摘要
为了实现小班逐年数据精准预测,使用LM-BP神经网络进行训练仿真,并用传统建模法与生长预测模型进行对比。结果表明:利用LM-BP神经网络法进行小班生长量模型训练精度高于传统方法,其训练方法简便易实施,可应用于森林资源调查工作中。
In order to achieve the subcompartment data accurate prediction year by year,the use of LM- BP neural network for training simulation,and comparing with traditional modeling method and growth forecast model,the results show that using LM- BP neural network training on subcompartment growth model has high precision than the traditional method,its training method is simple and easy to implement,so can be applied to forest resources investigation.
出处
《林业勘查设计》
2016年第1期85-87,共3页
Forest Investigation Design