This paper proves that the hierarchy neural network application is effective in simplifying the distinction process and accommodating dispersion of the gas sensitivity by each sensor and the deterioration of gas sensitivity by the aged degradation. I first transform the gas sensor's heater voltage to sawtooth waveforms and replace the sensor's output response for ethanol and methane mixed gases and ethanol and propane mixed gases with a quadratic experimental equation using regression analysis. I show the gas kinds distinction results obtained from three types of input elements—the regression coefficients, difference voltage, and output voltage. I then show how to obtain a reliable distinction result by using jointly distinction results obtained from the three different input elements.
Experiments using a particle-track laser sheet technique (PTLS) and a laser sheet tomography (LST) combined with a high speed video camera are conducted to reveal the flame structure and the flow vector diagram in pool fires. To improve our understanding of the scale effects on buoyancy controlled flame structure, we employed four different size of kerosene pool fire whose pan diameters are 10, 20, 30 and 50 cm. These results were compared with the numerical analysis previously presented by other researchers. The flow visualization results at the vertical plane along a center axis are in good agreement with the numerical results. However, air entrainment through the flame is estimated to be at most 20% larger than that of axisymmetric smooth flame because of the increase of the reaction surface area and the existence of the rotational velocity component around the wrinkled flame. For a medium-size pool fire below a 20 cm-diameter pan, the flame height and the puffing frequency are close to an assembled small-size pool fires in which 3 cm-diameter pans are lined around circle. The flame structure below a 20 cm-diameter pool fire seems to be composed of several flame cells. While the flame structure above a 30 cm-diameter pan is quite different from the aggregate of small-size pool fires. For a 50 cm-diameter pool fire wrinkled flame having high frequent disturbance of 12-14 Hz occurs near the base flame zone and grows along the flame surface. A pool fire beyond a 50 cm-diameter pan transits to a large-scale pool fire having turbulent flow structure.
In order to ensure the fire resistance efficiency of framework members, which were made using high strength concrete, loaded fire resistance tests were conducted using high strength RC columns containing PP (polypropylene) fibers in which the specified design strength (Fc) was increased from 60N/mm2 up to 150N/mm2. Our findings were as follows: 1) There was no explosive spalling at Fc60 regardless of whether or not PP fibers were added. High strength RC columns of Fc60 provided 3hours of fire resistive period capacity. 2) At Fc80 to Fc150, explosive spalling occurred when there was no PP fibers mixed in the concrete. As Fc increased, the scale of explosive spalling increased and fire resistive period capacity decreased. 3) The addition of PP fibers had the effect of suppressing explosive spalling. The amount of PP fibers needed to maintain 3hours of fire resistive period capacity is estimated to be 0. 055vol. -% or greater for Fc80; 0. 10vol. -% or greater for Fc100; and 0. 33vol. -% or greater for Fc130 and Fc150.