基于粗糙集神经网络的刀具磨损监测的研究

Fault Diagnosis of Tool Wear Based on Rough Set and Neural Network

针对多传感器刀具磨损监测系统输入维数较多、神经网络结构复杂、收敛速度慢等缺点,提出了粗糙集和遗传算法优化神经网络的模型。该模型首先利用粗糙集理论的属性约简对输入数据进行处理,从而达到减少神经网络输入维数、简化神经网络结构的目的。然后通过遗传算法优化神经网络的初始权值和阈值,以提高神经网络的收敛速度,避免神经网络陷入局部极值点。将该模型应用到刀具磨损监测,通过对声发射信号和电流信号进行处理,提取特征向量值,将特征值先通过自组织神经网络进行连续属性离散化,再通过粗糙集理论进行属性约简,最后通过遗传算法优化的BP神经网络进行识别,取得了很好的效果,证明了此模型的有效性和可行性。

To overcome the problem of structure complexity and long training time in neural network method for fault diagnosis of tool wear with multi-sensor, a new fault diagnosis method based on rough set and neural network is presented. At first, the rough set was used to select the influencing factors input into the neutral network. Then, the genetic algorithm is used to overcome the shortcoming of the BP algorithm, such as slowness and converge,we to local minimum. The model is applied to tool diagnosis, the self-organizing map method is used to get the discrete attributes fist, then an adaptive genetic algorithm is devised for attribute reduction, and finally the results of the attribute reduction is regard as the inputs of the neural network. The experimental results show that it is feasible and eflective in the tool diagnosis.