基于CNN-LSTM模型的土壤温湿度缺失数据填补算法  

Missing data filling algorithm for soil temperature and humidity based on CNN-LSTM Model

作　　者：张瑛进 史志强 古丽米拉·克孜尔别克[1] 库木斯·阿依肯 ZHANG Ying-jin;SHI Zhi-qiang;Gulimila Kezierbieke;Kumusi Ayiken(Computer and Information Engineering College,Xinjiang Agricultural University,Urumqi 830052,China;The Seventh Affiliated Hospital,Xinjiang Medical University,Urumqi 830001,China)

机构地区：[1]新疆农业大学计算机与信息工程学院,乌鲁木齐830052 [2]新疆医科大学第七附属医院,乌鲁木齐830001

出　　处：《湖北农业科学》2025年第2期179-183,196,共6页Hubei Agricultural Sciences

基　　金：科技部科技创新2030重大项目(2022ZD0115800);新疆维吾尔自治区重大科技专项(2022A02011-4)。

摘　　要：针对因恶劣环境、电池耗尽、硬件故障等原因导致的土壤温湿度传感器数据丢失问题,提出一种基于卷积神经网络的长短期记忆网络(CNN-LSTM)填补模型。以闪电河流域2019年土壤温湿度数据为试验数据,分别选用CNN、LSTM、TCN、CNN-TCN、CNN-LSTM 5个模型对土壤温湿度传感器网络缺失数据进行填补,并采用Adam算法优化模型,使用决定系数(R^(2))、均方根误差(RMSE)、平均绝对误差(MAE)指数对模型填补结果进行评价。结果表明,采用线性插补算法获得完整的数据,CNN-LSTM模型的R^(2)为0.999 9,高于其他模型,MAE、RMSE分别为0.001 85、0.019 70,均远低于其他模型。采用k近邻插补算法获得完整的数据,CNN-LSTM模型的MAE、RMSE分别为0.000 12、0.000 12,均远低于其他模型,R^(2)为0.999 4,高于CNN模型、TCN模型;CNN-LSTM模型对土壤温湿度传感器数据缺失值的填补效果最好。CNN-LSTM模型在处理大规模土壤温湿度传感器缺失数据问题时具有较好的可行性和精确度。A convolutional neural network-based long short-term memory network(CNN-LSTM)filling model was proposed to ad⁃dress the problem of soil temperature and humidity sensor data loss caused by harsh environments,battery depletion,hardware fail⁃ures,and other factors.Using the soil temperature and humidity data from the Shandian River Basin in 2019 as experimental data,five models including CNN,LSTM,TCN,CNN-TCN,and CNN-LSTM were selected to fill in the missing data of the soil temperature and humidity sensor network.The Adam algorithm was used to optimize the model,and the coefficient of determination(R^(2)),mean square root error(RMSE),and mean absolute error(MAE)index were used to evaluate the results of the model filling.The results showed that using the linear interpolation algorithm to obtain complete data,the R^(2) of the CNN-LSTM model was 0.9999,which was higher than that of other models.The MAE and RMSE were 0.00185 and 0.01970,respectively,which were much lower than those of other models.The K-nearest neighbor interpolation algorithm was used to obtain complete data.The MAE and RMSE of the CNNLSTM model were 0.00012 and 0.00012,respectively,which were much lower than those of other models.The R^(2) was 0.9994,which was higher than that of the CNN model,and TCN model;the CNN-LSTM model had the best filling effect on missing values in soil temperature and humidity sensor data.The CNN-LSTM model had good feasibility and accuracy in dealing with the problem of missing data from large-scale soil temperature and humidity sensors.

关 键 词：CNN-LSTM模型 土壤 温湿度 缺失数据填补算法 

分 类 号：TP389.1[自动化与计算机技术—计算机系统结构]