Abstract:

In recent years, fires have often occurred, such as the Notre Dame in Paris. Fire poses a great threat to human production and life and causes great losses to human life and property. In a fire, flashover is a special phenomenon. When an indoor fire occurs, all combustible surfaces in the building will ignite rapidly in a very short period of time, which greatly increases the difficulty of firefighting and rescue. In order to study the effect of fractal parameters of temperature on flashover discrimination in building fire, this paper uses fire dynamics simulator and direct numerical simulation method to calculate the situation of fire in a room with the corridor. According to the calculated results, the temperature field and flow field diagram of the physical model is obtained. On this basis, according to the basic principle of fractal theory, the nonlinearity of the temperature change process of the fire model is obtained. The characteristics of mutagenicity and fractal dimension reduction are analyzed and studied in depth. The temperature Hurst index of the fractal dynamic parameters of temperature sequence is determined by using the fractal theory R/S analysis method, and the Hurst index of the flashover occurs in the fire is calculated by taking the temperature of some point in the room as an example. The relationship between temperature Hurst index and flashover is analyzed and evaluated. It is found that with the occurrence of flashover, the Hurst index has obvious law of decreasing dimension and sudden change, and the time of dimensionality reduction is basically consistent with the time of flashover. This indicates that the temperature Hurst index is an effective nonlinear fractal dynamic parameter for evaluating the occurrence of fire flashover. The results show that the flame jet phenomenon is formed at the middle gate in 100 s and 200 s, and the temperature Hurst index is 0.22089 and 0.26781 at the corresponding time, respectively, which is in the state of anti-persistent series. In 300 seconds and after time, the temperature measured by a thermocouple in the room is in stable fluctuation state, and the Hurst index of temperature in the corresponding time is in a continuous series state. The fractal dimension of 100 s, 200 s and 300 s and later time is obviously reduced, that is, the Hurst index has a sudden change with the occurrence of flashover, and the time of dimensionality reduction is basically consistent with the time of flashover.

Key words: fire, flashover, numerical simulation, fractals, time series analysis, Hurst index, nonlinear dynamics

摘要：

为研究建筑火灾中温度的分形参数对轰燃判别的影响，利用火灾动力学模拟器对所构建的物理模型进行数值计算，在分析温度场和流场的基础上，根据分形理论基本原理，对该火灾模型温度变化过程所具有的非线性、突变性和分形降维特性进行分析与研究。运用分形理论R/S分析法，确定温度时序分形动力学参数的温度Hurst指数，并以室内某点温度为例，计算火灾轰燃发生时的 Hurst 指数。通过分析和评价温度Hurst指数与轰燃发生的关系，发现 Hurst 指数随着轰燃的发生出现降维突变规律，且降维时间与轰燃发生时间相吻合。这表明温度Hurst指数是评价火灾轰燃发生的一种有效的非线性分形动力学参数。

关键词: 火灾, 轰燃, 数值模拟, 分形, 时序分析, Hurst指数, 非线性动力学