Projects / Programmes
Fire safe accomodation of highly combustible materials in steel-framed structures: Development of models and experimental verification
| Code |
Science |
Field |
Subfield |
| 7.00.00 |
Interdisciplinary research |
|
|
| Code |
Science |
Field |
| T220 |
Technological sciences |
Civil engineering, hydraulic engineering, offshore technology, soil mechanics |
| Code |
Science |
Field |
| 2.01 |
Engineering and Technology |
Civil engineering |
fire, modelling, thermal degradation of material, pyrolysis, TGA, DSC, cone calorimetry, steel structures, creep of steel, hardening of steel
Researchers (1)
| no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
| 1. |
33293 |
PhD Jerneja Češarek Kolšek |
Civil engineering |
Head |
2016 - 2018 |
0 |
Organisations (1)
Abstract
The proposed postdoctoral project deals with the issue of fire safety in modern production-storage buildings. Nowadays, in terms of saving time and costs during their construction, these buildings are built often with prefabricated, mostly steel-framed structures which exhibit high sensitivity at elevated temperatures. On the other hand, higher and higher amounts of easily-combustible materials are accommodated in these buildings what increases their fire vulnerability additionally. As experience has shown, the traditional approaches to design of fire safety of buildings often fail to deal with the later satisfactorily, thus, development of corresponding new, advanced mathematical procedures is necessary. A suggestion of such will be prepared in the proposed project.
The mathematical models, developed for this purpose, will consist of two phases. The first phase will be dedicated to advanced CFD ('Computational Fluid dynamics') simulations of fire spread and the corresponding model for these will be prepared using the software FDS ('Fire Dynamics Simulator'). Within the latter, material models with suitable kinetic and thermal parameters of the selected materials (these will represent the fire load of the building) will be developed and properly embedded. These material models will be derived by numerical fitting of the Arrhenius models of material thermal degradation and the Fourier models of heat conduction to results of material thermoanalytical and fire experiments, i.e. TGA ('thermogravimetric analysis'), DSC ('differential scanning calorimatery'), bomb and cone calorimetry. Furthermore, the second phase of the models will deal with calculations of mechanical response of the building’s steel-framed structure due to fire predicted in the preceding phase. For this purpose, a suitable model will be prepared in ABAQUS FEA engaging 3D solid finite elements for representations of the entire structural assembly. In addition, special material models of high-temperature response of steel also accounting for new proposals for descriptions of high-temperature creep and isotropic-kinematic hardening of the material, the latter being caused by temperature-induced deformation reversals, will be coded in Fortran and embedded in Abaqus using the user subroutine UMAT. Finally, suitable stabilization algorithms will be checked for calibration of the Abaqus model and elimination of convergence issues.
In the second part of the project the prepared models for simulations of fire spread and for simulations of structural fire response will be suitably validated against experimental data.
In the last part of the project the models will be implemented in fire studies of a selected realistic example of a production-storage facility. The results will be compared against solutions of the traditional design procedures. Thereby, the advantages of the advanced procedures will be finally exposed along with the threats of ignoring the limitations of the present simplified approaches.
Significance for science
MAIN RESULT RELEVANT FOR SCIENCE AND PROFESSION: For future fire engineering practice to be properly upgraded and grown to challenges of higher fire vulnerability of modern buildings, special publications will be needed in which solutions of performance-based fire safety design will be presented on realistic practical examples and placed side by side with traditional solutions. Precisely such comparisons are also the ultimate result of the project and were presented in publications [1]-[5]. The conclusions presented here confirmed the initial hypothesis of the project, that in some cases simplified procedures (which are nowadays used almost exclusively in engineering practice because of the lack of knowledge or lack of understanding of their limitations) are on the unsafe side. In this regard, the author published the recommendations for a better definition of the limits of the applicability of simplified procedures in foreign [1-3] and Slovene literature [4-5] as well as at international and domestic events [9-10]. In particular in the light of recent tragic events (e.g. the Grenfell tower fire in London in June 2017), the significance of these results for science and the profession can be assessed as very high. OTHER RESULTS CONTRIBUTING TO SCIENCE: • Modelling fire spread: In the project new material submodels (i.e. pyrolysis and combustion models) were developed for selected materials/products, i.e. extruded polystyrene (XPS)[1,4], polyurethane foam (PUR) insulation [6], car rubber [7] and spruce-wood cribs [8]. The models are suitable for embedment in software FDS, which is one of the modern most widely used numerical tools for advanced fire analyses. The development of such sub-models will, thus, be useful to other researchers and engineers using FDS. The following questions were answered at the same time: - What quantity and distribution of thermal conditions of TGA/DSC tests and cone calorimetry must be tested for achieving certain accuracy of material submodels and of the final models of fire spread? - In determining the kinetic parameters of material submodels, is it necessary to use more precize calculations with genetic algorithms or are simplified analytical methods sufficient? - It has also been proven that two apparently equivalent samples of the same-type material can show a completely different burning behaviour. For this reason, it is clear that in the development of fire models (e.g. for simulations of past fires), only general information on the type of material is not sufficient but more detailed material descriptions are needed. How wrong can our solutions be otherwise, was in the project shown in connection to the tested XPS. In 2013 this material caused a fire in a factory of company FIBRAN when static electricity ignited one of the freshly produced XPS panels on the production line. Since fresh plates (in contrast to the older ones) emit large quantities of a highly flammable pentane (this is used in production of the plates), it was clear that this fire would have occurred unlikely if the plates were old (not fresh). The laboratory tests of fresh and old XPS samples also showed that using the results of old samples in material sub-model development would lead to completely false results of the simulation of this fire. The summary of the three issues presented above have been described in references [1] and [4] and will be presented in a more detail in a paper aimed for publication in a Category 1 SCI journal (in preparation). • Modelling structural response: New material sub-models were also developed for embedment in simulations of fire response of larger steel structural systems (software Abaqus), i.e.: - new material sub-models for description of 3D stress-strain states in steel at elevated temperatures considering isotropic-kinematic hardening and creep of steel [3,5] and - new (modified) models for calculations of steel temperatures behind an intumescent fire-protective coating [2].
Significance for the country
• DEVELOPMENT OF COMPANIES: By raising awareness of the limitations of current simplified fire design procedures through domestic professional and scientific publications and events [4-7], the project is particularly important for the progress of Slovenian fire engineering companies. The project is also important in terms of the progress of the development of material submodels for the simulation of the development of the fire. It is expected that this kind of modelling will be in the future used with benefit also in development of new materials and products. With its help and the help of softwares such as FDS it will be, namely, possible to predict fire risks of these materials/products on a larger scale before testing the in large-scale settings. This will speed up the process of material development and reduce its costs thereby contribute significantly to higher competitiveness of the material developers. • EDUCATION OF SLOVENIAN ENGINEERS: Beside in papers in international scientific journals and in conferences the results of the project were and will be also shown in national professional journals and at national events [4-7]. Thus, the project also contributed significantly to education of Slovenian engineers. While Slovenia at date remains one of the countries where a specific engineering (undergraduate) study programme of fire science is not yet established, the research and knowledge in the field of advanced fire science is limited to specialised postgraduate research dispersed among faculties from different scientific fields (structural engineering, chemistry, mechanical engineering, architecture...) and research institutes. The knowledge provided by these institutions and further published in national popular-science journals is, thus, of crucial importance. In order to increase the knowledge and awarness of future Slovenian engineers in Slovenia, the project leader also contributed 36 hours of special lectures, which she carried out in the winter semester of the school year 2017/2018 as part of the course IPMK (Selected Chapters in Solid Structures) of the Master's Degree Program of Civil Engineering at Faculty of Civil and Geodetic Engineering in Ljubljana. An important part of these lectures was devoted to presenting advances in modern fire design approaches and presenting the results of the FIRESIM project. • PROMOTION OF SLOVENIA: On the project a new test equipment of ZAG, i.e. cone calorimeter with FTIR analysers representing the key equipment of fire science and the first case of such equipment in Slovenia, was used and explored in details. With the gained knowledge, Slovenia will be able to finally take a competitive position in a wider international context. • ACCESS TO NEW KNOWLEDGE: During the project, the project leader also actively participated in COST Action FP1404: Fire safe use of bio-based building products, which was a source of establishing important research connection for the needs of the proposed project as well as for the need of possible future collaborations. • ENVIRONMENTAL PROTECTION: Fires in production-storage facilities and other industrial facilities can be particularly damaging to the environment as evidenced, e.g., also in the recent (May 2017) fire in Kemis company in Vrhnika. Due to burning of large amounts of stored materials in such fires large quantities of hazardous gases can be released into the air during a short time period or spills of dangerous liquids can be caused due to fire-induced damage on production equipment. The proposed project aimed at development of advanced numerical models for fire-spread simulations whereby with using suitable postprocessing analyses such kinds of undesirable fire consequences could also be foreseen along with their measures for their prevention and management. In this way, the project is also important from the perspective of environmental protection.
Most important scientific results
Interim report,
final report
Most important socioeconomically and culturally relevant results
Interim report,
final report
