摘要
Research work on rural landscape planning was conducted at Nanhua State Farm, which is located in the south of the Leizhou Peninsula, Guangdong Province. Based on its soil nutrient status, rainfall and mean yield of major crops, the soil suitability and land adaptability were evaluated. An artificial neural network was introduced as an alternative for modeling the landscape planning. A model was developed using back propagation as the learning procedure, the sigmoid function as the transfer function and 6 patch types and its number of three phases as input factors. Three optimum planning schemes were selected by using the model.Strategies were proposed for improving diversity and heterogeneity, productivity and sustainability within the system through the planning scheme.
Research work on rural landscape planning was conducted at Nanhua State Farm, which is located in the south of the Leizhou Peninsula, Guangdong Province. Based on its soil nutrient status, rainfall and mean yield of major crops, the soil suitability and land adaptability were evaluated. An artificial neural network was introduced as an alternative for modeling the landscape planning. A model was developed using back propagation as the learning procedure, the sigmoid function as the transfer function and 6 patch types and its number of three phases as input factors. Three optimum planning schemes were selected by using the model.Strategies were proposed for improving diversity and heterogeneity, productivity and sustainability within the system through the planning scheme.
