基于BP神经网络的中药饮片生产风险预测
Risk prediction of traditional Chinese medicine decoction pieces production based on BP neural network
娄黎明 1白莹1
作者信息
- 1. 北京信息科技大学经济管理学院 北京 100192
- 折叠
摘要
本文对中药饮片生产风险问题进行深入研究,采用BP神经网络构建风险预测模型.通过调查大量生产数据,选取关键变量,模型在隐藏层节点数为 14 时表现出色,迭代 11 次后误差值显著降低,相关系数高达 0.940 35,展现出高拟合和泛化能力.相较于传统方法,该模型能更早发现风险,为中药饮片生产风险预测提供新方法,有助于提升生产效率、降低成本、减少企业损失,对中药饮片行业的安全生产具有重要意义.
Abstract
This article conducts an in-depth study on the risk issues in the production of traditional Chinese medicine decoction pieces,and utilizes BP neural networks to construct a risk prediction model.Through investigating a large amount of production data and selecting key variables,the model has demonstrated excellent performance when the number of hidden layer nodes is set to 14.After 11 iterations,the error value has significantly decreased,and the correlation coefficient is as high as 0.940 35,showing high fitting and generalization capabilities.Compared with traditional methods,this model can detect risks earlier,providing a new approach for risk prediction in the production of traditional Chinese medicine decoction pieces.It helps to improve production efficiency,reduce costs and minimize losses for enterprises,which is of great significance to the safe production of the traditional Chinese medicine decoction piece industry.
关键词
中药饮片/生产风险/BP神经网络/风险预测Key words
Chinese medicine decoction pieces/Production risk/BP neural network/Risk prediction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据