Search

In the past decades, water droplet-based suppression systems (i.e. fire sprinklers, water mist) have been extensively utilized as building fire suppression systems. Nevertheless, current suppression systems operate under considerable design limitations due to high-rise structures and rapid increase in building complexity, such as pressure supply and water storages. Challenges can also be foreseen in suppressing water-reactive chemicals (i.e. alkali metals, hydrides) as the violent explosive reaction will be triggered. Therefore, it is vital to investigate potential suppression agents to cope with the increasing building fire risk associated with complex building materials and hazardous combustibles.A large eddy simulation (LES) model incorporated with novel user-defined functions (UDFs) to consider particle expansion and charring was proposed in this thesis. This model has been utilised to numerically study an experimental case to investigate the fire suppression behaviour of expandable graphite (EG) in building structures. This approach has provided an in-depth characterization of EG's thermophysical properties. This includes the expansion, barrier effect from char formation and the decomposition of EG. It was discovered that EG is more effective in fire suppression compared to natural graphite. Among the diameter range of EG (400 µm - 1000 µm), the smaller diameter of EGs tend to be efficient in suppressing metallic fire. The WALE SGS model has provided the most accurate temperature prediction among other SGS models, with average relative errors of 7.71% and 8.93%. The novel multiphase model was comprehensively validated by material testing and other experiments, and proven to be an effective tool to investigate particle-infused suppression in a structural fire. Moreover, the thermophysical properties (i.e. pyrolysis kinetics) of graphite have also been characterized through Molecular Dynamics (MD) simulations. The extracted Arrhenius kinetics parameters (i.e. act

Intumescent coating reinforce mica was developed and studied in terms of their hysical appearance, thermal properties, composition, and morphology. 6 samples, including control formulation, were developed by varying percentage of mica to binder from 0 to 5%. Control fire test shows an increase of expansion when the percentage of mica increases but decreases when more than 4% of mica was used. The thermal stability of the samples with 3% mica was improved with a degradation rate of 0.047 wt.%/min compared to the control formulation 0.091wt%/min suggested coming from silicon phosphate oxide composition emerged on XRD analysis. Silicon phosphate oxide was known to enhance the antioxidation of the char and thermal stability of the char. A homogenous surface with a smaller pore was also observed on the char structure of MC3\ coming from the ceramization process, which creates a ceramics layer on the surface, thus improve the insulation properties of the coating.

A comprehensive literature review on the development of 3D inkjet printing technologies has revealed that inkjet printing is an effective method for additive manufacturing with advantages in specific applications such as microstructures and printable electronics. However, the knowledge of the flow development of the inkjet from the flow formation inside the nozzle to the dropping process is in severe dearth, so are the predictive models for the printing process. A CFD (computational fluid dynamics) model is presented to study the jetting and dropping characteristics for needle-valve inkjets. The model is developed by using a dynamic mesh method and experimentally verified for both a low viscosity fluid typified by a distilled water, and a relatively high viscosity PEDOT:PSS polymer nano-particle containing ink. The model-calculations show a good agreement with the experimental data under the corresponding conditions for the both types of inks. The results from the model show that the parameter window for generating a single droplet from the low viscosity fluid is narrower than that using the high viscosity ink. It also reveals that the droplet size increases with the nozzle diameter and a droplet with a diameter slightly smaller than the nozzle diameter can be achieved with a low viscosity ink under a low dropping velocity. The verified CFD model is then used to study the flow and dropping characteristics of an ink containing organic conductive polymer (PEDOT:PSS) nano-particles. It is revealed that three types of droplet defects can occur in the needle-valve ink jetting process, e.g. excess liquid suction, liquid accumulation, and satellite droplet formation. It is further found that an increase in the valve striking time, nozzle diameter or valve seat inclination results in a dramatic decrease in the droplet velocity, while an increase in the valve stroke or inlet air pressure causes an increase in the droplet velocity. The droplet volume and equilibrium depositio

Bushfire is a devastating tragedy occurring in forest or wildland area causing hardly recoverable environmental and ecological disruptions, as well as the billions worth of property damages. In Australia, the majority of landscape comprises of grasslands area and are frequently ravaged by bushfires due to the hot and dry climate. Strong knowledge of the bushfire behaviour is essential for prevention, planning, response and recovery. A novel fire field model has been proposed for bushfires modelling under the Large Eddy simulation (LES) framework incorporating detailed chemical kinetics combustion modelling, in associated with soot formation, turbulence and radiation models which are fully coupled and interactive. Two fire spread model has been developed and tested: (i) a level set model and (ii) pyrolysis model. Numerical simulations of the fire spread on a pine needle board with adjustable slope angles have been performed. The fire spread rate and temperature predictions were in good agreement with experimental data. The model was capable of capturing the increase in fire spread rate for larger slopes. Furthermore, the flame tilting behaviour caused by the unbalanced air entrainment was also successfully replicated. The tilting of the flame front caused the fire to lean towards the unburned region of the pine needle board, which increased the radiative heat flux from the flame to the solid fuel surface leading to an increase in fire spread rate. This physical phenomenon was fully captured by applying the pyrolysis model which has been found to more accurately predicting the fire spread rate compared to the level set model for steeper slope angles. With the consideration fluid mechanics and heat transfer for the turbulent reacting gas mixture, solid-phase decomposition, detailed chemical kinetics for gas-phase combustion with comprehensive description of the generation processes for asphyxiant gases including CO, CO2 and soot particles, as well as the radiation heat

Combustion-generated soot particulates is a subject of great interest due to its past and ongoing applications in research and industrial sectors. An in-house Direct Quadrature Method of Moments (DQMOM) based population balance model has been developed to investigate the evolution of soot particulate. Most commercial packages assume the particle size is monodispersed, while the proposed model enables the evaluation of real-time soot particle size distribution which further enhances prediction accuracies. The proposed model is fully coupled with all essential fire sub-modelling components and integrated with the strained laminar flamelet model considering detailed chemical kinetics. Numerical simulation with the implementation of the proposed DQMOM soot model has been validated against an in-house co-flow burner experimental study and compared with other numerical studies. The results demonstrate that the improved DQMOM soot model has a significant improvement to the accuracy of simulation when compared with Moss-Brookes soot model. It has also discovered an optimised combination of numerical configurations, e.g., nucleation and oxidation kinetics, as well as fractal dimension and Schmidt number, that yield the most agreeing results against experimental data.The validated DQMOM model is designed to practically model realistic fire scenarios, e.g., fire whirl, a catastrophic fire phenomenon driven by eddy-generation associated with radiation feedback. Owing to the intriguing coupling of the flow dynamic and combustion kinetic, pilot investigation of fire whirl formulated under various entrainment conditions has been carried out. The test was conducted in a numerical environment decoupling all interacting, to characterise the flame temperature, flame height, velocity in axial, radial and tangential direction over a 50 s timeframe. The result demonstrates the significant enhancement of the combustion process, i.e., flame temperature and flame height, due to the intensif