欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对化工生产过程中质量指标无法在线监测的问题,多模型软测量建模方法往往能取得不错的模型估计精度。然而传统的模糊聚类算法都假定样本的各维特征对聚类的贡献相同,影响了聚类效果和模型估计精度。为了考虑样本各维特征对聚类的不同影响,提出一种新的特征加权模糊聚类算法。该算法在模糊聚类迭代的基础上,逐步调整特征权值,最终有效改善了聚类效果。利用一个实际生产装置的操作数据进行建模仿真实验,结果显示了该方法的优越性。
英文摘要：
As for the problem that quality index can＇t be monitored on line in chemical processing,the modeling method of multi-model soft-sensor can always get good estimated accuracy.However,the traditional fuzzy clustering assumes that each feature of the samples is the same to the contribution of clustering.In order to consider the particular contributions of different features,a feature-weighted fuzzy clustering algorithm is proposed.This algorithm adjusts feature weights gradually on the basis of the iterative fuzzzy clustering,improves the clustering result finally.The simulation experiments are carried out with the data from an industerial plant.The simulation result shows that the modeling method is superior.
同期刊论文项目
同项目期刊论文
Parameter estimation for the forgotten factor least squares algorithm for dual-rate time-varying sys
Exponential synchronization of a class of neural networks with mixed time-varying delays and impulsi
Robust stability of uncertain Markovian jumping Cohen-Grossberg neural networks with mixed time-vary
Novel global robust exponential stability criterion for uncertain BAM neural networks with time-vary
An auxiliary model based recursive least squares parameter estimation algorithm for non-uniformly sa
Delay-dependent robust stability for uncertain stochastic fuzzy Hopfield neural networks with time-v
期刊信息
位置：