The features are inputted into BP neural network to diagnose the type of working conditions.The trained network can be used to monitor and diagnos the aluminum reduction cells on line.

• 利用小波包多维多分辨率特性，对铝电解槽电压波动信号进行分解与重构，实现与工作状态干扰因素相关的特征信息的提取，将其作为特征向量输入BP神经网络，对其进行训练。