Soft Computing ( IF 3.1 ) Pub Date : 2021-07-12 , DOI: 10.1007/s00500-021-06009-4 Akram Seifi 1 , Fatemeh Soroush 1 , Mohammad Ehteram 2 , Fatemeh Nayebloei 3 , Bahram Gharabaghi 4 , Ali Torabi Haghighi 5
|
Accurate prediction of soil temperature (Ts) is critical for efficient soil, water and field crop management. In this study, hourly Ts variations at 5, 10, and 30 cm soil depth were predicted for an arid site (Sirjan) and a semi-humid site (Sanandaj) in Iran. Existing machine learning models have high performance, but suffer from uncertainty and instability in prediction. Therefore, GLUE approach was implemented to quantify model uncertainty, while wavelet coherence was used to assess interactions between Ts and meteorological parameters. Standalone machine learning models (adaptive neuron fuzzy interface system (ANFIS), support vector machine model (SVM), radial basis function neural network (RBFNN), and multilayer perceptron (MLP)) were hybridized with four optimization algorithms (sunflower optimization (SFO), firefly algorithm (FFA), salp swarm algorithm (SSA), particle swarm optimization (PSO)) to improve Ts prediction accuracy and reduce model uncertainty. For both arid and semi-humid sites, ANFIS-SFO produced the most accurate performance at studied soil depths. At best, hybridization with SFO (ANFIS-SFO, MLP-SFO, RBFNN-SFO, SVM-SFO) decreased RMSE by 5.6%, 18%, 18.3%, and 18.2% at 5 cm, 11.8%, 10.4%, 10.6%, and 12.5% at 10 cm, and 9.1%, 12.1%, 13.9%, and 14.2% at 30 cm soil depth compared with the respective standalone models. GLUE analysis confirmed the superiority of hybrid models over the standalone models, while the hybrid models decreased the uncertainty in Ts predictions. ANFIS-SFO covered 95%, 94%, and 96% observation data at 5, 10, and 30 cm soli depths, respectively. Wavelet coherence analysis demonstrated that air temperature, relative humidity, and solar radiation, but not wind speed, had high coherence with Ts at different soil depths at both sites, and meteorological parameters mostly influenced Ts in upper soil layers. In conclusion, uncertainty analysis is a necessary and powerful technique to obtain an accurate and realistic prediction of Ts. In contrast, wavelet coherence analysis is a useful tool to investigate the most effective variables that strongly affect predictions.
Graphic abstract
中文翻译：
预测每小时土壤温度的混合模型的 GLUE 不确定性分析和应用小波相干分析与气象变量的相关性
土壤温度（准确预测Ť小号）是高效土壤，水和大田作物管理至关重要。在这项研究中，预测了伊朗干旱地区 (Sirjan) 和半湿润地区 (Sanandaj) 在 5、10 和 30 cm 土壤深度处的每小时T s变化。现有的机器学习模型具有高性能，但存在预测的不确定性和不稳定性。因此，实施了 GLUE 方法来量化模型的不确定性，而小波相干性用于评估T s之间的相互作用和气象参数。独立机器学习模型（自适应神经元模糊接口系统 (ANFIS)、支持向量机模型 (SVM)、径向基函数神经网络 (RBFNN) 和多层感知器 (MLP)）与四种优化算法（向日葵优化 (SFO)）混合、萤火虫算法（FFA）、salp swarm 算法（SSA）、粒子群优化（PSO））来提高T s预测准确性并减少模型的不确定性。对于干旱和半湿润地点，ANFIS-SFO 在研究的土壤深度上产生了最准确的性能。在最好的情况下，与 SFO（ANFIS-SFO、MLP-SFO、RBFNN-SFO、SVM-SFO）的杂交在 5 cm、11.8%、10.4%、10.6% 处将 RMSE 降低了 5.6%、18%、18.3% 和 18.2%和 10 cm 处的 12.5%，以及 30 cm 土壤深度处的 9.1%、12.1%、13.9% 和 14.2% 与各自的独立模型相比。GLUE 分析证实了混合模型优于独立模型，而混合模型降低了T s的不确定性预测。ANFIS-SFO 分别覆盖了 5、10 和 30 cm 固体深度的 95%、94% 和 96% 的观测数据。小波相干性分析表明，气温、相对湿度和太阳辐射（而非风速）与两个站点不同土壤深度的T s具有高度相干性，气象参数主要影响上层土层的T s。总之，不确定性分析是获得准确和现实的T s预测的必要且强大的技术。相比之下，小波相干分析是研究强烈影响预测的最有效变量的有用工具。
京公网安备 11010802027423号