摘要
针对废杂铜冶炼过程炉膛温度难以准确测量的问题,提出一种基于加权LS-SVM的炉膛温度软测量方法。该方法通过对过程主要输入输出变量的误差平方赋予不同权重用于克服异常训练样本的影响,并利用粒子群算法对加权LS-SVM的参数进行寻优,增强了动态模型对非线性时变特性的适应能力,提高了温度预测的准确性。最后,通过对废杂铜冶炼过程的实际运行数据进行仿真研究,验证了方法的有效性。
Aimed to the difficult temperature measurement of scrap copper smelting process, this paper proposed a method of soft measurement of furnace temperature based on weighted least squares support vector machine (WLS-SVM). In this method, the main input and output variables of the process squared error is given different weights to overcome the impact of the training sample anomalies, and use PSO for WLS-SVM parameters optimization, enhanced ability to adapt of dynamic model for the nonlinear time-varying characteristics, improved the prediction accuracy of the model. Finally, simulated through actual operating data of scrap copper smelting process, and verified the effectiveness of the method.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2014年第12期1457-1460,共4页
Computers and Applied Chemistry
基金
国家863计划资助项目(2009AA04Z154)
国家青年基金资助项目(61304081)
关键词
废杂铜
最小二乘
支持向量机
软测量
scrap copper smelting
least squares
support vector machine
soft measurement