高温高压蒸汽改性落叶松木材力学性能预测模型的建立

摘要

We used RBF neural network and support vector machine to study the relationship between Larix gmelini modification process parameters and its mechanical properties of high temperature and pressurized Steam .With the heat treatment tem-perature, relative humidity, processing time as network input , we established the RBF neural network and support vector machine forecasting model , and compared the two models .The support vector machine ( SVM) model in network structure , convergence speed and generalization ability has great significance .%利用RBF神经网络和支持向量机两种算法建模，分析落叶松高温高压蒸汽改性工艺参数与其力学性能关系；以落叶松热处理的温度、相对湿度、处理时间3个主要工艺参数作为网络输入，建立了RBF神经网络和支持向量机预测模型，并对两者进行比较。结果表明：支持向量机模型，在网络建立结构、收敛速度和泛化能力上更具优势。