Monitoring and predicting the microclimate in naturally ventilated barns (NVB) is important given the adverse effects of high summer temperatures on dairy cows in the context of global climate change. The aim of the work was to verify the accuracy of the microclimate forecast in a NVB using linear regression (LR). Our working hypothesis suggested that multiple periodic measurements of air temperature and relative humidity outside and inside the barns at the same time will allow us to build LR models for predicting the temperature-humidity index (THI). This was done not only for a specific dairy barn based on this indicator outside, but also in other dairy barns with a similar design, located in similar weather conditions. The results of the research indicate that the use of LR had a high accuracy of forecasting (93-96%) the THI in NVB of various designs during the summer heat. At the same time, differences were found between traits (air temperature, relative humidity as well as resulting THI) provided by meteorological weather stations and these data measured simultaneously next to the dairy barns. The proposed LR models can be used to predict THI in NVBs of different designs.

中文翻译：

---

炎热季节自然通风奶牛舍的小气候建模：检查预测的准确性

鉴于在全球气候变化的背景下，夏季高温对奶牛的不利影响，监测和预测自然通风牛舍 (NVB) 的小气候非常重要。这项工作的目的是使用线性回归 (LR) 验证 NVB 中小气候预测的准确性。我们的工作假设表明，同时对谷仓内外的空气温度和相对湿度进行多次定期测量将使我们能够建立用于预测温湿度指数 (THI) 的 LR 模型。这不仅适用于基于此外部指标的特定奶牛场，而且还适用于具有类似设计、位于相似天气条件的其他奶牛场。研究结果表明，在夏季炎热期间，使用LR对各种设计的NVB中的THI具有较高的预测精度（93-96%）。同时，气象气象站提供的性状（气温、相对湿度以及由此产生的 THI）与在奶牛场附近同时测量的这些数据之间存在差异。所提出的 LR 模型可用于预测不同设计的 NVB 中的 THI。