论文部分内容阅读
针对非线性、高维工业过程,提出一种基于核主成分分析(KPCA)与正交最小二乘(OLS)的软仪表建模方法。该方法首先采用KPCA技术对,在特征空间中对高维输入数据进行降维处理,消除噪声等不利因素的影响;然后采用OLS处理输入输出之间的非线性关系,在最大化泛化能力的同时,实现模型的稀疏性。将此软仪表模型应用于柴油凝点的预报,结果表明,较其他方法,提出的方法有较好的泛化能力及稀疏性。
Aimed at nonlinear and high dimensional industrial processes, a method of modeling soft instruments based on kernel principal component analysis (KPCA) and orthogonal least squares (OLS) is proposed. Firstly, KPCA technique is used to reduce the dimensionality of high-dimensional input data in feature space to eliminate the influence of noise and other unfavorable factors. Then, OLS is used to deal with the nonlinear relationship between input and output, At the same time, the sparsity of the model is realized. The soft instrument model is applied to forecast the pour point of diesel. The results show that the proposed method has better generalization ability and sparsity than other methods.