On the basis of Artificial Neural Network theory, a back propagation neural network with one middle layer is building in this paper, and its algorithms is also given, Using this BP network model, study the case of Mal...On the basis of Artificial Neural Network theory, a back propagation neural network with one middle layer is building in this paper, and its algorithms is also given, Using this BP network model, study the case of Malian-River basin. The results by calculating show that the solution based on BP algorithms are consis- tent with those based multiple - variables linear regression model. They also indicate that BP model in this paper is reasonable and BP algorithms are feasible.展开更多
The nomograph by Wischmeier et al.(1971)for calculating the K-factor in the USLE was extremely useful when there was low access to calculators.However,the generalised calculation of this factor requires the developmen...The nomograph by Wischmeier et al.(1971)for calculating the K-factor in the USLE was extremely useful when there was low access to calculators.However,the generalised calculation of this factor requires the development of analytic procedures.This paper presents a detailed analysis of the nomograph and its underlying equation,which is applicable only when the silt plus very find sand fraction does not exceed 70%.We also examined the quality of fit on the nomograph of the adaptations to the equation that have been proposed,as a means of dealing with those areas where the original equation is not applicable.All models are shown to have areas where the fit is deficient or even unacceptable.Besides,the family of curves on the nomograph for the various values taken by the organic matter are not coincident with the mathematical function from which they presumably derive.The study also identifies those areas of the textural triangle in which the soils originally used in developing the USLE are located,with a view to according a lower predictive value to the contrasting areas in which calculations of the K-factor will necessarily be extrapolations.Finally,a new equation for calculating the K-factor is presented,which accurately reproduces the different sections of the nomograph,and allows the poorly functioning graph to be dispensed with.The paper ends with a link to a tool in R for simplifying the procedure for calculating the K-factor,taking into account varying situations of data availability.展开更多
基金Supported by Brilliant Youth Fund in Hebei Province
文摘On the basis of Artificial Neural Network theory, a back propagation neural network with one middle layer is building in this paper, and its algorithms is also given, Using this BP network model, study the case of Malian-River basin. The results by calculating show that the solution based on BP algorithms are consis- tent with those based multiple - variables linear regression model. They also indicate that BP model in this paper is reasonable and BP algorithms are feasible.
文摘The nomograph by Wischmeier et al.(1971)for calculating the K-factor in the USLE was extremely useful when there was low access to calculators.However,the generalised calculation of this factor requires the development of analytic procedures.This paper presents a detailed analysis of the nomograph and its underlying equation,which is applicable only when the silt plus very find sand fraction does not exceed 70%.We also examined the quality of fit on the nomograph of the adaptations to the equation that have been proposed,as a means of dealing with those areas where the original equation is not applicable.All models are shown to have areas where the fit is deficient or even unacceptable.Besides,the family of curves on the nomograph for the various values taken by the organic matter are not coincident with the mathematical function from which they presumably derive.The study also identifies those areas of the textural triangle in which the soils originally used in developing the USLE are located,with a view to according a lower predictive value to the contrasting areas in which calculations of the K-factor will necessarily be extrapolations.Finally,a new equation for calculating the K-factor is presented,which accurately reproduces the different sections of the nomograph,and allows the poorly functioning graph to be dispensed with.The paper ends with a link to a tool in R for simplifying the procedure for calculating the K-factor,taking into account varying situations of data availability.
