基于数据降噪的冷轧板形调控功效系数获取

 针对轧制过程实际数据噪声大、难以获取准确板形调控功效系数的问题,提出了一种融合集成经验模态分解(EEMD)和小波变换(WT)的数据降噪方法。将含有噪声的实际生产数据经过EEMD分解后,利用小波变换方法对噪声主导的本征模态分量(IMF)进行降噪处理,处理后的噪声分量与其余分量重构得到降噪后数据,并结合结构方程模型(SEM)计算得到板形功效系数。利用1 450 mm五机架冷连轧生产线实际数据进行试验,结果表明,EEMD-WT-SEM方法可以有效降低数据噪声,有效提升板形调控功效系数的准确性。

Cold-rolled flatness actuator efficiency coefficient obtaining based on data noise reduction

Aiming at the problem that it is difficult to obtain accurate flatness control efficiency coefficients because of the large noise in the actual data of the rolling process,a data noise reduction method combining ensemble empirical modal decomposition(EEMD) and wavelet transform(WT) is presented. Firstly,EEMD is used to decompose the original data and the intrinsic mode function(IMF) divided into two parts, the noise dominant components and the signal dominant components. Then,the wavelet transform denoising method is used to conduct noise reduction of the noise dominant components. Finally,the flatness actuator efficiency coefficients are calculated from the reconstructed data by the structural equation model(SEM) method. The actual data of a 1 450 mm five-stand cold tandem rolling production line was used for testing,the results show that the EEMD-WT-SEM method can effectively reduce the data noise and improve the accuracy of the efficiency coefficient.