论文部分内容阅读
由于基于频域的经典小波变换运算时间较长,不能很好地满足轧辊偏心信号在线实时控制的要求,提出了用提升结构小波变换对偏心信号进行不同分辨率下分解处理的新方法.通过对轧制力信号和厚差信号的分析,利用提升和对偶提升原理将偏心信号从干扰信号和噪声信号中提取出来并通过参数自校正控制实现对轧辊偏心的在线动态控制.仿真结果表明,该方法获得了比较理想的效果,并且在同样数据长度下,提升小波变换运算速度比经典小波变换至少提高1倍以上.
Due to the time-consuming classical wavelet transform based on frequency domain and the in-line real-time control of roll eccentricity, which can not meet the requirements of on-line real-time control of roller eccentricity, a new method of decomposing eccentric signals with different resolutions is proposed. Rolling force signal and thickness difference signal, the eccentric signal is extracted from the interference signal and the noise signal by lifting and dual lifting principle and the on-line dynamic control of roll eccentricity is realized by the parameter self-tuning control.The simulation results show that this method Obtained the ideal result, and under the same data length, enhance the speed of wavelet transform operation than the classical wavelet transform at least doubled.