We try to construct an evaluation method on evacuation safety based on the probabilistic theory. At first, we try to specify the probability distribution functions of important random variables like as walking speed, walking time, and then, it is shown that the logarithmic normal distribution is desirable to adopt as these distributions. Next, we get the probability of evacuation failure using the result. Using this result, we deduce the evacuation safe criterion formula only for simple cases. In addition, we compare the deterministic evaluation and our probabilistic evaluation, it is shown that both evaluation will be perfectly agreed, if the safety factors used in deterministic evaluation are decided on the basis of standard deviation which shows the dispersion of each variable got from measured value.
Feasibility study of temperature measurement of pool fire was performed with various measurement techniques ; e.g. thermocouple, spectroscopy in visible range, single-color pyrometer with single band in infrared range (center : 5 μm, 10 μm) and two-color pyrometer with two narrow-bands in visible range (0.65/0.80 μm). Targeted flames were small and middle size of kerosene pool fire. Experiments were performed in fire facility at Fire Research Laboratory of Nagoya City Fire Bureau. It is found that thermocouple might not be adequate for the measurement of luminous flame temperature unless the accumulated soot on its surface, which causes an unrecoverable measurement error, is properly removed. Spectroscopy analysis and two-color pyrometer give the similar temperature and its profile is rather uniform. On the other hand, single-color pyrometers with unity emissive coefficient give lower temperature than one obtained by spectroscopy or two-color pyrometer and its profile is not uniform; rather highest "thermal spot" is formed inside the flame. This difference is due to the corresponding emissive objects; two-color pyrometer counts only the radiation from soot, i.e. luminous flame part, but single-color pyrometer counts not only the radiation of soot but also the heated gas species surrounded by the flame. Such difference is more pronounced when the optical thickness of the radiation object becomes thinner, e.g. small scale flames. As consequence we should pay attention when a single-color pyrometer is applied to temperature measurement of small laboratory flames.
In general, when doing the compression test of concrete with high temperature, we used specimens heated in the furnace attached to compressor and reaching them to the target temperature, however this method needs much time and efforts. In this paper, a new test method by moving the specimens heated with the pre-heating furnace to compressor was proposed. Furthermore, the temperature fall in the case of movement was verified, and the validity of the test method was shown. At last, compression test results (parameters: 36∼180 N/mm2, coarse aggregate, polypropylene, steel fiber) using the proposed method were reported. The following results were obtained: In the case of normal strength concrete, compressive strength remaining ratio was similar to the EUROCODE and AIJ proposal value. On the other hand, compressive strength remaining ratio of high-strength concrete showed the special tendency which grew up at 100 °C differ from normal concrete's.
When the flame propagation of a forest fire is estimated, it is important to quickly know the wind speed and direction at the place where the fire is broken out. This study investigated an estimation method for the wind speed and direction at AMeDAS stations using the wind data observed at the weather stations. From this study, it is found that the averaging method with the weight of the inverse of the square of the distances between the AMeDAS and the weather stations gives reasonable estimates. The targeted areas of this study are Kanto and Shizuoka areas. The data with the wind speed of more than 10 m/s in March and April, 2001, are used as the weather data. In spite of the simplicity of the method, it can reasonably estimate the wind at an arbitrary location on the flat land using the wind data observed at the adjacent weather stations. This is an advantage of the method.