用于碳通量长时间缺失值插补的深度学习模型ARformer  

Constructing ARformer:A deep learning model for the long-term gap-filling of carbon flux data

作　　者：齐建东[1,2] 吴鹏 查天山[3] QI Jiandong;WU Peng;ZHA Tianshan(School of Information Science and Technology,Beijing Forestry University,Beijing 100083,China;Engineering Research Center for Forestry-oriented Intelligent Information Processing of National Forestry and Grassland Administration,Beijing 100083,China;School of Soil and Water Conservation,Beijing Forestry University,Beijing 100083,China)

机构地区：[1]北京林业大学信息学院,北京100083 [2]国家林业和草原局林业智能信息处理工程技术研究中心,北京100083 [3]北京林业大学水土保持学院,北京100083

出　　处：《农业工程学报》2024年第24期127-136,共10页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家重点研发计划项目(2020YFA0608100);国家自然科学基金项目(32071842)。

摘　　要：为提高净生态系统交换量(net ecosystem exchange,NEE)在长期缺失下的插补精度,为地球碳循环的后续研究提供关键数据,该研究提出了一种用于NEE长时间缺失值插补的模型Adapter-Reverseformer(ARformer)。首先,为了更好地捕捉NEE数据的时序特征,重新设计和优化了层归一化(layer normalization)、前馈网络(feed-forward network)和自注意力(self-attention)3个模块的功能,模型能够使用更长的回望窗口,从更长的时间序列中捕获时序特征;其次,设计了特征融合模块Attention-MLP,以拟合NEE与环境因子的即时响应关系,提高模型在不同土地利用类型NEE数据上的插补精度。在全球长期通量观测网络(FLUXNET)65个站点、10种土地利用类型的数据上进行试验,结果显示,ARformer模型的R^(2)在0.762~0.913之间,均方根误差(root mean square error,RMSE)、平均绝对误差(mean absolute error,MAE)和偏差(bias)在0.668~2.724、0.410~1.751和−0.024~0.067μmol/(m^(2)·s)之间。在不同缺失长度和土地利用类型下插补精度均高于边际分布采样法(marginal distribution sampling,MDS)、随机森林(random forest,RF)、DLinear、PatchTST和iTransformer。该研究结果可为NEE数据在长时间缺失场景下的插补提供参考。Net ecosystem exchange(NEE)measurement is very critical to understanding carbon flux in ecosystems.But some gaps are still common in data collection,due to the harsh weather or sensor malfunction.Traditional interpolation can struggle with the long-term missing data,leading to inaccuracies in carbon flux analysis.Long-term missing data can also occur in NEE measurements.In this study,the Adapter-Reverseformer(ARformer)model was proposed to enhance the accuracy of NEE gap-filling,especially for extended periods of data loss.A multi-layer perceptron(MLP)was integrated with the Reverseformer module.The longer data gaps were effectively utilized to leverage both environmental data and the temporal patterns in the NEE.A novel system was established to focus on the non-linear relationships between NEE and environmental factors.The time-dependent trend was characterized by NEE data.The obtained model was tested using the FLUXNET 2015 dataset.Half-hourly carbon flux data was collected from 65 sites across 10 types of land use.Five artificial gap scenarios were generated by randomly removing data for continuous periods of 1,7,15,30,and 90 days.The performance of ARformer was compared with marginal distribution sampling(MDS),random forest(RF),and three advanced deep learning models:DLinear,PatchTST,and iTransformer.The results demonstrated that the ARformer model consistently outperformed the baseline,especially when dealing with long-term missing data.Specifically,the performance of RF decreased significantly,when the missing data spanned 90 days,and MDS failed to reasonably estimate the model.In contrast,the ARformer model maintained a high accuracy,with R^(2) values ranging from 0.762 to 0.913.The root mean square error(RMSE)ranged between 0.668 and 2.724μmol/(m^(2)·s),the mean absolute error(MAE)ranged from 0.410 to 1.751μmol/(m^(2)·s),and bias values remained between−0.024 and 0.067μmol/(m^(2)·s).The ARformer model demonstrated superior performance across different land-use types,including closed shrublands,de

关 键 词：碳通量 Adapter-Reverseformer 深度学习 长缺失插补 

分 类 号：S24[农业科学—农业电气化与自动化]