Process and Plant Safety E-Book

• Herausgeber: Wiley-VCH
• Kategorie: Fachliteratur
• Sprache: Englisch

Seitenzahl: 554

Veröffentlichungsjahr: 2012

Beschreibung

The safe operation of plants is of paramount importance in the chemical, petrochemical and pharmaceutical industries. Best practice in process and plant safety allows both the prevention of hazards and the mitigation of consequences. Safety Technology is continuously advancing to new levels and Computational Fluid Dynamics (CFD) is already successfully established as a tool to ensure the safe operation of industrial plants. With CFD tools, a great amount of knowledge can be gained as both the necessary safety measures and the economic operation of plants can be simultaneously determined. Young academics, safety experts and safety managers in all parts of the industry will henceforth be forced to responsibly judge these new results from a safety perspective. This is the main challenge for the future of safety technology. This book serves as a guide to elaborating and determining the principles, assumptions, strengths, limitations and application areas of utilizing CFD in process and plant safety, and safety management. The book offers recommendations relating to guidelines, procedures, frameworks and technology for creating a higher level of safety for chemical and petrochemical plants. It includes modeling aids and concrete examples of industrial safety measures for hazard prevention.

Leseprobe

Contents

Cover

Related Titles

Title Page

Copyright

Series Page

Preface

List of Contributors

Chapter 1: Computational Fluid Dynamics: The Future in Safety Technology!

Chapter 2: Organized by ProcessNet: Tutzing Symposion 2011 ‘CFD – its Future in Safety Technology’

2.1 ProcessNet – an Initiative of DECHEMA and VDI-GVC

2.2 A Long Discussed Question: Can Safety Engineers Rely on Numerical Methods?

Chapter 3: CFD and Holistic Methods for Explosive Safety and Risk Analysis

3.1 Introduction

3.2 Deterministic and Probabilistic Design Tasks

3.3 CFD Applications on Explosions and Blast Waves

3.4 Engineering Methods: The TNT Equivalent

3.5 QRA for Explosive Safety

3.6 Summary and Outlook

References

Part One: CFD Today – Opportunities and Limits if Applied to Safety Techology

Chapter 4: Status and Potentials of CFD in Safety Analyses Using the Example of Nuclear Power

4.1 Introduction

4.2 Safety and Safety Analysis of Light Water Reactors

4.3 Role and Status of Fluid Dynamics Modeling

4.4 Expected Benefits of CFD in Nuclear Reactor Safety

4.5 Challenges

4.6 Examples of Applications

4.7 Beyond-Design-Based Accidents

4.8 Summary

List of Acronyms

References

Part Two: Computer or Experimental Design?

Chapter 5: Sizing and Operation of High-Pressure Safety Valves

5.1 Introduction

5.2 Phenomenological Description of the Flow through a Safety Valve

5.3 Nozzle/Discharge Coefficient Sizing Procedure

5.4 Sizing of Safety Valves Applying CFD

5.5 Summary

List of symbols

Subscripts

References

Chapter 6: Water Hammer Induced by Fast-Acting Valves – Experimental Studies, 1D Modeling, and Demands for Possible Future CFX Calculations

6.1 Introduction

6.2 Multi-Phase Flow Test Facility

6.3 Extension of Pilot Plant Pipework PPP for Software Validation

6.4 Experimental Set-Up

6.5 Experimental Results

6.6 2 Case Studies of Possible Future Application of CFX

6.7 Possible Chances and Difficulties in the Use of CFX for Water Hammer Calculations

6.8 CFD – The Future of Safety Technology?

References

Chapter 7: CFD-Modeling for Optimizing the Function of Low-Pressure Valves

References

Part Three: Fire and Explosions – are CFD Simulations Really Profitable?

Chapter 8: Consequences of Pool Fires to LNG Ship Cargo tanks

8.1 Introduction

8.2 Evaluation of Heat Transfer

8.3 CFD-Calculations

8.4 Conclusions

References

Chapter 9: CFD Simulation of Large Hydrocarbon and Peroxide Pool Fires

9.1 Introduction

9.2 Governing Equations

9.3 Turbulence Modeling

9.4 Combustion Modeling

9.5 Radiation Modeling

9.6 CFD Simulation

9.7 Results and Discussion

9.8 Conclusions

9.9 CFD – The Future of Safety Technology?

Nomenclature

References

Chapter 10: Modeling Fire Scenarios and Smoke Migration in Structures

10.1 Introduction

10.2 Hierarchy of Fire Models

10.3 Balance Equations for Mass, Momentum, and Heat Transfer (CFD Models)

10.4 Zone Models

10.5 Plume Models

10.6 Computational Examples

10.7 Conclusions

10.8 CFD – The Future of Safety Technology?

Annex 1: Set of Partial Differential Equations for CFD Models (differential form)

List of symbols

References

Part Four: CFD Tomorrow – The Way to CFD as a Standard Tool in Safety Technology

Chapter 11: The ERCOFTAC Knowledge Base Wiki – An Aid for Validating CFD Models

11.1 Introduction

11.2 Structure of the Knowledge Base Wiki

11.3 Content of the Knowledge Base

11.4 Interaction with Users

11.5 Concluding Remarks

AC Index

UFR Index

Chapter 12: CFD at its Limits: Scaling Issues, Uncertain Data, and the User's Role

12.1 Numerics and Under-Resolved Simulations

12.2 Uncertainties

12.3 Theory and Practice

12.4 Conclusions

References

Chapter 13: Validation of CFD Models for the Prediction of Gas Dispersion in Urban and Industrial Environments

13.1 Introduction

13.2 Types of CFD Models

13.3 Validation Data

13.4 Wind Tunnel Experiments

13.5 Summary

Acknowledgements

References

Chapter 14: CFD Methods in Safety Technology – Useful Tools or Useless Toys?

14.1 Introduction

14.2 Characteristic Properties of Combustion Systems

14.3 Practical Problems

14.4 Outlook

References

Part Five: Dynamic Systems – Are 1D Models Sufficient?

Chapter 15: Dynamic Modeling of Disturbances in Distillation Columns

15.1 Introduction

15.2 Dynamic Simulation Model

15.3 Case Study

15.4 CFD- The Future of Safety Technology?

15.5 Nomenclature

References

Chapter 16: Dynamic Process Simulation for the Evaluation of Upset Conditions in Chemical Plants in the Process Industry

16.1 Introduction

16.2 Application of Dynamic Process Simulation

16.3 Conclusion

16.4 Dynamic Process Simulation – The Future of Safety Technology?

Chapter 17: The Process Safety Toolbox – The Importance of Method Selection for Safety-Relevant Calculations

17.1 Introduction – The Process Safety Toolbox

17.2 Flow through Nitrogen Piping During Distillation Column Pressurization

17.3 Tube Failure in a Wiped-Film Evaporator

17.4 Phenol-Formaldehyde Uncontrolled Exothermic Reaction

17.5 Computational Fluid Dynamics – Is It Ever Necessary?

17.6 Computational Fluid Dynamics – The Future of Safety Technology?

Acknowledgments

References

Chapter 18: CFD for Reconstruction of the Buncefield Incident*

18.1 Introduction

18.2 Observations from the CCTV Records

18.3 CFD Modeling of the Vapor Cloud Dispersion

18.4 Conclusions

18.5 CFD: The Future of Safety Technology?

Acknowledgments

References

Part Six: Contributions for Discussion

Chapter 19: Do We Really Want to Calculate the Wrong Problem as Exactly as Possible? The Relevance of Initial and Boundary Conditions in Treating the Consequences of Accidents

19.1 Introduction

19.2 Models

19.3 Case Study

19.4 Conclusions

Appendix

References

Chapter 20: Can Software Ever be Safe?

20.1 Introduction

20.2 Basics

20.3 Software Errors and Error Handling

20.4 Potential Future Approaches

20.5 CFD - The Future of Safety Technology?

References

Chapter 21: CFD Modeling: Are Experiments Superfluous?

References

Index

Related Titles

Stoessel, F.Thermal Safety of Chemical ProcessesRisk Assessment and Process Design 2008 ISBN: 978-3-527-31712-7

Reniers, G. L. L.Multi-Plant Safety and Security Management in the Chemical and Process Industries 2010 ISBN: 978-3-527-32551-1

Nemerow, N. L., Agardy, F. J., Salvato, J. A.Environmental EngineeringEnvironmental Health and Safety for Municipal Infrastructure, Land Use and Planning, and Industry 2009 ISBN: 978-0-470-08305-5

Center for Chemical Process Safety (CCPS)Inherently Safer Chemical ProcessesA Life Cycle Approach 2009 ISBN: 978-0-471-77892-9

Center for Chemical Process Safety (CCPS)Guidelines for Hazard Evaluation Procedures 2008 ISBN: 978-0-471-97815-2

Center for Chemical Process Safety (CCPS)Guidelines for Performing Effective Pre-Startup Safety Reviews 2007 ISBN: 978-0-470-13403-0

Bender, H. F., Eisenbarth, P.Hazardous ChemicalsControl and Regulation in the European Market 2007 ISBN: 978-3-527-31541-3

Bhagwati, K.Managing SafetyA Guide for Executives 2006 ISBN: 978-3-527-31583-3

Vogel, G. H.Process DevelopmentFrom the Initial Idea to the Chemical Production Plant 2005 ISBN: 978-3-527-31089-0

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.: applied for

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de.

© 2012 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Composition Thomson Digital, Noida, India

Cover Design Schulz Grafik-Design, Fugönheim

Print ISBN: 978-3-527-33027-0

ePDF ISBN: 978-3-527-64574-8

ePub ISBN: 978-3-527-64573-2

Mobi ISBN: 978-3-527-64575-6

This Book was Published with the Generous Support from the Following Companies

Linde AG, Pullach, Germany

BASF SE, Ludwigshafen, Germany

Merck KGaA, Darmstadt, Germany

Braunschweiger Flammenfilter GmbH, Braunschweig, Germany

LESER GmbH & Co. KG, Hamburg, Germany

ANSYS Germany GmbH, Darmstadt, Germany

Germanischer Lloyd SE, Hamburg, Germany

Bopp & Reuther, Mannheim, Germany

consilab Gesellschaft für Anlagensicherheit mbH, Frankfurt, Germany

EPSC, Rugby, Warwickshire, Great Britain

REMBE GmbH, Brilon, Germany

Siemens AG, Frankfurt, Germany

Preface

Nice toy or reliable toolbox? There is no precise opinion about Computational Fluid Dynamic (CFD) results. Today, Computational Fluid Dynamics (CFD) calculations are standard in many applications with the exception of the conservative subject area of safety engineering. This was the focus of discussions between experts in the fields of safety engineering and CFD at a symposium in 2011 entitled “CFD - the future in safety technology?” (50th Tutzing symposium, 2011). When human lives or comparable commodities are at risk, then typically only conventional calculation methods are employed. As demonstrated by the eruption of the Eyjafjallajökull volcano on Iceland in 2010, it did not make sense to simulate the trajectory of the ash cloud with CFD tools when the effects of the ash particles on the jet engines of an airplane could not be satisfactorily modeled. On the other hand, CFD is already being used successfully in many areas of safety engineering, for example, for the investigation of an incident in Hemel Hempstead, England, during which an explosion led to a devastating fire at the Buncefield oil-products storage depot (the fifth largest of its type in the United Kingdom). So, when should CFD simulations be used in safety engineering?

The possible applications and limitations of CFD modeling in the area of safety engineering are discussed in this book, which covers a variety of topics, including the simulation of flow-through fittings, the consequences of fire and the dispersion of gas clouds, pressure relief of reactors, and the forensic analysis of incidents.

The contributions to this book propose the establishment of CFD programs as reliable standard tools in safety engineering. In order to accomplish this goal in the future, experts from the fields of safety engineering and CFD simulation must regularly exchange their knowledge and especially their different methodologies for dealing with certain topics. Hopefully this book will act as a catalyst for the development of deeper synergy between the two groups.

2011

Jürgen Schmidt

List of Contributors

G.T. Atkinson Health & Safety Laboratory Mathematical Sciences Unit Fluid Dynamics Team Harpur Hill Buxton SK17 9JN UK

Henning Bockhorn Karlsruhe Institute of Technology Engler-Bunte-Institute Division of Combustion Technology (EBI/VBT) Engler-Bunte-Ring 1 76131 Karlsruhe Germany

Andreas Dudlik Fraunhofer-Institut für Umwelt-, Sicherheits- und Energietechnik (Fraunhofer UMSICHT) Osterfelder Straße 3 46047 Oberhausen Germany

Robert Fröhlich Fraunhofer-Institut für Umwelt-, Sicherheits- und Energietechnik (Fraunhofer UMSICHT) Osterfelder Straße 3 46047 Oberhausen Germany

Simon E. Gant Health & Safety Laboratory Mathematical Sciences Unit Fluid Dynamics Team Harpur Hill Buxton SK17 9JN UK

Ulrich Hauptmanns Otto-von-Guericke-Universität Magdeburg Fakultät für Verfahrens- und Systemtechnik Institut für Apparate- und Umwelttechnik (IAUT) Abteilung Anlagentechnik und Anlagensicherheit Universitätsplatz 2 39106 Magdeburg Germany

Frank Helmsen Braunschweiger Flammenfilter GmbH Industriestr. 11 38110 Braunschweig Germany

W. Henk Linde AG Dr.-Carl-von-Linde-Str. 6–14 82049 Pullach Germany

Tobias Kirchner Braunschweiger Flammenfilter GmbH Industriestr. 11 38110 Braunschweig Germany

Rupert Klein Freie Universität Berlin FB Mathematik und Informatik Institut für Mathematik Arnimallee 6 14195 Berlin-Dahlem Germany

Arno Klomfass Fraunhofer Institute for High Speed Dynamics, Ernst-Mach-Institute Freiburg Germany

Christian Knaust BAM Bundesanstalt für Materialforschung und -prüfung Division 7.3 ‘Fire Engineering’ Unter den Eichen 87 12205 Berlin Germany

M. Koch Linde AG Dr.-Carl-von-Linde-Str. 6–14 82049 Pullach Germany

Ulrich Krause BAM Bundesanstalt für Materialforschung und -prüfung Division 7.3 ‘Fire Engineering’ Unter den Eichen 87 12205 Berlin Germany

Andy Jones Evonik Degussa Corporation Process Technology and Engineering 4301 Degussa Rd. Theodore, AL 36590-0606 USA

B. Jörgensen LESER GmbH & Co. KG Wendenstrasse 133–135 20537 Hamburg Germany

Bernd Leitl University of Hamburg Meteorological Institute KlimaCampus Bundesstr. 55 20146 Hamburg Germany

Tina Mattes Technische Universität München Institute of Automation and Information Systems Automation Group Boltzmannstr. 15 85748 Garching Germany

D. Moncalvo LESER GmbH & Co. KG Wendenstrasse 133–135 20537 Hamburg Germany

Aristides Morillo BASF SE Carl-Bosch-Straße 38 67063 Ludwigshafen Germany

Matthais Münch Freie Universität Berlin FB Mathematik und Informatik Institut für Mathematik Arnimallee 6 14195 Berlin-Dahlem Germany

Karl Niesser Linde AG Dr.-Carl-von-Linde-Str. 6–14 82049 Pullach Germany

Wolfgang Peschel BASF SE Carl-Bosch-Straße 38 67063 Ludwigshafen Germany

Norbert Pfeil BAM Federal Institute for Materials Research and Testing ProcessNet Safety Engineering Section 12200 Berlin Germany

Horst-Michael Prasser ETH Zürich Institut für Energietechnik ML K 13 Sonneggstr. 3 8092 Zürich Switzerland

Frederik Rabe BAM Bundesanstalt für Materialforschung und -prüfung Division 7.3 ‘Fire Engineering’ Unter den Eichen 87 12205 Berlin Germany

Wolfgang Rodi Karlsruhe Institute of Technology Institute for Hydromechanics Kaiserstr.12 76128 Karlsruhe Germany

Stefan Schälike University of Duisburg-Essen Institute of Chemical Engineering I Universitätsstr. 5–7 45141 Essen Germany

and

BAM Federal Institute for Material Research and Testing Division 2.2 ‘Reactive Substances and Systems’ Unter den Eichen 87 12205 Berlin Germany

Michael Schatzmann University of Hamburg Meteorological Institute KlimaCampus Bundesstr. 55 20146 Hamburg Germany

Frank Schiller Technische Universität München Institute of Automation and Information Systems Automation Group Boltzmannstr. 15 85748 Garching Germany

Jürgen Schmidt [email protected] Karlsruhe Institute of Technology Faculty of Chemical and Process Engineering Engler Bunte Institute Engler Bunte Ring 1 76131 Karlsruhe Germany

and

BASF SE Safety an Fluid Flow Technology 67056 Ludwigshafen Germany

Benjamin Scholz Germanischer Lloyd SE Department of Environmental Research Brooktorkai 18 20457 Hamburg Germany

Axel Schönbucher University of Duisburg-Essen Institute of Chemical Engineering I Universitätsstr. 5–7 45141 Essen Germany

Daniel Staak Technische Universität Berlin Berlin Institute of Technology Sekretariat KWT 9 Straße des 17. Juni 135 10623 Berlin Germany

Klaus Thoma Fraunhofer Institute for High Speed Dynamics, Ernst-Mach-Institute Freiburg Germany

Iris Vela University of Duisburg-Essen Institute of Chemical Engineering I Universitätsstr. 5–7 45141 Essen Germany

Klaus-Dieter Wehrstedt BAM Federal Institute for Material Research and Testing Division 2.2 ‘Reactive Substances and Systems’ Unter den Eichen 87 12205 Berlin Germany

Anton Wellenhofer Linde AG Process & Environmental Safety – TS Dr.-Carl-von-Linde-Str. 6–14 82049 Pullach Germany

Günter Wozny Technische Universität Berlin Berlin Institute of Technology Sekretariat KWT 9 Straße des 17. Juni 135 10623 Berlin Germany

Gerd-Michael Wuersig Germanischer Lloyd SE Department of Environmental Research Brooktorkai 18 20457 Hamburg Germany

Chapter 1

Computational Fluid Dynamics: The Future in Safety Technology!

Jürgen Schmidt

Safety engineering is based on reliable and conservative calculations. With Computational Fluid Dynamics (CFD) tools, the knowledge of certain physical processes is deepened significantly. However, such programs are currently not standard. In safety engineering more stringent demands for accuracy must be set, for example, as compared to methods for the optimization of plants. The methods must, among other things, be sufficiently validated by experiences or experimental data and fully documented (method transparency). In addition, they must be comprehensible, reproducible, and economical to apply. The necessary demands on precision can usually only be met by model developers, program suppliers, and users of the CFD codes (common sense application).

The developers of models must document their models, and the assumptions under which the models were derived must be fully understandable. Only if the application range is carefully described can a responsible transfer to other fluids and parameter rages take place at some later time. Unlike simple empirical correlations, CFD models, with their many sub-models, often appear complex and not transparent. The validation is usually done only on certain individual data points or by measuring global parameters such as pressures and mass flows. This makes it difficult to assess whether a method is more generally applicable in practice. Margins of error cannot be estimated, or only very roughly. There are relatively for model validations for typical questions in the field of safety engineering. However, even there only models and methods with sufficiently well-known uncertainties should be applied.

It is still not enough if only the model application ranges are transparent. In addition it should be possible to review the CFD program codes. Most codes are not currently open source. Moreover, frequent version changes and changes in the program codes complicate any review. Generally accepted example calculations which can be used for revalidation (safety-relevant test cases) are usually lacking. There are often demands for open-source programs among the safety experts. This certainly facilitates the testing of models. On the other hand, in practice it is then only barely comprehensible what changes were made in a program in any particular case.

CFD calculations are reasonably possible in safety technology only with a good education and disciplined documentation of the results.

A university education should provide any students with:

A fundamental knowledge of numerical modeling, including an understanding of the mathematical solution procedures, their use and application boundaries, and the influence of initial and boundary conditions.Experience in the application of safety-related models and methods.Analytical skills to be able to evaluate safety-related calculation results for abnormal operation conditions on the basis of experimental studies performed with other fluids and under normal conditions.A training in how to assess the self-evident plausibility of calculation results with the help of shortcut methods.A technical safety mindset and approach in dealing with computational methods and the evaluation of results.

These requirements are currently being taught in their entirety in hardly any of the major universities. Students often lack the mathematical skills of numerical modeling, a deeper understanding of turbulence models, or simply the experimental experience to assess calculation results. At some institutions, CFD codes are used as black boxes. Student training needs to be adjusted. A major effort to teach these necessary skills is essential.

Particularly in safety engineering, CFD programs are currently (still) used by a relatively small circle of experts. Careful documentation of results in this area is particularly important. In addition to input and output data, the initial and boundary values as well as the chosen solution method and model combinations must be recorded. These data are often very extensive. It may therefore be useful to keep all programs and necessary files long-term on appropriate computers. Again, it would be helpful if certain practices were well established and documented as standard – this is lacking in safety engineering.

In addition to the required computational results, sensitivity analysis of individual parameters is desirable. With CFD programs a deeper understanding of the physical processes can arise from that analysis. Alternatively it may turn out that the chosen combination of models is not suitable to solve a specific problem. Even with sensitivity analysis, the user has the duty to responsibly perform and document them as an additional part of the actual calculations.

For a third party, the CFD calculation results can in principle only be evaluated and understood from a safety point of view with much more effort. Even the inspection authorities must have sufficient expertise. For the industrial application of CFD programs in the field of safety technology, the exchange between learners and experts, and training specifically with experts from both safety engineering and CFD, are necessary. At the symposium in Tutzing, a “CFD's license” was proposed. The ensuing discussion revealed the following applications for CFD calculations in safety engineering:

1. To gain additional in-depth knowledge and understanding of physical processes. This is especially true if the effect of individual parameters to be investigated or detailed information about the spatial and temporal distribution of individual parameters is required.

2. To visualize process operational work flows.

3. To use as the sole source of information in areas where no experiments are possible (hazardous materials, very high pressures).

4. To examine boundary conditions as specified for conventional models.

5. To interpolate experimental results.

6. To question conventional methods and standards. This includes the improvement of these methods and the reduction of safety margins due to higher accuracy of the models.

CFD programs are already used in the field of safety technology for the optimization of valve operations, the investigation of fires and explosions, the examination of single-phase fluid flows, the propagation of liquid pools from leaks, and generally for the investigation of incidents. In contrast, there are also some areas where the CFD computer codes should not be used, namely:

when simple models are adequate,if they are the only source of information to design safety devices,for unknown or fluctuating initial and boundary conditions,for extrapolations beyond a range with experimentally validated data,if only insufficient property data are available, orto solve very complex problems with many parameters.

Typically, established and conventional methods are applied to design safety measures and to size safety devices. With increasing risk, these standards are more important. For most of the safety experts it is currently not viable to size safety devices solely on the basis of CFD simulations. It is however expected that this will change in the future.

According to the information of the participants, 48% of the participants of the Tutzing symposium in 2011 with a safety-related background and 68% of the numerically trained participants trust in CFD simulations applied for safety engineering tasks. Training and experience, experimental validation of models, and the definition of standards (Best Practice) are the relevant criteria in order to further increase confidence.

In summary, the discussion in the Tutzing symposium has shown that CFD computer codes are used in safety technology with different intensity according to specific tasks. CFD has arrived in safety technology! The advantages of these tools show up in all areas of technology. But the dangers in the application of safety technology have also been impressively demonstrated, for example:

1. The extrapolation of validated results from highly non-linear CFD models can lead to extreme errors.

2. In safety technology, initial and boundary value problems often cannot be defined with the necessary accuracy. This may result in large errors or large uncertainties in the results of a CFD simulation. In many cases these uncertainties cannot be quantified.

3. The most often used eddy viscosity turbulence models dampen smaller fluctuations and in principle do not allow for the adequate resolution of a problem in some cases. In contrast, Large Eddy Simulation (LES) or Direct Numerical Simulation (DNS) are typically more precise but increase the computation time enormously.

BASF SE, Ludwigshafen

Only with sufficient training and experience in dealing with the CFD models and their solution methods can questions in the field of safety technology be answered responsibly. Any ‘black box’ CFD application mentality in which results are firstly obtained by systematic variation of models and adaptation of internal model parameters to very few experimental data and secondly presented as validated results and subsequently used for extrapolations must be strictly avoided. Of course, this applies to any type of modeling in safety technology. An extended study program at German and international universities is needed to inculcate the necessary safety skills and mindset in the next generation of students. At the same time, interdisciplinary numerical, experimental, and safety skills must be taught – just a new kind of computational Safety Engineering. For practical application in industry the idea of a CFD license or quality labels should be pursued.

The CFD computer codes should be supplemented as a standard tool by best practice guidelines in the field of safety technology and by many test cases from the professional safety community. The research and development work on the way to such standard tools (and common sense) can only enhance the training of safety engineers in the field of CFD, the acceptability of the methods, and ultimately the current state of safety technology.

With CFD tools, the demand for necessary safety measures and economic operation of plants can be merged. The knowledge so gained is considerable, and the trend toward making increasing use of these tools is already equally considerable. As long as the results are physically based on a meaningful theory and are responsibly weighted by safety considerations, this is the right way into the future of safety technology.

The 50th Tutzing Symposium 2011, organized by the community of safety technology of the Dechemás ProcessNet initiative, has brought experts from the fields of safety engineering and numerical modeling together for a first major exchange of views. Only when these two disciplines grow closer together will CFD be able to establish itself as a standard tool in all areas of safety technology.

Members of the safety community in Germany who participated in the 50th Tutzing Symposium in 2011.

Chapter 2

Organized by ProcessNet: Tutzing Symposion 2011 ‘CFD – its Future in Safety Technology’

Norbert Pfeil

This contribution to the present book is intended to act as a bridge between the continuous work of ProcessNet's Safety Engineering Section and the 50th Tutzing Symposion entitled ‘CFD – its Future in Safety Technology?’ held in May 2010 at the Evangelische Akademie Tutzing at Lake Starnberg, Bavaria. It may also hopefully make both ProcessNet and its Safety Engineering Section better known, particularly in the international process and plant safety community.

2.1 ProcessNet – an Initiative of DECHEMA and VDI-GVC

In December 2006 DECHEMA (the German society for chemical engineering and biotechnology) and VDI-GVC (the German society of engineers and society for chemical and process engineering) united all their chemical and process engineering activities under a common umbrella: ProcessNet. Nearly a hundred committees of both societies with a wide variety of scientific and technical tasks of various kinds organized their work within the eight ProcessNet sections

Chemical Reaction Technology

Fluid Dynamics and Separation

Particle Technology and Product Design

Process, Apparatus, and Plant Technology

Safety Engineering

SuPER (Sustainable Production, Energy and Resources)

Materials, Construction, Lifetime

Education and Innovation.

Since then, ProcessNet has acted successfully as the one German platform on chemical engineering with more than 5000 members from the sciences, industry, and administration, exchanging ideas and experiences within ProcessNet's committees and at the numerous events organized year by year. Papers dealing with strategic topics have been published to support ongoing scientific and societal discussions and also to trigger funding policy initiatives. Further details on ProcessNet are available on its website: www.procesnet.org.

2.1.1 The ProcessNet Safety Engineering Section

Safety engineering was first treated as a distinct topic within DECHEMA and GVC in 1978, by establishing a joint Research Committee. In the inaugural meeting of this Committee on 5 July 1978, its tasks were characterized as

development of safety concepts for chemical engineeringpublic relations by conferences, publications, and discussioninitiation and coordination of working items including research work

aiming to tackle the increasingly demanding safety requirements in chemical plants by taking on direct responsibility for these matters instead of reacting to requirements imposed from outside.

The first two working parties ‘Safe Designing’ and ‘Risk, Damage Analysis, and Reliability’ were established in 1978 as a first step toward today's structure of the section with working parties covering all relevant areas of safety engineering, as follows:

Safety ParametersChemical Process SafetySafe Design of Chemical PlantsPreventive Industrial Fire SafetySafeguarding of Industrial Process Plants by Means of Process Control EngineeringReleases and Impacts of Hazardous MaterialsElectrostaticsLessons from Process Safety IncidentsRisk Management

The working parties listed above, with about 130 appointed members all in all, observe recent scientific, technical and legal developments and identify topics which need further consideration, for instance, by research, by more intensive exchange of ideas, or by making specific knowledge available to everybody responsible in the field of process and plant safety.

However, the working parties are only one part of the Safety Engineering Section. Another no less important part is the group of people who have subscribed to the Section, currently including 550 members, from industry, science, and administration.

A steering board of elected members plus the chairpersons of the working parties or temporary working groups takes care of both the safety engineering community – for instance by initiating events like the Tutzing Symposion 2011 – and the strategic development of the working structure of the Section. Aims and tasks of the currently existing working parties/groups are to be found on the Section's website, together with much relevant information for and from the community concerning institutions, events, publications, training courses and so on – most of it in the German language.

2.2 A Long Discussed Question: Can Safety Engineers Rely on Numerical Methods?

Safety engineers are conservative, for sure. Their recommendations need to err on the safe side. Therefore, they tend to suspect the results of numerical simulations of insufficiently known accuracy carried out with complex software the performance of which they do not understand completely. This subject has been been discussed repeatedly within the Safety Engineering Section and its working parties.

However, methods like CFD (Computational Fluid Dynamics) are successfully used in the chemical industry for optimizing process, apparatus, and plant design. Therefore, it is not surprising that several working parties regard numerical modeling or CFD as current working items. The question which has to be answered today seems no longer to be whether or not CFD could be used in process and plant safety, but how it is to be used most appropriately. And when the steering board of the Safety Engineering Section was required once again to organize DECHEMA's traditional Tutzing Symposium and to decide what the general topic would be, the unanimous vote was CFD and safety engineering. Tutzing Symposia are very popular because Tutzing Castle is a perfect venue for about 100 experts to gather together over a couple of days for a meeting which is more a conclave than a normal conference. It was very much hoped that the special atmosphere in the Evangelische Akademie Tutzing would lead to substantial answers to the question ‘CFD – its Future in Safety Technology?’. The contributions in this book will show that this hope was well justified.

Chapter 3

CFD and Holistic Methods for Explosive Safety and Risk Analysis

Arno Klomfass and Klaus Thoma

3.1 Introduction

After more than 30 years of intense development Computational Fluid Dynamics (CFD) has become an indispensable tool in many scientific and engineering disciplines. Early CFD development was strongly promoted by the aerospace and defense industries. In the 1980s, numerical solutions to the three-dimensional Euler equations of inviscid flow became possible, and in the 1990s the full Navier–Stokes equations became solvable for an entire aircraft [1]. While CFD in the aeronautics community was mainly concerned with external flows of gases and combustion processes in rocket motors and jet engines, the defense industry developed so-called hydrocodes or wave propagation codes, which enabled the simulation of high speed processes in compressible fluids for the analysis of explosion and penetration processes [2]. The characteristic features of these codes are the explicit time integration and the capability to treat multiple materials within the same computational grid and to model materials with strength, that is, solids. The first computational methods for the simulation of large-scale gas explosions for the oil and gas industry also appeared in the mid-1980s [3, 4]. Since the early days, CFD has developed into a versatile tool for many industrial branches. Today, an evolution into multi-physics simulation can be recognized, as modern software concepts permit the coupling of simulation codes from different disciplines: fluid–structure coupling is the most prominent example [5, 6]. This successful evolution was enabled through three factors: the development of advanced software architectures, the invention of versatile numerical discretization and integration techniques, and the incredible and ongoing development of computer performance. While a majority of early CFD codes only worked on Cartesian grids, most modern codes are based on flexible unstructured grid concepts, which enable the usage of body-fitted grids even for complex geometries. Further to this, mesh-free methods for particular applications such as flows through complex geometries with changing topologies have been developed. Such methods are used, for example, in the simulation of airbag inflation in automotive crash simulation [7].