<p>In this thesis, gas explosions inside pipes are considered. Laboratory experiments and numerical simulations are the basis of the thesis. The target of the work was to develop numerical models that could predict accidental gas explosions inside pipes.</p><p>Experiments were performed in circular steel pipes, with an inner diameter of 22.3 mm, and a plexiglass pipe, with an inner diameter of 40 mm. Propane, acetylene and hydrogen at various equivalence ratios in air were used. Pressure was recorded by Kistler pressure transducers and flame propagation was captured by photodiodes, a SLR camera and a high-speed camera. The experiments showed that acoustic oscillations would occur in the pipes, and that the frequencies of these oscillations are determined by the pipe length. Several inversions of the flame front can occur during the flame propagation in a pipe. These inversions are appearing due to quenching of the flame front at the pipe wall and due to interactions of the flame front with the longitudinal pressure waves in the pipe. Transition to detonation was achieved in acetylene-air mixtures in a 5 m steel pipe with 4 small obstructions.</p><p>Simulations of the flame propagation in smooth pipes were performed with an 1D MATLAB version of the Random Choice Method (RCMLAB). Methods for estimation of quasi 1D burning velocities and of pipe outlet conditions from experimental pressure data were implemented into this code. The simulated pressure waves and flame propagation were compared to the experimental results and there are good agreements between the results.</p><p>Simulations were also performed with the commercial CFD code FLACS. They indicated that to properly handle the longitudinal pressure oscillations in pipes, at least 7 grid cells in each direction of the pipe cross-section and a Courant number of maximum 1 should be used. It was shown that the current combustion model in FLACS gave too high flame speeds initially for gas explosions in a pipe with an inner width of 40 mm.</p>

In this thesis, gas explosions inside pipes are considered. Laboratory experiments and numerical simulations are the basis of the thesis. The target of the work was to develop numerical models that could predict accidental gas explosions inside pipes. Experiments were performed in circular steel pipes, with an inner diameter of 22.3 mm, and a plexiglass pipe, with an inner diameter of 40 mm. Propane, acetylene and hydrogen at various equivalence ratios in air were used. Pressure was recorded by Kistler pressure transducers and flame propagation was captured by photodiodes, a SLR camera and a high-speed camera. The experiments showed that acoustic oscillations would occur in the pipes, and that the frequencies of these oscillations are determined by the pipe length. Several inversions of the flame front can occur during the flame propagation in a pipe. These inversions are appearing due to quenching of the flame front at the pipe wall and due to interactions of the flame front with the longitudinal pressure waves in the pipe. Transition to detonation was achieved in acetylene-air mixtures in a 5 m steel pipe with 4 small obstructions. Simulations of the flame propagation in smooth pipes were performed with an 1D MATLAB version of the Random Choice Method (RCMLAB). Methods for estimation of quasi 1D burning velocities and of pipe outlet conditions from experimental pressure data were implemented into this code. The simulated pressure waves and flame propagation were compared to the experimental results and there are good agreements between the results. Simulations were also performed with the commercial CFD code FLACS. They indicated that to properly handle the longitudinal pressure oscillations in pipes, at least 7 grid cells in each direction of the pipe cross-section and a Courant number of maximum 1 should be used. It was shown that the current combustion model in FLACS gave too high flame speeds initially for gas explosions in a pipe with an inner width of 40 mm.

Introduction. The police and the investigation team has been manually doing behavioural analysis and connecting different crimes to an offender. With the help of computers technologies, databases, and automated system, the statistical analysis of the offender’s behaviour significantly improved. There we can transfer from a manual process to an automated one, and the investigator can allocate time and resources better by prioritising the offenders to investigate. In this study, we create and experiment with a proof of concept system that ranks and prioritise different offenders using the Random Choice method in combination with the state of the art Spatial-Temporal method. Objectives. In experimenting with the proof of concept system, we are aiming to understand the effect of different offender’s behaviour having on the offenders ranking and the effect of having multiple different numbers of reference crimes in the database. The objective is also to understand the role of consistency and distinctiveness in offenders ranking. Moreover, understanding the performances of our proof of concept system comparing to already existing methods such as Random Choice, Spatial-Temporal and a baseline method that based on pure randomness. Method. The method we chose for this study was an experimental study. With an experimental environment with independent and dependent variables, we presented and evaluated the system. We were using the experimenting approach because it has a stable presence and widely used in similar studies in this field. Results. After the experiments, we found that different Modus Operandi (MO)categories have a different effect on the ranking results and different distinctive combinations of MO categories also has different accuracy when ranking the offenders. Offenders that were consistent with more references crime in the database were often higher ranked and were linked more correctly. Our proof of concept system shows significant improvement over Random Choice method and the Spatial-Temporalmethod. Conclusion. From the results, we concluded that the proof of concept system displays a significant accuracy in ranking and prioritising offenders, there different MO categories and combinations of them has a different effect on the accuracy of the ranking. The ranking system was also affected by the number of reference cases that exist in the database. Future works can extend the study by trying to improve different aspects of the proof of concept systems, such as the Random Choice aspect or the Spatial-Temporal aspect.

In this study, I introduce the alternative-dependent focal Luce model (ADFLM), a random choice model generalizing the well-known Luce model (1959). In the ADFLM, focal alternatives are chosen more frequently relative to their utilities. I identify utilities, focal sets, and the magnitude of focal biases from choice data. Additionally, I axiomatically characterize the ADFLM by weakening the independence of irrelevant alternatives (IIA) axiom. This model can explain the well-known behavioral phenomena, the attraction and compromise effects. Furthermore, I also study the seller's profit maximization problem in the ADFLM. I also study an asymmetric dynamic patent race with a deadline under complete information. In my model, two firms decide whether to invest in RandD. The patent arrives randomly according to a Poisson process, and the large firm has a higher hazard rate than the small firm. I find the unique sub-game perfect Nash equilibrium strategy for this game. At the equilibrium, the large firm will stay longer in the race, while the small firm will quit earlier. The large firm's optimal stopping time is not affected by the competition, while the small firm's stopping time is reduced. Additionally, I find that companies will remain longer in the race if the investigation cost is lower, the winning premium is higher, the deadline is extended further, and the hazard rate is more prominent. Moreover, the market becomes more efficient with the competition since the patent is easier to realize. / Doctor of Philosophy / In this research, I study the consumer's behavior when individuals have limited cannot or do not give the same attention to each alternative available to them. In my study, I characterize the alternative-dependent focal Luce model (ADFLM), a consumer behavior model. Moreover, I solve the seller's profit maximization problem when the consumer's behavior follows the ADFLM. Meanwhile, I also study a dynamic patent race problem that occurs when firms compete for a patent with a deadline. If no firms achieve the patent, the stopping time (when the firm quits the patent race) of the large firm's (with a higher success rate every period) is not affected by the introduction of the small firm. However, the small firm quits earlier when the large firm is introduced. The competition between the two companies increases the overall probability of receiving a patent.