Accurate internal air distribution prediction facilitates the analysis of occupant thermal comfort and the control of gaseous contaminants within naturally ventilated livestock buildings (NVLB). However, the dynamic and random change in external wind conditions makes it difficult to predict the air distribution. This study compares three scenarios based on real-time monitored external wind conditions: the quasi-static simulation method with moving average data, the transient simulation method with raw data, the transient simulation method with Kalman filtering, and the continuous fast Fourier transform. The quasi-static simulation was verified for computational efficiency and reliability within different time spans. The simulation results were validated with measurement data in a low-rise stand-alone NVLB surrounding a low-density building complex. The quasi-static simulation results of internal air distribution in oblique wind directions partially agreed with the hourly averaged wind speed measurements and turbulent kinetic energy. Additionally, the quasi-static simulation agrees with the direct transient method for predicting the air distribution within a long-time span. The quasi-static simulation shows promise for the future simulation of internal air distribution in naturally ventilated livestock buildings under dynamic boundary conditions.