Kohyu Satoh

Artificial neural network approach for modeling the impact of population density and weather parameters on forest fire risk.

The risk of forest fire occurrence is affected by the interactions among forest fuels, weather, human activities, etc. In the present paper, we try to build a method to model and forecast forest fire risk based on artificial neural networks. The data considered include population density and several weather parameters, i.e. average relative humidity, wind velocity and daily sunshine hours. With an interpolation method, these data have been expanded into 1 by 1 km meshes that are calculated according to the standard mesh code system in Japan, where the Japanese territory is divided into a lattice by latitude and longitude. Different parameter combinations and corresponding fire probabilities are computed. The correlations between forest fire probability and population density, and sequentially that between forest fire probability and combinations of population density together with one or several weather parameters are analyzed with three back-propagation neural networks in comparison with polynomial regression investigations. The results indicate that non-linear relationships exist among the influential factors and forest fire probability; artificial neural networks could better capture the non-linearity and give closer results to the test set compared with polynomial regression. The proposed method may be used to investigate and forecast forest fire risk providing there are enough data.

Experiments and Analysis of Interaction Among Multiple Fires in Equidistant Fire Arrays

The study of dynamics of multiple fires is important to gain a physical insight of the conditions under which destructive phenomena could result in city fires such those caused by earthquakes. Particularly, heavy populated cities such as Tokyo are highly vulnerable. Unfortunately, previous studies on multiple fires and their dynamics are rather limited. An extensive experimental study has been carried out to examine the fire interactions among freestanding equidistant multiple fires in square arrays, to supplement the authors’ previous related studies. Four square arrays, namely, 5×5, 9×9, 15×15 and 17×17, with various inter-fuel pan distances were treated. The burnout time (BOT) from ignition at every fire in the array was experimentally recorded and expressed as multiples of the BOT of a single free-standing fire as a reference. Since the BOT at any fire location in an array is inversely proportional to an average burning rate (BR) at that location, the local BR can then be directly inferred, and their comparisons thus indications of the physical interactions as affected by the fire location, inter-fuel pan distance and size of the fire array. It is shown that all these parameters play remarkable roles in the interactions among multiple fires in square fire arrays.Copyright

Numerical and Experimental Study of Fire Whirl Generated in 15 × 15 Square Array Fires Placed in Cross Wind

Fire whirls in large city fires and forest fires, which are highly dangerous and destructive, can cause substantial casualties and property damages. It is important to examine under what conditions of weather and geography such merging fires and fire whirls are generated. However, detailed physical characteristics about them are not fully clarified yet. Therefore, we have conducted preliminary studies about merging fires and swirling fires and found that they can enhance the fire spread. If sufficient knowledge can be obtained by relevant experiments and numerical computations, it may be possible to mitigate the damages due to merged fires and fire whirls. The objective of this study is to investigate the swirling conditions of fires in square arrays, applying wind at one corner, in laboratory experiments and also by CFD numerical simulations. Varying the inter-fire distance, heat release rate and mass flow rate by a wind fan, ‘swirling’ or ‘non-swirling’ in the array were judged. It has been found that the fire whirl generation is highly affected by the inter-fire distance in the array, the total heat release rate and also the mass flow rate by a fan. We obtained the conditions of swirling fire generation in 15 × 15 square array for (1) the ratio between the upward mass flow rate vs. applied mass flow rate in the upward swirling plume and (2) a non-dimensional relationship between the heat flow rate in the swirling plume and the applied mass flow rate.Copyright

An experimental study on thermal radiation of fire whirl

Fire whirl is frequently observed in wildland fires, and may cause serious difficulty in firefighting owing to its significant turbulent flow. In this paper, the radiation of fire whirl is investigated through experiments using a fire whirl facility made up of an air curtain apparatus, with five different sizes of n-heptane pools (25, 30, 35, 40 and 45 cm). The flame contour was extracted by image processing. By using infrared methods, the flame emissivity of fire whirl at different heights for different pool diameters was measured, and thereby a correlation was developed between the flame emissivity and the flame diameter. The soot volume fraction in the luminous flame is estimated to range within 2.5 × 10−6 to 4.0 × 10−6, much higher than that of general heptane pool fires, which provides an explanation of the higher flame emissivity of fire whirl. The emissive power profile v. normalised height is deduced from flame emissivity and flame temperature data. A multizone flame model (in which each zone is assumed as a grey body) is used, based on the measured data of flame emissivity, to predict the radiation of fire whirl. Comparison between the predicted and measured data of radiative flux shows good agreement.