基于BP神经网络的西藏纯牧区草地生态承载力评价及预警  

Evaluation and early warning of grassland ecological carrying capacity based on BP neural network in pure pastoral area of Tibet

作　　者：王娜[1] 雒伟群[1] 周章金[1] 张燕华[1] WANG Na;LUO Wei-qun;ZHOU Zhang-jin;ZHANG Yan-hua(College of Information Engineering,Xizang Minzu University,Xianyang 712082,Shaanxi,China)

机构地区：[1]西藏民族大学信息工程学院,陕西咸阳712082

出　　处：《湖北农业科学》2022年第19期141-144,共4页Hubei Agricultural Sciences

基　　金：西藏自治区自然科学基金项目(XZ202001ZR0086G)。

摘　　要：以西藏15个纯牧业县(区)为研究对象,选取与草地生态承载力相关的14个指标,运用主成分分析进行指标精简,探讨影响草地生态承载力的主要因素;然后利用BP神经网络对纯牧区草地生态承载力进行仿真预测及预警。结果表明,色尼区、巴青县、嘉黎县和当雄县草地生态承载力超载严重,处于重警状态,聂荣县处于中警状态,安多县、申扎县、班戈县等7个县处于轻警状态,尼玛县、双湖县和革吉县处于安全状态。最后,针对不同预警状态,提出相关措施建议。Taking 15 pure pastoral husbandry counties(districts)in Tibet as the research objects,14 indexes related to grassland ecological carrying capacity were selected and simplified by principal component analysis to explore the main factors affecting grassland ecological carrying capacity.Then BP neural network was used to simulate and forecast the grassland ecological carrying capacity in the pure pastoral area.The results showed that the grassland ecological carrying capacity in Seni District,Baqing County,Jiali County and Dangxiong County was seriously overloaded,and was in the severe warning state.Nierong County was in the medium warning area.Seven counties,including Anduo County,Shenza County and Bange County,were in the light warning area.Nima County,Shuanghu County and Geji County were in the safe area.Finally,some suggestions were put forward for different early warning states.

关 键 词：西藏 纯牧区 生态承载力 BP神经网络 主成分分析 

分 类 号：S812[农业科学—草业科学]