基于插值与神经网络的包装薄膜热封参数分析

目的研究塑料薄膜热封工艺中热封参数之间的非线性关系,建立一种可用于自动包装机热封过程的数学模型。方法通过实验采集样本数据,并用附加动量法训练BP神经网络,建立热封参数之间的非线性数学模型,最后通过神经网络预测热封时间,并采用插值算法建立目标热封强度下热封温度和热封时间之间的多项式数学模型。结果通过插值算法与神经网络的结合运用,较为精确地描述了热封温度和热封时间之间的数学关系,插值函数实现了神经网络模型的简化,两者误差较小。结论通过文中方法确定了包装材料热封参数之间的非线性关系,将其用于热封包装设备,可提高设备的智能化程度。

Heat Sealing Parameters of Packaging Film Based on Interpolation and Neural Network

The work aims to study the nonlinear relationship between the heat sealing parameters in the heat sealing process of plastic film, and establish a mathematical model which can be used in the heat sealing process of automatic packaging machine. The BP neural network was trained by the additional momentum method under the sample data collected from experimental data to establish the nonlinear mathematical model between the heat sealing parameters. Finally, the neural network was used to predict the heat sealing time, and the interpolation algorithm was used to establish the polynomial mathematical model between the heat sealing temperature and the heat sealing time under the targeted heat sealing strength. Through the combination of interpolation algorithm and neural network, the mathematical relationship between heat sealing temperature and heat sealing time was described accurately; at the same time, the interpolation function realized the simplification of the neural network model, and the error of the both was small. The nonlinear relationship between the heat sealing parameters of packaging materials can be determined in the proposed method, and it can be applied to the heat sealing packaging equipment, which can improve the intelligence level of the equipment.