摘要
利用羧甲基纤维素钠和甘油为主要原料配制的水溶性胶黏着剂制作3种颜色粘板,在露天菜地的同一处地点进行田间蓟马诱捕测试。将测试结果作为BP神经网络模型的训练样本与预测样本进行建模分析,结果表明,黏合剂中诱虫物质的添加因素对葱蓟马田间诱捕量的影响权重最大,蓝色大型粘板诱捕蓟马效果最好,模型对预测样本的预测值与实际观测值基本吻合。
The sticky plates of three colors were made out by water-soluble adhesive prepared with carboxymethylcellulose sodium and glycerine as main raw materials, and the field experiment of trapping thrips was carried out in one place of a vegetable garden by different adhesives. Finally the data of this experiment as the training samples and the forecasting samples were analyzed by using BP neural network model. The results showed that the attrahent added had the most weight to the number of entrapped Thrips alliorum on the field, and the forecast results of model were consistent with the actual results perfectly. Generally, the bigger blue sticky plates had the best effect of trapping Thrips alliorum. The predictive value of samples by using this model was accordant with the observed value.
出处
《安徽农业科学》
CAS
北大核心
2009年第31期15294-15296,共3页
Journal of Anhui Agricultural Sciences
基金
河南科技学院重点资助基金项目(0500136)
关键词
BP神经网络
水溶黏着剂
葱蓟马
BP neural network
Water-soluble adhesive
Thrips alliorum
