基于主元分析和RBF神经网络的火灾模拟实验炉温软测量

Temperature Soft Sensor of Fire Simulated Experiment Furnace Based on Principal Component Analysis and RBF Neural Network

软测量也称为软仪表技术,采用主元分析和RBF神经网络相结合的融合模型构成火灾模拟实验炉温软测量。主元分析(PCA)实现输入变量的降维,RBF神经网络采用K-均值聚类算法进行隐层中心和连接权调节的学习,实现快速收敛。该融合模型使炉温估计精度比常规的最小二乘方法拟合精度提高2倍以上。

Soft sensor is also called soft measurement. Temperature soft sensor of fire simulated experiment furnace is constructed based on a data fusion model with a method combining principal component analysis and RBF neural network. Principle component analysis can reduce the dimension of input variable. RBF neural network may achieve fast convergence of learn algorithm by adjusting the center of hide layer and the connective weigh with K-mean cluster algorithm. This model improves the precision of furnace temperature two times more than least square method.