基于KPCA-LSSVM的硅锰合金熔炼过程炉渣碱度预测研究

针对硅锰合金熔炼过程中炉渣碱度在线检测困难、离线化验滞后大,难以实现实时控制的问题,提出了一种基于核主元分析(KPCA)与最小二乘支持向量机(LSSVM)相结合的预测方法。该方法通过KPCA去除样本数据的噪声,提取输入数据空间中的非线性主元,然后利用LSSVM回归算法建立硅锰合金熔炼炉炉渣碱度预测模型,工业生产过程数据仿真结果表明,与SVM或LSSVM建模方法相比,KPCA-LSSVM预测模型的测量精度高、跟踪性能好,能满足炉渣碱度的在线测量要求。

KPCA and LSSVM model-based slag basicity prediction for silicomanganese smelting process

To overcome the difficulty that slag basicity can not be effectively controlled in silicomanganese smelting process due to lack of real-time on-line instrumentation,a predicting method is proposed by combining the Kernel Principal Component Analysis(KPCA)with the Least Square Support Vector Machine(LSSVM).Using KPCA,it is possible to denoise the input data and capture the high-ordered nonlinear principal components in input data space,and with LSSVM we can establish a prediction model between the featured principal components and the primary variable for the slag basicity in a silicomanganese furnace smelting process.Simulation results show that the KPCA-LSSVM model has higher accuracy and better tracking performance compared with SVM or LSSVM models,so the proposed method can satisfy the requirements of on-line measurement of slag basicity.