摘要
利用小波包分析与BP（Back Propagation）神经网络相结合的算法，对氧化铝熟料检测的应用进行了研究。通过对回转窑中的氧化铝熟料下落碰撞窑壁产生的声音信号进行采集，利用小波包分析提取特征向量，根据氧化铝的烧结状况与声音信号特征向量的对应关系，提出建立BP神经网络模型。经过MATLAB对测试样本进行验证，结果表明BP神经网络模型在氧化铝熟料检测的可行性，而且具备一定的准确率。
This paper has a research on the application of wavelet packet analysis combined with BP neural network in alumina clinkers detection. The method includes gathering the sound signal produced by alumina clinkers collide the rotary kiln wall and extracting feature vectors using wavelet packet analysis from sound signal. According to corresponding relationship between the alumina clinkers sintered condition and the sound signal feature vectors, BP（Back Propagation）neural network model is presented. After MATLAB simulation, test results show that BP（Back Propagation）neural network model is a feasible plan in alumina clinkers detection, but also has high accuracy rate.
出处
《电子测试》
2013年第10期36-39,共4页
Electronic Test
关键词
小波分析
BP网络
特征向量
熟料检测
Wave]et Analysis
BP network
Feature vectors
Clinker detection
