Time Series Data Generation Method Simulating Natural Wind

Abstract

In this paper, a method is proposed for generating time-series data simulating natural wind. This method can be applied to indoor ventilation and air conditioning. The simulation of natural wind described in this paper does not demonstrate a reproduction of the waveform of the sampling data, but rather a reproduction of the statistical characteristics of natural wind. First, we clarified the fundamental relation between the turbulence Intensity and frequency distribution of natural wind. In addition, we analyzed the characteristics of the time series data of natural wind based on a logistic map. We then clarified the stochastic structure of the coefficients used in the recurrence formula of the logistic map. Finally, we proposed a new BCM method through which natural wind is simulated at any mean velocity. In the BCM (Fig. 6) proposed in this paper, the frequency distribution of the wind speed is approximated based on a Weibull distribution. The turbulence intensity and mean wind velocity can be reproduced through a Weibull distribution approximation. The inverse function of the cumulative distribution of the Weibull distribution can be used to generate the elements of the probability distribution. The elements are always newly generated using the inverse function. A logistic map is used to reconstruct the elements of the probability distribution, that is, the wind speed, as time series data. In the logistic map, the probability distribution of the coefficient used in a recurrence formula is approximated using the corrected sigmoid function (Equation (10)). As a result of BCM, the waveform of the time series data (Fig. 10) was similar to that of natural wind. The frequency distribution of the wind speed and the reproduction of the turbulence intensity (Fig. 11) were also properly simulated. In addition, Chapter 5 shows how to adjust the average wind speed as required for ventilation and air conditioning. The purpose of simulating natural wind with air conditioning and ventilation is to provide unsteady wind comfort. Chaos is related to the comfort of unsteady winds, and thus, as an advantage of BCM, the characteristics of chaos can be easily evaluated using the method described in Section 3.1. The acceleration can also be approximated using the BCM * shown in Section 4.3. As a subject for future research, it is necessary to consider the change in wind speed and waveform from fan to living area. This may be achieved by modeling the change in waveform owing to the difference in velocity.