[Objective] The aim was to study the coupling effect of water and phosphate on economic traits of sugarcane. [Method] Taking sugarcane variety ROC22 as tested material,coupling effects of different levels of water sup...[Objective] The aim was to study the coupling effect of water and phosphate on economic traits of sugarcane. [Method] Taking sugarcane variety ROC22 as tested material,coupling effects of different levels of water supply quantity and different levels of phosphorus fertilizer on the yield and quality of sugarcane were studied. Among them,water supply quantity had 3 levels,that was,the water supply quantity per 10 days from the early tillering stage of sugarcane to the end of elongation was 199.5 m3/hm2 (A1),400.5 m3/hm2 (A2) and 600.0 m3/hm2 (A3),respectively; Phosphorus fertilizer as basic fertilizer had 4 levels:P2O5 0 kg/hm2 (B1),120 kg/hm2 (B2),240 kg/hm2 (B3) and 360 kg/hm2 (B4). [Result] Treatment A3B2 in water-fertilizer coupling was more suitable to improve economic traits of sugarcane. [Conclusion] The research results provide theoretical basis for the efficient utilization of water and phosphorus fertilizer in production of Guangxi sugarcane and the cultivation of high-yield and high-glucose sugarcane.展开更多
由于水质数据特征复杂、关联度参差不齐而导致溶解氧浓度预测难度较大,为提高水质溶解氧浓度预测的准确性,提出了一种基于特征工程和北方苍鹰优化算法的长短期记忆网络(Feature Engineering-Northern Goshawk Optimization-Long Short T...由于水质数据特征复杂、关联度参差不齐而导致溶解氧浓度预测难度较大,为提高水质溶解氧浓度预测的准确性,提出了一种基于特征工程和北方苍鹰优化算法的长短期记忆网络(Feature Engineering-Northern Goshawk Optimization-Long Short Term Memory,FE-NGO-LSTM)混合模型。首先对水质数据集进行缺失值补齐、特征筛选与特征多项式构造,然后基于NGO-LSTM模型优化模型参数,提升预测性能;对不同多项式阶数下的特征预测效果进行分析之后,将该模型与基于灰狼优化算法、鲸鱼优化算法及粒子群优化算法的LSTM模型进行对比;最后,在太湖流域东苕溪城南监测断面对该模型进行了验证,计算FE-NGO-LSTM模型预见期为4,8,12,16,20,24 h的预测结果。试验结果显示:当多项式阶数为2阶时,模型预测效果最好,FE-NGO-LSTM模型相比基于其他优化算法的LSTM模型,平均绝对误差、均方误差、均方根误差分别至少降低9.0%,12.9%及6.3%,且随着预见期的增加,预测误差仍在可接受范围内,说明FE-NGO-LSTM模型在预测溶解氧浓度时具有一定优势与泛化性。展开更多
基金Supported by National Science and Technology Project of China(2007BAD30B04)~~
文摘[Objective] The aim was to study the coupling effect of water and phosphate on economic traits of sugarcane. [Method] Taking sugarcane variety ROC22 as tested material,coupling effects of different levels of water supply quantity and different levels of phosphorus fertilizer on the yield and quality of sugarcane were studied. Among them,water supply quantity had 3 levels,that was,the water supply quantity per 10 days from the early tillering stage of sugarcane to the end of elongation was 199.5 m3/hm2 (A1),400.5 m3/hm2 (A2) and 600.0 m3/hm2 (A3),respectively; Phosphorus fertilizer as basic fertilizer had 4 levels:P2O5 0 kg/hm2 (B1),120 kg/hm2 (B2),240 kg/hm2 (B3) and 360 kg/hm2 (B4). [Result] Treatment A3B2 in water-fertilizer coupling was more suitable to improve economic traits of sugarcane. [Conclusion] The research results provide theoretical basis for the efficient utilization of water and phosphorus fertilizer in production of Guangxi sugarcane and the cultivation of high-yield and high-glucose sugarcane.
文摘由于水质数据特征复杂、关联度参差不齐而导致溶解氧浓度预测难度较大,为提高水质溶解氧浓度预测的准确性,提出了一种基于特征工程和北方苍鹰优化算法的长短期记忆网络(Feature Engineering-Northern Goshawk Optimization-Long Short Term Memory,FE-NGO-LSTM)混合模型。首先对水质数据集进行缺失值补齐、特征筛选与特征多项式构造,然后基于NGO-LSTM模型优化模型参数,提升预测性能;对不同多项式阶数下的特征预测效果进行分析之后,将该模型与基于灰狼优化算法、鲸鱼优化算法及粒子群优化算法的LSTM模型进行对比;最后,在太湖流域东苕溪城南监测断面对该模型进行了验证,计算FE-NGO-LSTM模型预见期为4,8,12,16,20,24 h的预测结果。试验结果显示:当多项式阶数为2阶时,模型预测效果最好,FE-NGO-LSTM模型相比基于其他优化算法的LSTM模型,平均绝对误差、均方误差、均方根误差分别至少降低9.0%,12.9%及6.3%,且随着预见期的增加,预测误差仍在可接受范围内,说明FE-NGO-LSTM模型在预测溶解氧浓度时具有一定优势与泛化性。