ANN在聚丙烯酸酯乳液性质预测中的应用

在MATLAB中的图形用户界面下,用人工神经网络(ANN)对聚丙烯酸酯乳液的硬度、附着力和耐冲击性3种性能进行了预测。选用三层的误差反向传递网络(BP网络),讨论了隐含层节点数,训练目标goal值和传递函数等相关参数对预测结果的影响。在隐含层节点数分别为19、16和20,传递函数为logsig函数和purelin函数,训练目标值goal为1.0×10-5的条件下,对17个样品进行了预测。结果表明:硬度预测值与实验值相对误差的绝对值的平均值为5.90%,附着力预测准确率为100%,耐冲击性预测准确率为100%。

Application of ANN in property prediction of polyacrylate emulsion

Based on the graphical user interfaces in MATLAB, the artificial neural network was used to predict the hardness,adhesion and impact resistance of the polyacrylate emulsion. A back propagation network with three layers was selected to discuss the effects of the number of the hidden layer node, the value of the training goal and transfer functions on the prediction results. When the number of the hidden layer node was 19, 16 and 20, respectively, transfer functions were logsig function and purelin function, and the value of training goal was 1.0×10-5, 17 samples were predicted. The results show that the average absolute value of relative errors between predictive and measured values of hardness is 5.90%, the prediction accuracies of adhesion and impact resistance are 100%.