华南理工大学学报(自然科学版) ›› 2009, Vol. 37 ›› Issue (2): 158-162.

• 动力与电气工程 • 上一篇    

基于LSSVM—MODE的水煤浆生产优化控制

刘定平叶向荣邓华裕2   

  1. 1. 华南理工大学 电力学院, 广东 广州 510640; 2. 茂名热电厂, 广东 茂名 525011
  • 收稿日期:2007-10-30 修回日期:2007-11-19 出版日期:2009-02-25 发布日期:2009-02-25
  • 通信作者: 刘定平(1965-),男,副教授,博士生,主要从事燃烧优化与控制研究. E-mail:liudingping@126.com
  • 作者简介:刘定平(1965-),男,副教授,博士生,主要从事燃烧优化与控制研究.

Optimization Control of Preparation of Coal Water Mixture Based on LSSVM-MODE

Liu Ding-ping1  Ye Xiang-rong1  Deng Hua-yu2   

  1. 1, School of Electric Power, South China University of Technology, Guangzhou 510640, Guangdong, 2. Maoming Thermal Power Plant, Maoming 525011, Guangdong, China
  • Received:2007-10-30 Revised:2007-11-19 Online:2009-02-25 Published:2009-02-25
  • Contact: 刘定平(1965-),男,副教授,博士生,主要从事燃烧优化与控制研究. E-mail:liudingping@126.com
  • About author:刘定平(1965-),男,副教授,博士生,主要从事燃烧优化与控制研究.

摘要: 水煤浆(CWM)制造过程中,生产成本的降低和水煤浆性能的提高之间存在着矛盾.文中利用最小二乘支持向量机(LSSVM)对球磨机电流和水煤浆浓度进行多目标建模,并采用基于Pareto最优概念的多目标微分进化(MODE)算法对运行工况进行寻优,然后根据模糊集理论在Pareto解集中求得满意解,获得了水煤浆浓度的优化调整方式和提高水煤浆生产效益的策略.

关键词: 水煤浆, 优化运行, 最小二乘支持向量机, 多目标微分进化算法

Abstract:

In the production of coal water mixture ( CWM), there exists an inconsistency between the production cost and the product performance. In order to solve this problem, the least-square support vector machine is em- ployed to establish a muhi-objective optimization model for CWM concentration and ball mill current, and the multi- objective differential evolution algorithm based on Pareto optimal concept is used to optimize the operation condi- tions, Moreover, the fuzzy set theory is introduced to obtain the satisfactory solutions in Pareto solution set. An op- timized adjustment mode of CWM concentration and some strategies to improve the CWM production benefit are fi- nally proposed in the paper.

Key words: coal water mixture, optimal operation, least-square support vector machine, multi-objective differential evolution algorithm