Division of Computing and Software Systems, University of Washington Bothell, Bothell, Washington 98011, United States.
Department of Civil and Environmental Engineering, University of Washington, Seattle, Washington 98195, United States.
Environ Sci Technol. 2020 Dec 1;54(23):15320-15328. doi: 10.1021/acs.est.0c02549. Epub 2020 Nov 17.
Although the exposure to PM has serious health implications, indoor PM monitoring is not a widely applied practice. Regulations on the indoor PM level and measurement schemes are not well established. Compared to other indoor settings, PM prediction models for large office buildings are particularly lacking. In response to these challenges, statistical models were developed in this paper to predict the PM concentration in well-mixed indoor air in a commercial office building. The performances of different modeling methods, including multiple linear regression (MLR), partial least squares regression (PLS), distributed lag model (DLM), least absolute shrinkage selector operator (LASSO), simple artificial neural networks (ANN), and long-short term memory (LSTM), were compared. Various combinations of environmental and meteorological parameters were used as predictors. The root-mean-square error (RMSE) of the predicted hourly PM was 1.73 μg/m for the LSTM model and in the range of 2.20-4.71 μg/m for the other models when regulatory ambient PM data were used as predictors. The LSTM models outperformed other modeling approaches across the performance metrics used by learning the predictors' temporal patterns. Even without any ambient PM information, the developed models still demonstrated relatively high skill in predicting the PM levels in well-mixed indoor air.
尽管 PM 暴露对健康有严重影响，但室内 PM 监测并不是广泛应用的实践。关于室内 PM 水平和测量方案的规定尚未得到很好的制定。与其他室内环境相比，大型办公楼的 PM 预测模型尤其缺乏。针对这些挑战，本文开发了统计模型来预测商业办公楼内混合良好的室内空气中的 PM 浓度。比较了不同建模方法的性能，包括多元线性回归（MLR）、偏最小二乘回归（PLS）、分布滞后模型（DLM）、最小绝对收缩和选择算子（LASSO）、简单人工神经网络（ANN）和长短时记忆（LSTM）。将各种环境和气象参数组合用作预测因子。当使用监管环境 PM 数据作为预测因子时，LSTM 模型预测的每小时 PM 的均方根误差（RMSE）为 1.73μg/m，而其他模型的 RMSE 在 2.20-4.71μg/m 范围内。通过学习预测因子的时间模式，LSTM 模型在使用的性能指标上优于其他建模方法。即使没有任何环境 PM 信息，所开发的模型在预测混合良好的室内空气中的 PM 水平方面仍表现出相对较高的技能。
