基于全局优化支持向量机的多类别高炉故障诊断

发布时间：2018-01-28 23:51

本文关键词： 高炉故障诊断最小二乘分析支持向量机全局优化　出处：《工程科学学报》2017年01期 　论文类型：期刊论文

【摘要】：针对高炉故障诊断系统快速性和准确性的要求,提出基于全局优化最小二乘支持向量机的策略.首先,采用变尺度离散粒子群对最小二乘支持向量机的参数和故障特征的选取进行优化;然后,利用核主元分析法对选取的特征向量进行压缩整理;最后,构造了以Fisher线性判别率为标准的启发式纠错输出编码.仿真结果表明,通过对故障训练样本有意义地分割重组,用较少的最小二乘支持向量机分类器,得到较高的故障判断准确率且增强了整个系统的实时性.
[Abstract]:Aiming at the requirement of the speed and accuracy of blast furnace fault diagnosis system, a strategy based on global optimization least squares support vector machine (LS-SVM) is proposed. Variable scale discrete particle swarm optimization is used to optimize the parameters and fault features of the least squares support vector machine (LS-SVM). Then, the kernel principal component analysis is used to compress the selected feature vectors. Finally, a heuristic error correction output code based on Fisher linear discriminant rate is constructed. The simulation results show that the fault training samples are segmented and reorganized meaningfully. By using less least squares support vector machine classifier, the accuracy of fault diagnosis is higher and the real-time performance of the whole system is enhanced.
【作者单位】：北京科技大学自动化学院;北京科技大学钢铁流程先进控制教育部重点实验室;
【分类号】：TF549
【正文快照】： 钢铁工业是我国国民经济的基础产业和支柱产业[1].高炉炼铁在钢铁工业中处于举足轻重的地位.高炉炼铁系统生产设备繁多,具有多耦合、大延时、非线性等特点.尽管在高炉本体上安装了很多自动化的检测装置,然而由于高炉运行炉况复杂,无法建立准确的机理模型,在高炉自动控制决策过

【相似文献】