非线性部分最小二乘方法的研究与应用

该文将部分最小二乘算法与径向基函数神经网络相结合,给出了一种非线性部分最小二乘建模方法,可以更加有效地处理过程非线性和数据共线性等复杂特性,提高模型的精度和推广能力。该方法在确定径向基函数神经网络的隐节点时,采用了一种改进型的k-均值聚类算法来自动确定最优的聚类区数。对煤气化炉合成气组分浓度软测量建模的应用结果表明,采用该方法建立的软测量模型在模型精度和推广能力等方面明显优于二次型多项式部分最小二乘方法建立的模型,并且计算精度满足工业生产的实际要求。

Least Squares and Its Applications

In this paper,a nonlinear partial least squares(NLPLS) is proposed by combining partial least squares(PLS) with RBF network.It is more effective to deal with complicated processes with nonlinearity and data collinearity.For obtaining the Gaussian centers in RBF network,an improved k-means clustering algorithm is developed to fine the optimal clusters automatically.The proposed method is applied to the actual plant to estimate the generated gas compositions of coal gasifier.It is shown that the accuracy and ...