煤调湿系统蒸汽消耗量的多模型建模研究  

Research on the Multi-model Modeling for Vapor Consumption of Coal Moisture Control System

作　　者：王建华[1] 罗雷涛 耿佳灿 

机构地区：[1]上海应用技术学院电气与电子工程学院,上海201418 [2]华东理工大学化工过程先进控制与优化技术教育部重点实验室,上海200237

出　　处：《自动化仪表》2014年第12期88-93,共6页Process Automation Instrumentation

摘　　要：煤调湿工艺是炼焦过程的关键技术,建立煤调湿系统的蒸汽消耗量软测量模型对节约生产成本、减少环境污染和提高焦炭质量具有促进作用。针对某炼铁厂煤调湿装置的非线性、强耦合、工况波动大等特点,提出了一种多模型建模方法。对模糊核C均值聚类引入了基于密度的聚类中心初始化方法和聚类数目自适应策略,并将其用于生产工况最优划分;使用最小二乘支持向量机对每个子工况进行数据驱动建模,并利用贝叶斯证据框架优化最小二乘支持向量机的超参数。检验结果表明,所提出的煤调湿系统蒸汽消耗量多模型具有良好的跟踪性能与较高的预测精度。Coal moisture control technology is the key technique of the coking process. Establishing the soft sensing model for vapor consumption in coal moisture control system may save production costs, reduce environmental pollution and improve the quality of coke. In accordance with the features of the coal moisture control equipment in certain ironworks, e. g. , nonlinearity, strong coupling and large fluctuations of working condition, the multi-model modeling method is proposed. The density-based clustering center initialization method and number of clusters adaptive strategy are integrated into the fuzzy C-means clustering to get optimal division of the production working conditions. The least squares support vector machines （ LSSVM ） is applied to conduct data-driven modeling for each sub-working conditions, and the parameters of the LSSVM is optimized by adopting Bayesian evidence framework. The checked results indicate that the proposed multi-model possesses good tracking performance and higher prediction accuracy.

关 键 词：煤调湿 模糊核C均值聚类 最小二乘支持向量机(LSSVM) 贝叶斯证据框架 多模型 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]