区域农作制优先序是区域农业结构调整的主要依据。本研究通过对西北旱作农区农业自然资源、农田生态环境和农村经济发展水平等现状的调研,总结出西北旱作农区主要的7种农作制度。并应用BP神经网络模型,以随机技术模拟生成的评价指标序列与其所属的评价等级值进行网络训练。网络训练后,将不同农作制度的评价指标为网络输入,通过计算可以得到相应的农作制优先序的评价等级值。结果表明,西北旱作农区农作制的优先序为:粮草果畜复合制〉粮棉(油)轮作制〉特色产业(苹果)型农作制〉抗旱节水型农作制〉小杂粮、草畜主导型农作制〉两粮一肥型农作制〉粮油豆草肥田制。本方法只要给定相应的评价指标值,通过BP神经网络模型的计算,可直接得出农作制度优先序的评价等级值,可以用于农作制度优先序的评价。
The research of regional farming system's priority sequence is the main basis of regional agricultural structural adjustment.This study is based on the investigation of the status quo of agricultural natural resources,agro-ecological environment and the level of economic development in rural areas of dryland farming areas in Northwest China,and summs up seven kinds of major farming systems of dryland area in northwest China.BP neural network model and network training is applied in evaluation index generated by sequence of random technology and evaluation of their own level of value.After the network training,different evaluation of farming systems are input.By calculating,we can get the evaluation of grade value of farming system priority.The results show that: the farming system priority of dryland farming area in Northwest China is a complex farming system of animal food fruit cotton(oil) rotation farming system farming system of industry characteristics(apple)-type drought water-saving farming system small grains,grass-animal based farming system two grains and a pasture farming system farming system of grain and oil beans grass fertilizer.As long as given values of the corresponding evaluation indexes,by the calculation of BP neural network model,the grade value of evaluation of farming system priority can be directly got.This model can be used for the evaluation of farming system priority.