摘要
板形模式识别是板形高精度控制过程中的技术难点之一。本文提出了一种基于小波分析和神经网络相结合的方法,可以较好地解决这个难题。首先利用非线性小波变换阈值法有效去除板形检测数据中噪声干扰,从而提高模式识别系统的准确率;然后利用神经网络的鲁棒性使目标的识别更加接近实际。该方法不仅能有效地对复合板形进行正确分类,而且能分辨出所属类型的程度,为制定出相应的优化控制策略提供了重要的依据。
Pattern recognition for flatness is one of the difficult techniques in flatness control system of high precision. A novel pattern recognition method based on wavelet analysis and neural networks is presented in this paper. In this method, the nonlinear wavelet threshold denoising method is used to filter the noise of the measure data availably. Then the filtered data are applied to recognize the flatness through the model based on Neural Networks. The combination of wavelet and neural networks can improve veracity and robustness ability in pattern recognition system for flatness. This method can classify the complex flatness correctly and recognize the extent of a certain classification. It is important to provide information for deciding the control strategy.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2005年第1期103-106,共4页
Pattern Recognition and Artificial Intelligence
基金
国家自然科学基金(No.60274024)
关键词
小波分析
神经网络
板形
模式识别
Wavelet Analysis
Neural Network
Flatness
Pattern Recognition