Importance of ozone precursors information in modelling urban surface ozone variability using machine learning algorithm

Abstract

Surface ozone (O$$_3$$ 3 ) is primarily formed through complex photo-chemical reactions in the atmosphere, which are non-linearly dependent on precursors. Even though, there have been many recent studies exploring the potential of machine learning (ML) in modeling surface ozone, the inclusion of limited available ozone precursors information has received little attention. The ML algorithm with in-situ NO information and meteorology explains 87% (R$$^{2}$$ 2 = 0.87) of the ozone variability over Munich, a German metropolitan area, which is 15% higher than a ML algorithm that considers only meteorology. The ML algorithm trained for the urban measurement station in Munich can also explain the ozone variability of the other three stations in the same city, with R$$^{2}$$ 2 = 0.88, 0.91, 0.63. While the same model robustly explains the ozone variability of two other German cities’ (Berlin and Hamburg) measurement stations, with R$$^{2}$$ 2 ranges from 0.72 to 0.84, giving confidence to use the ML algorithm trained for one location to other locations with sparse ozone measurements. The inclusion of satellite O$$_3$$ 3 precursors information has little effect on the ML model’s performance.

Automatic fault location method for power communication network based on a machine learning algorithm

“…Since the individual fault signal output by the electricity communication system cannot be used as the main criterion to measure the fault characteristics, only when the scale of the fault signal reaches a certain magnitude can it be used as the judgment criterion to judge the power communication network fault problem. Considering that there are grade differences in the degree of abnormality of power communication networks, it is difficult to characterize the degree of abnormality by the quantitative results of fault signal scale [11]. In this regard, this paper takes the single fault signal of the electricity communication group system as the basis, and differentiates the weight parameters of different abnormality degrees so as to realize the effective judgment of the abnormal status and the particular implementation approach.…”

Spatiotemporal modeling of air pollutant concentrations in Germany using machine learning

“…Machine learning (ML) models have been shown to be an effective complement to these computationally expensive CTMs (Vlasenko et al, 2021). The performance of machine learning models for modeling air pollutants is promising (Balamurugan et al, 2022a;Cheng et al, 2022;Lee et al, 2020;Li et al, 2023;Liang et al, 2020;Liu et al, 2022;Zaini et al, 2022;Zhao et al, 2023). Meteorological variables such as solar radiation and temperature have been shown to be important parameters in near-surface ozone modeling using machine learning (Diao et al, 2021;Hu et al, 2021).…”