煤矿旋转机电设备的量子神经网络故障诊断技术

Quantum neural network fault diagnosis technology for coal mine rotating electromechanical equipment

针对煤矿旋转机电设备故障模式相互干扰的问题,基于量子神经网络理论,提出了一种量子神经网络故障诊断算法。以量子学中的相移门和受控非门为基本计算单元,构造出3层量子神经网络故障诊断模型,采用梯度下降法作为该模型的学习算法,对刮板输送机减速器的多种故障进行识别诊断。初步研究结果表明,所提出的量子神经网络故障诊断技术是可行的,有助于提高煤矿旋转机电设备的故障诊断率。

In view of problem of mutual interference of failure mode for rotating electromechanicalequipment in coal mine, a quantum neural network fault diagnosis algorithm was proposed based on quantum neural network theory, a quantum neural network fault diagnosis model with three-layer was constructed by using the phase-shift gate and controlled-not gate of quantum theory. A gradient descent algorithm was taken as learning algorithm for the model which was applied to recognize the fault diagnosis of scraper conveyor reducer. The preliminary research results show that the algorithm is feasible and helpful to improve fault diagnosis rate of rotating electromechanical equipment used in coal mine.