基于能量分布轧机振痕振动特征识别方法研究

在轧制过程中,带钢表面经常会出现明暗交替的振痕,严重影响带钢表质量;由于轧制速度等轧制参数波动会导致振痕特征频率的变化,难以准确捕捉振痕特征频率,无法实现振痕特征频率的自动识别和实时监测。分析了带钢表面振痕特征,根据轧制参数确定了带钢表面振痕出现的特征频率,提出了对振动加速度信号的自相关处理和小波包分解相结合的方法,得到了小波包分解能量分布情况,克服了轧制参数波动对振痕特征频率的影响,实现了对带钢表面振痕的有效识别。试验证明该方法不仅便于计算机自动识别,而且为提高生产效率和经济效益提供了新的理论依据和实践支撑。

Study on chatter marks vibration characteristics of rolling mill based on energy distribution

Light chatter marks alternating with dark chatter marks on the surface of steel strip in the rolling process may seriously affect the final product quality. It still difficult to identify and monitor the chatter marks frequency accurately in time because of the variations of the rolling parameters. In the paper, the calculation of chatter marks frequency is introduced based on the analysis of the characteristics of the chatter marks on the steel strip. The autocorrelation processing and the wavelet packet decomposition are integrated to obtain the wavelet packet decomposition energy distribution, which overcomes the influence of rolling parameters fluctuation on chatter marks frequency and realizes the effective indention of chatter marks. Experimental results show that the proposed method not only facilitates the computer automatic identification, but also provides a theoretical and practical support for improving the production efficiency and economic benefits.