首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到19条相似文献,搜索用时 109 毫秒
1.
基于BP神经网络的土壤水力学参数预测   总被引:8,自引:1,他引:7  
为了获取区域土壤水分和溶质运移模拟所需的土壤水力学参数,利用黄淮海平原曲周县的试验资料建立基于BP神经网络的土壤转换函数模型。本文采用土壤粒径分布、容重、有机质含量等土壤基本理化性质,来预测土壤饱和导水率Ks、饱和含水量sθ、残余含水量θr、以及van Genuchten公式参数α、n的对数形式ln(α)和ln(n),并与多元线性逐步回归方法进行比较。t检验结果表明,BP神经网络训练和预测得到的模拟值与实测值之间吻合很好,该方法具有较高的预测精度。通过对平均相对误差的比较,得出在粒径分布的基础上增加容重、有机质含量等输入项目,可以提高部分土壤水力学参数的预测精度,而有些参数的预测精度反而降低。以误差平方和为标准的比较结果表明,BP神经网络模型的预测效果总的来看要优于多元线性回归法。  相似文献   

2.
基于网格搜索和交叉验证支持向量机的地表土壤容重预测   总被引:1,自引:1,他引:0  
为了改善支持向量机(SVM)对地表(0~2 cm)土壤容重预测的应用可行性及其效果,针对传统经验法选择SVM惩罚因子C和核函数参数g可能造成的较大误差的问题,提出了一种对SVM参数进行优化的方法—网格搜索与交叉验证相结合的方法。本文采用所提出的参数优化方法,以黄土高原区旱作农田土壤表层容重年度跟踪试验数据样本为依据,选取土壤(0~2 cm)粒径分布、有机质含量、体积含水率、累积接受水量和全盐量为输入变量,建立了地表土壤容重的SVM预测模型。结果表明:所建SVM模型预测值和试验实测值之间不存在显著性差异,利用SVM预测地表土壤容重是可行的;采用网格搜索与交叉验证相结合的方法对SVM参数进行优化,明显降低了模型的预测误差;在粒径分布、有机质含量、体积含水率为输入变量的基础上,增加全盐量为输入因子并不能显著提高模型的预测效果,而增加累积接受水量为输入因子的预测效果明显优于前几种情况,其训练样本和测试样本相对误差的平均值分别为6.23%和6.95%,都在可接受范围。研究成果可为土壤表层容重的实时预测提供有力支撑。  相似文献   

3.
以科尔沁沙地典型坨-甸相间地区为研究区,野外布设240个采样点,对流动沙丘、半固定沙丘、固定沙丘、沙丘区杨树林、沙丘区耕地、低覆盖度草甸、高覆盖度草甸、草甸区耕地、撂荒地9种地貌类型下的表层土壤进行了采样,测定了其含水率、干容重、有机质、饱和导水率等理化特性,分析了不同地貌类型下表层土壤理化参数差异。选取Campbell、Cosby、Wosten等、Saxton等4种土壤饱和导水率传递函数,对该地区表土饱和导水率进行了预测。结果显示这几种土壤传递函数预测值与实测值偏差较大,相关系数均小于0.3,精度难以满足本地区应用。在此基础上,选取土壤容重、有机质含量、饱和含水率、平均粒径、粒径标准偏差5种土壤特性参数作为输入变量,采用主成分分析与非线性回归分析相结合的方法,重新建立了预测本地区表土饱和导水率的土壤传递函数,结果显示预测值与实测值相关系数为0.661,该传递函数可用于科尔沁沙地表层土壤饱和导水率的预测。  相似文献   

4.
本研究以科尔沁沙地典型坨-甸相间地区为研究区,野外布设240个采样点,对流动沙丘、半固定沙丘、固定沙丘、沙丘区杨树林、沙丘区耕地、低覆盖度草甸、高覆盖度草甸、草甸区耕地、撂荒地九种地貌类型下的表层土壤进行了采样,测定了这些样点的含水率、干容重、有机质、饱和导水率等理化特性参数,分析了不同地貌类型下表层土壤理化参数差异。选取Campbell、Cosby、Wosten1997/1999、Saxton四种土壤传递函数,对该地区表土饱和导水率进行了预测,结果显示这几种土壤传递函数预测值与实测值偏差较大,相关系数均小于0.3,由此可见这几种传统的土壤传递函数在本地区应用具有一定的局限性。在此基础上,选取土壤容重、有机质含量、饱和含水率、平均粒径、粒径标准偏差五种土壤特性参数作为输入变量,采用逐步多元统计回归、主成分分析及非线性回归分析相结合的方法,重新建立了预测本地区表土饱和导水率的土壤传递函数,结果显示预测值与实测值相关系数为0.648,该传递函数可应用于科尔沁沙地表层土壤饱和导水率的预测。  相似文献   

5.
曾胤  陆宇振  杜昌文  周健民 《土壤学报》2014,51(6):1262-1269
快速测定土壤有机质含量对作物生产和土壤肥力评价具有重要意义,红外光声光谱技术的应用为土壤有机质快速测定提供了可能。本研究以江苏省南京市溧水区水稻土土样为材料,探究了红外光声光谱技术在有机质测定中的应用。采用主成分分析、偏最小二乘和独立成分分析,分别提取了土壤光谱的主成分、偏最小二乘潜变量和独立成分,并以提取的信息输入支持向量机,从而构建了三种支持向量机校正模型。同时,偏最小二乘也被用于建立校正模型,作为支持向量机模型的对照。预测结果表明,基于独立成分的支持向量机模型效果最好,预测相关系数R2、均方根误差RMSEP和实际测量值的标准差与光谱模型预测值标准差的比值即RPD值分别为0.808、0.575和2.28。F检验表明,该模型显著优于基于主成分的支持向量机模型,但与基于偏最小二乘潜变量的支持向量机模型,以及经典偏最小二乘模型没有显著差异。t检验表明,各校正模型对有机质的预测结果与化学测定结果没有显著差异。因此,红外光声光谱技术为土壤有机质的快速测定提供了新的技术手段。  相似文献   

6.
水蚀风蚀交错带土壤剖面水力学性质变异   总被引:6,自引:0,他引:6  
土壤剖面水力学性质的确定是土壤水分动态预测的基础。该文在水蚀风蚀交错区六道沟流域分别对居于坡中和坡上两块样地160 cm土层不同深度未扰动土壤的水分特征曲线进行了测定,将Van Genuchtens水分特征曲线模式与Mualem导水模式相结合,确定了两样地土壤剖面的水力学参数,对水力学参数在剖面的变化进行了分析。结果表明,土壤剖面饱和含水率、滞留含水率、进气吸力倒数和孔隙大小分布因子沿剖面变化不大,滞留含水率、进气吸力倒数属于中等程度变异,饱和含水率和孔隙大小分布指标属于弱变异,但经方差检验均不显著,说明该地区160 cm土壤剖面可以处理成均质剖面。  相似文献   

7.
科尔沁沙地土壤水分特征曲线传递函数的构建与评估   总被引:8,自引:2,他引:6  
为了快速准确的获取某一区域的水力性质,该文以科尔沁沙地典型沙丘-草甸相间地区为研究区,在对该区49个不同地貌类型采样点土壤水分特征曲线与土壤基本物理化学特性参数测试分析的基础上,采用函数参数非线性规划法构建了土壤干容重、粒径分布、有机质、pH值、电导率值等基本参数与水分特征曲线之间的传递函数,并进行了精度评估与分析。结果表明:1)研究区土壤水分特征曲线的坡度陡峭,不同地貌类型与土地利用方式下,土壤水分特征曲线有较大差异,相同负压下,土壤持水量按照流动沙丘-半固定沙丘-固定沙丘-农田-草甸的顺序递增;土壤的供水能力按照流动沙丘-半固定沙丘-固定沙丘-草甸-农田的顺序递减;2)利用土壤基本物化特性参数通过函数参数非线性规划法建立了研究区土壤水分特征曲线的传递函数,干容重和砂粒含量是预测土壤水分特征曲线模型参数的主要变量,增加土壤的理化指标可以提高预测精度,然而有机质含量、pH值、电导率值对本区土壤水分常数的影响并不大,对水分特征曲线模型3个参数的影响略微增加;3)通过对传递函数的检验与精度评估分析,各参数的平均误差均在0附近;饱和含水率、残余含水率的均方根误差分别为0.017、0.023;土壤水分常数的相关系数在0.95附近,饱和含水率、残余含水率的误差比的几何标准偏分别为1.04、1.27。表明所建土壤水分特征曲线传递函数的精度较高,可用于该区土壤水分特性研究。该研究可为该区水分、溶质运移、水-热-盐耦合运移模拟提供技术支持和理论保证。  相似文献   

8.
采用Gompertz函数的水稻土压缩特性研究   总被引:4,自引:1,他引:3  
土壤压实模型是预测压实破坏的常用方法,但土壤压实模型的应用常因输入参数(土壤压缩特性及其与不同土壤物理性质之间的关系)的缺乏而受到限制。为定量地评价土壤水力学性质和土壤结构对土壤压缩特性的影响,该文利用土壤固结仪对25种不同含水率和容重的重塑土样进行单轴压缩试验,并采用Gompertz函数对试验数据进行拟合以获取土样的回弹指数、压缩指数和先期固结压力。试验结果表明,Gompertz函数对水稻土试验数据的拟合效果较优,决定系数为0.991~0.999。水稻土回弹指数为0.003~0.138,与容重呈负相关,与含水率呈正相关。水稻土压缩指数为0.115~0.839,与容重呈负相关,与含水率呈二次多项式关系。水稻土先期固结压力为33~127k Pa,与容重呈正相关,与含水率呈负相关。该研究建立的土壤压缩特性与含水率和容重之间的传递函数,可用于大尺度范围内水稻土压缩特性的预测;同时这些传递函数可作为土壤压实模型的输入参数,用于农业机械作业引起的压实破坏的量化和土壤压实风险的评估。  相似文献   

9.
应用土壤质地预测干旱区葡萄园土壤饱和导水率空间分布   总被引:7,自引:4,他引:3  
田间表层土壤饱和导水率的空间变异性是影响灌溉水分入渗和土壤水分再分布的主要因素之一,研究土壤饱和导水率的空间变化规律,有助于定量估计土壤水分的空间分布和设计农田的精准灌溉管理制度。为了探究应用其他土壤性质如质地、容重、有机质预测土壤饱和导水率空间分布的可行性,试验在7.6 hm2的葡萄园内,采用均匀网格25 m×25 m与随机取样相结合的方式,测定了表层(0~10 cm)土壤饱和导水率、粘粒、粉粒、砂粒、容重和有机质含量,借助经典统计学和地统计学,分析了表层土壤饱和导水率的空间分布规律、与土壤属性的空间相关性,并对普通克里格法、回归法和回归克里格法预测土壤饱和导水率空间分布的结果进行了对比。结果表明:1)土壤饱和导水率具有较强的变异性,平均值为1.64 cm/d,变异系数为1.17;2)表层土壤饱和导水率60%的空间变化是由随机性或小于取样尺度的空间变异造成;3)土壤饱和导水率与粘粒、粉粒、砂粒和有机质含量具有一定空间相关性,而与土壤容重几乎没有空间相关性;4)在中值区以土壤属性辅助的回归克里格法对土壤饱和导水率的预测精度较好,在低值和高值区其与普通克里格法表现类似。研究结果将为更好地描述土壤饱和导水率空间变异结构及更准确地预测其空间分布提供参考。  相似文献   

10.
基于图像处理和SVR的土壤容重与土壤孔隙度预测   总被引:5,自引:5,他引:0  
杨玮  兰红  李民赞  孟超 《农业工程学报》2021,37(12):144-151
土壤容重和土壤孔隙度是衡量土壤结构的重要参数。传统的土壤容重、土壤孔隙度获取方法费时费力,且大多数预测模型的输入变量获取难度较高。该研究利用土壤粗糙度、土壤阻力与土壤容重的相关关系,以土壤表面图像的颜色参数和纹理参数表征土壤粗糙度,同使用车载式土壤阻力测量系统获得的土壤阻力一起,从信息融合的角度构建了支持向量机回归(Support Vector Regression,SVR)土壤容重预测模型和SVR土壤孔隙度预测模型。图像处理使用HSV颜色空间进行阈值分割,得到HSV颜色参数,纹理参数使用灰度共生矩阵的能量、熵、对比度和逆方差。使用主成分分析对颜色参数和纹理参数进行主成分提取。将SVR模型的预测结果同环刀法测得的标准值进行相关性分析,决定系数R2达到了0.867。土壤孔隙度SVR预测模型决定系数R2达到了0.743。在相同的运行环境下,将SVR模型与决策树回归模型结果做了对比,决策树回归对土壤容重和土壤孔隙度的预测精度R2分别为0.734和0.690,验证得到SVR预测模型具有更好的预测精度。研究可为节省试验成本,以及快速、有效预测土壤容重和土壤孔隙度提供方法和参考。  相似文献   

11.
为筛选和构建适合苏北沿海滩涂围垦农田耕层土壤饱和水力传导率间接估算的土壤转换函数,在典型地块实测土壤饱和导水率和相关土壤基本性质的基础上,分析了11种根据基本土壤性质预测饱和导水率的转换函数方法的适用性,同时探讨了基于人工神经网络方法的土壤转换函数的预测效果。结果表明:滩涂围垦农田耕层土壤平均饱和导水率为10.04 cm/d,属低透水强度;在现有的土壤饱和导水率转换函数中,Vereecken函数是最适合滩涂围垦农区土壤、具有最佳预测精度的转换函数,其预测均方根误差为8.154,其次是Li、Campbell和Rawls函数;以砂粒、粘粒、容重和有机质作为输入因子,基于人工神经网络的土壤转换函数较Vereecken函数其预测均方根误差降至7.920,在输入因子中增加土壤盐分指标可进一步提高饱和导水率的预测精度,其预测均方根误差降为7.634。本文的研究结果显示利用人工神经网络方法建立的转换函数可有效提高滩涂盐渍农田土壤饱和导水率的预测精度。  相似文献   

12.
ABSTRACT

Pedotransfer functions (PTFs), as an indirect forecasting method, offer an alternative for labor-intensive bulk density (BD) measurements. In order to improve the forecasting accuracies, support vector machine (SVM) method was first used to develop PTFs for predicting BD. Cross-validation and grid-search methods were used to automatically determine the SVM parameters in the forecasting process. Soil texture and organic matter content were selected as input variables based on results of predecessors, coupled with gray correlation theory. And additional properties were added as inputs for improving PTF's accuracy and reliability. The performance of the PTF established by SVM method was compared with artificial neural network (ANN) method and published PTFs using two indexes: root-mean-square error (RMSE) and coefficient of determination(R2). Results showed that the average RMSE of published PTFs was 0.1053, and the R2 was 0.4558. The RMSE of ANN–PTF was 0.0638, and the R2 was 0.7235. The RMSE of SVM–PTF was 0.0558, and the R2 was 0.7658. Apparently, the SVM–PTF had better performance, followed by ANN–PTF. Additionally, performances could be improved when accumulated receiving water was added as predictor variable. Therefore, the first application of SVM data mining techniques in the prediction of soil BD was successful, improved the accuracy of predictions, and enhanced the function of soil PTFs. The idea of developing PTFs using SVM method for predicting soil BD in the study area could provide a reference for other areas.  相似文献   

13.
14.
Abstract

Pedotransfer functions (PTFs), predicting the soil water retention curve (SWRC) from basic soil physical properties, need to be validated on arable soils in Norway. In this study we compared the performance of PTFs developed by Riley (1996), Rawls and Brakensiek (1989), Vereecken et al. (1989), Wösten et al. (1999) and Schaap et al. (2001). We compared SWRCs calculated using textural composition, organic matter content (SOM) and bulk density as input to these PTFs to pairs of measured water content and matric potential. The measured SWRCs and PTF input data were from 540 soil horizons on agricultural land in Norway. We used various statistical indicators to evaluate the PTFs, including an integrated index by Donatelli et al. (2004). The Riley PTFs showed good overall performance. The soil specific version of Riley is preferred over the layer specific, as the latter may introduce a negative change in water content with increasing matric potential (h). Among the parameter PTFs, Wösten's continuous PTF showed the overall best performance, closely followed by Rawls&B and Vereecken. The ANN-based continuous PTF of Schaap showed poorer performance than its regression based counterparts. Systematic errors related to both particle size and SOM caused the class PTFs to perform poorly; these PTFs do not use SOM as input, and are therefore inappropriate for soils in Norway, being highly variable in SOM. The PTF performance showed little difference between soil groups. Water contents in the dry range of the SWRC were generally better predicted than water contents in the wet range. Pedotransfer functions that included both SOM and measured bulk density as input, i.e. Wösten, Vereecken and Rawls&B, performed best in the wet range.  相似文献   

15.
16.
河北省土壤容重的传递函数研究   总被引:1,自引:1,他引:0  
由于非破坏性取样测定土壤容重是一项困难、昂贵、费时和不符合实际的工作。为克服这些困难,国外学者已经研究出多种与土壤有机质或有机碳含量和土壤质地有关的传递函数去估计其研究范围内的土壤容重。本文利用河北省第二次土壤普查资料中的土壤有机质和质地数据,对已有的13种土壤传递函数模拟的结果进行对比,确定最佳传递函数形式后,用SPSS统计分析软件进行回归分析,求得适宜河北省土壤容重的传递函数表达式。  相似文献   

17.
Background, Aims, and Scope  During the last decades, different methods have been developed to determine soil hydraulic properties in the field and laboratory. These methodologies are frequently time-consuming and/or expensive. An indirect method, named Pedotransfer Functions (PTFs), was developed to predict soil hydraulic properties using other easily measurable soil (physical and chemical) parameters. This work evaluates the use of the PTFs included in the Rosetta model (Schaap et al. 2001) and compares them with PTFs obtained specifically for soils under two different vegetation covers. Methods  Rosetta software includes two basic types of pedotransfer functions (Class PTF and Continuous PTF), allowing the estimation of van Genuchten water retention parameters using limited (textural classes only) or more extensive (texture, bulk density and one or two water retention measurements) input data. We obtained water retention curves from undisturbed samples using the ‘sand box’ method for potentials between saturation and 20 kPa, and the pressure membrane method for potentials between 100 and 1500 kPa. Physical properties of sampled soils were used as input variables for the Rosetta model and to determine site-specific PTFs. Results  The Rosetta model accurately predicts water content at field capacity, but clearly underestimates it at saturation. Poor agreement between observed and estimated values in terms of root mean square error were obtained for the Rosetta model in comparison with specific PTFs. Discrepancies between both methods are comparable to results obtained by other authors. Conclusions  Site-specific PTFs predicted the van Genuchten parameters better than Rosetta model. Pedotransfers functions have been a useful tool to solve the water retention capacity for soils located in the southern Pyrenees, where the fine particle size and organic matter content are higher. The Rosetta model showed good predictions for the curve parameters, even though the uncertainty of the data predicted was higher than for the site-specific PTFs. Recommendations and Perspectives  The Rosetta model accurately predicts the retention curve parameters when the use is related with wide soil types; nevertheless, if we want to obtain good predictors using a homogenous soil database, specific PTFs are required. ESS-Submission Editor: Prof. Zhihong Xu, PhD (zhihong.xu@griffith.edu.au)  相似文献   

18.
Abstract. Water retention properties of 219 horizons were measured in Cambisols, Luvisols and Fluvisols, mainly from the Paris basin. We derived class pedotransfer functions (class PTFs) based on texture alone and in a second stage class PTFs based on classes combining texture and clod bulk density. The performance of these two types of PTFs were discussed at −330 and −15000 hPa water potential on an independent set of 221 horizons. Results showed that PTFs based on sets grouped by texture and clod bulk density provide estimates with an accuracy that is (i) greater than with class PTFs based on texture alone, and (ii) similar to the estimation accuracy recorded with continuous PTFs. As a consequence, the lack of interest in class PTFs should be reconsidered to bridge the gap between the available basic soil data and hydraulic properties which are generally missing, particularly when pertinent soil characteristics can be derived from the data available in soil databases. The two types of class PTFs providing gravimetric water contents at seven water potentials ranging from −10 to −15 000 hPa were converted to volumetric water content using the soil bulk density. Finally, the parameters of van Genuchten's water retention curve model were computed for every class PTF.  相似文献   

19.
灰色关联及非线性规划法构建传递函数估算黑土水力参数   总被引:2,自引:2,他引:0  
土壤水分特征曲线和饱和导水率是重要的水力参数,为了简便准确获取这些参数,以松嫩平原黑土区南部为研究区域,采集136个采样点土样用于测定不同土层土壤水分特征曲线、饱和导水率以及土壤理化性质,并运用灰色关联分析确定影响土壤水力参数的主要土壤理化性质,采用非线性规划构建土壤分形维数、有机质、干容重、土壤颗粒组成与土壤水分特征曲线、饱和导水率之间的土壤传递函数,并通过与现有土壤传递函数对比分析进行精度验证。结果表明:1)土壤分形维数是估算土壤水分特征曲线模型参数和饱和导水率的主要参数之一,同时,干容重和有机质含量也在不同土层土壤传递函数中起到重要的作用;2)通过验证分析,不同土层各参数平均绝对误差接近于0,均方根误差值也都较小,其中在不同土层土壤传递函数估算的土壤含水率均方根误差分别为0.022、0.017cm~3/cm~3;3)对比分析其他已存的土壤水分特征曲线和饱和导水率的土壤传递函数,该文构建的土壤传递函数均方根误差值均较小,决定系数值都在0.66以上,表明估算精度较高,均好于其他方法估算精度,具有良好的区域适应性。综上,所构建的土壤水分特征曲线和饱和导水率土壤传递函数可以用于松嫩平原黑土区土壤水力参数估算。  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号