论文部分内容阅读
由于以快速傅立叶变换为基础的轧辊偏心补偿控制算法具有时域和频域局部化矛盾的局限性 ,且不能处理非平稳信号 ,本文利用在小波变换下奇异信号和随机噪声在多尺度空间中的模极大值传递特性的不同 ,提出了一种基于小波消噪的轧辊偏心补偿控制算法 ,并设计了一种轧辊偏心的小波分析器 (WARE)。仿真研究表明 ,对于提高轧件的厚度精度 WARE优于轧辊偏心的傅立叶分析器 (FARE)和其改进算法 (MFFT和 L S算法 ) ,并可使轧辊偏心对厚度的影响减小 90 %。通过对基于小波分析的轧辊偏心补偿控制的探索 ,为开发高精度轧机控制系统指出了一个新的研究方向。它对设计新型轧机和改造旧轧机都具有现实意义 ,有利于促进轧制过程自动化的发展
Because of the limitations of time-domain and frequency-domain localization, the roll eccentricity compensation control algorithm based on Fast Fourier Transform can not deal with the non-stationary signal. In this paper, the singular signal and random noise under multi-scale space Based on the difference of maximum transmission modes, a roll compensation algorithm based on wavelet denoising is proposed. An eccentric wavelet analyzer (WARE) is designed. Simulation studies have shown that reducing the influence of roll eccentricity on the thickness by 90% can improve the precision of WARE over roll eccentricity FARE and its improved algorithms (MFFT and LS algorithms). Through the exploration of roller eccentricity compensation control based on wavelet analysis, a new research direction for developing high precision rolling mill control system is pointed out. It is of practical significance to the design of new rolling mills and the transformation of the old rolling mills, which contributes to the automation of the rolling process