Heat Release Rate of Fires Using Point-Based Sampling

Ho, Hsin-Hsiu

Etd

Heat Release Rate of Fires Using Point-Based Sampling

Public Deposited

Conventional hood-based calorimeters rely on total capture of the combustion products to measure HRR. Because of this requirement, the fire size is often limited to a certain length scale. A new Outdoor Gas Emission Sampling (OGES) system was developed to serve as an alternative to expensive industrial gas sampling equipment to measure the heat release rate (HRR) of large-scale fires in outdoor scenarios. The low cost and lightweight nature of the OGES design introduced the potential of deploying multiple systems simultaneously and a new technique utilizing point-based capture. Before performing experiments, FDS simulations at three different length scales demonstrated the viability of point-based sampling for fire calorimetry. When the fire centerline is known, only a single sampling point is needed at the centerline to characterize the flow and calculate HRR. When the location of the centerline is unknown, another method using least-squares regression was created to solve for the sampling distance to the fire centerline. In this situation, results showed that a minimum of three points are needed to characterize the flow and obtain HRR. Furthermore, a rule of thumb was determined for suitable sampling location: (1) the first sampling point (r1) should be between the centerline and the half-radius of the anticipated visible smoke plume (r1 ≤ 0.50σ), and (2) sampling point spacing (x) should be at least one-third but no more than the half-radius of the anticipated visible smoke plume (0.33σ ≤ x ≤ 0.50σ). These methods were then validated with experiments at three different length scales. The first set of experiments consisted of meso-scale (75 cm diameter) pan fire experiments aimed at validating OGES’ accuracy when measuring a pool fire compared to industrial gas sampling equipment (SERVOMEX 4200). Results showed comparable readings between the two. CO2 and CO concentrations versus height showed the suitable sampling height for OGES to be within the intermittent zone and lower plume zone of the fire. Moreover, this stage of the study did not have measurements to establish ṁ and relied on existing plume theory correlations, which were applicable because the sampling location was at the fire centerline. Results showed good agreement for HRR, but overestimations using the CDG method could be attributed to the uniform, or top-hat, profile assumption in plume theory correlations to estimate ṁ. Therefore, it was understood that multiple sampling points and additional measurements were needed to achieve an improved understanding of ṁ. For the second set of experiments, the conclusions from the meso-scale work regarding suitable sampling height were applied to large-scale (1.9 m by 1.7 m) outdoor oil-on-water fires. Because of certain limitations, only a single sampling point along the centerline was used for analysis. Although limited, OGES demonstrated potential as a low-cost but effective tool for HRR measurement, especially in situations where total capture of the plume is unfeasible and only point sampling is possible. However, similar to the meso-scale experiments, many of the discrepancies in HRR between gas analysis methods and the fuel regression rate data could be attributed to the use of existing plume theory correlations to estimate ṁ. To obtain ṁ through measurements, temperature and vertical velocity measurements were added to the OGES sampling apparatus. Velocity measurements were achieved by using bi-directional probes, differential pressure transducers, and a Raspberry Pi mounted on the apparatus. In the last set of experiments, by utilizing gas concentration, velocity, and temperature data in addition to video footage, results from two large-scale (3.6 m by 1.8 m) burn experiments agreed with the FDS conclusions that a minimum of three discrete sampling points along the radial direction of the smoke plume are needed to accurately determine the position of the sampling points relative to the centerline of the fire. This point-based sampling technique was most noteworthy in its ability to perform HRR measurement even when the burn parameters deviated from the scope of existing correlations. This was evident when comparing well-established flame height correlations during burn enhancement technology deployment. Partial premixing of the fuel and air by the apparatus is believed to have culminated in higher HRR values as shown by fuel regression rates and gas concentration data. Through the use of OGES, this study has exhibited the first instance of removing the total capture requirement that has been present in fire calorimetry for over 50 years and the full implementation of the point-based capture concept for HRR measurement.

Creator