Abstract:

Due to the complex and changeable working conditions in chemical production process, a single soft sensor model cannot meet the requirements of the system for estimation accuracy. A new method of multi-manifold modeling is proposed in this paper based on a modified expanding search clustering algorithm. This algorithm uses the distance between manifolds instead of the Euclidean distance, adaptively determines the neighborhood radius, and introduces the local density to determine the center of clustering. The features of sub-manifold obtained after clustering are extracted by kernel isometric mapping method in manifold leaning respectively, and develop sub-models based on Gaussian process regression. The method was applied to the soft measurement modeling of a bisphenol A production device. The simulation results verify the effectiveness of the method.

Key words: manifold learning, algorithm, model, soft sensor, expanding search clustering, computer simulation

摘要：

针对化工生产过程工况复杂多变，单一的软测量模型难以满足系统对估计精度的要求，提出了一种基于改进的扩张搜索聚类算法的多流形软测量建模的方法。该算法采用流形距离来代替欧氏距离，自适应地确定邻域半径，并引入局部密度用于确定聚类中心，对聚类后得到的各个子流形分别采用流形学习中的核等距映射法进行特征提取，建立基于高斯过程回归的子模型。将该方法应用于某双酚A生产装置的软测量建模，仿真结果验证了该方法的有效性。

关键词: 流形学习, 算法, 模型, 软测量, 扩张搜索聚类, 计算机模拟