A long-term N₂O dataset from a full-scale biological process was analysed for knowledge discovery. Non-parametric, multivariate timeseries changepoint detection techniques were applied to operational variables (i.e. NH₄-N loads) in the system. The majority of changepoints, could be linked with the observed changes of the N₂O emissions profile. The results showed that even three-day sampling campaigns between changepoints have a high probability (>80%) to result to an emission factor (EF) quantification with ~10% error. The analysis revealed that support vector machine (SVM) classification models can be trained to detect operational behaviour of the system and the expected range of N₂O emission loads. The proposed approach can be applied when long-term online sampling is not feasible (due to budget or equipment limitations) to identify N₂O emissions “hotspot” periods and guide towards the identification of operational periods requiring extensive investigation of N₂O pathways generation.

分析了来自大规模生物过程的长期N ₂ O数据集，以发现知识。将非参数多元时间序列变化点检测技术应用于系统中的操作变量（即NH ₄ -N负荷）。大多数变化点都可以与N ₂ O排放曲线的观测变化联系在一起。结果表明，即使在变更点之间进行三天的采样活动，也有很高的可能性（> 80％）导致排放因子（EF）定量，误差约为10％。分析显示，可以训练支持向量机（SVM）分类模型来检测系统的操作行为和N ₂的预期范围O排放负荷。当无法进行长期在线采样（由于预算或设备限制）而不能确定N ₂ O排放的“热点”时期并指导确定需要大量研究N ₂ O途径产生的运行时期时，可以采用建议的方法。