm was developed.

Fault propagation behavior study of CSTR in HAZOP

This paper discusses the role of process modeling in safety analysis. Process modeling is applied in the fault propagation behavior study of CSTR chemical production. For that purpose, HAZOP methodology and continuation analysis were used. The proposed hazard identification methodology involves analysis of steady-state multiplicity and safe operating conditions as well as those which can shift process units from one steady state to another. All presented case studies are also supported by system dynamic simulations, essential to detect oscillatory thermal instability. In this paper, N-oxide alkylpyridines production process was chosen to identify potential hazard and operational problems. Presented dynamic simulations represent an analysis of the system response to step changes in the key operating parameters. The effect of deviations of three key parameters on the reactor safe operation was investigated. The proposed numerical algorithms represent a mathematical engine of the simulation module within an automated model-based HAZOP analysis tool.

Smart software system solution for model-based hazard identification of complex industrial processes

Abstract To satisfy the ever-growing needs of modern civilization, society and industry are experiencing transformation through automation and digitalization. The present work deals with process safety automation issues in chemical industry with a particular focus on computer aided hazard identification based on mathematical modeling and process simulation. In this paper, a smart software system solution combining HAZOP (HAZard and OPerability) study principles and computer simulation of complex industrial processes employing Aspen HYSYS is proposed. An ammonia industrial production unit has been chosen as a case study to demonstrate the applicability and application procedure of the proposed software tool for model-based HAZOP study. The results also indicate that the proposed software tool can supplement process design and intensification studies employing Aspen HYSYS.

Estimation of thermal effects on receptor from pool fires

Abstract The aim of this contribution is to provide an overview of the calculation procedures of risk analysis, that is, the effects and consequences of pool fires. Fires and explosions are the most significant and most common causes of damage to equipment and of injuries and death in industry. Damages are a direct consequence of the generated heat flux. Mathematical tools for the prediction of heat flux at a distance can be divided into four classes: semi-empirical models, field models, integral models and zone models. Semi-empirical modeling is a relatively simple technique providing models predicting heat flux at a distance. There are two types of semi-empirical models: point source models and surface emitter models. By their nature, semi-empirical models depend strongly on experimental data. Correlations are able to describe the general features of a fire. Semi-empirical models are ideal for routine hazard assessment purposes because they are mathematically simple, and hence easily understood. However, if more models describing the same phenomenon are available, significant differences in the heat flux prediction can be expected. In this contribution, differences in the prediction of the heat flux from pool fires are discussed.