论文部分内容阅读
针对线控转向汽车的可靠性和安全性以及故障诊断方法的不足,提出了一种基于软计算的汽车线控转向故障诊断方法,该方法利用软计算中的粗糙集和粒子群优化的径向基神经网络进行结合。将粗糙集作为径向基神经网络的输入处理,对样本数据进行属性约简,约简后的属性集作为径向基神经网络的输入以达到缩短网络训练时间的目的。采用粒子群算法对径向基神经网络的基函数中心值和宽度进行编码和寻优,并使用得到的最优中心值和宽度组建径向基神经网络,使得径向基神经网络的样本训练误差相比未优化之前有一定程度的降低。然后使用训练好的神经网络对故障样本进行测试,测试结果表明,该方法加快了神经网络的训练速度,提高了神经网络的诊断准确度。
Aiming at the reliability and safety of the steer-by-wire vehicles and the shortcomings of the fault diagnosis methods, a soft-based fault diagnosis method of steer-by-wire steering is proposed. This method uses the rough set in soft computing and the radial Basis neural network to combine. The rough set is treated as the input of RBF neural network, and the attribute reduction of sample data is carried out. The reduced attribute set is used as the input of RBF neural network to shorten the network training time. Particle swarm optimization algorithm is used to encode and optimize the basis function center value and width of radial basis function neural network. The RBF neural network is constructed by using the optimal center value and width, so that the sample training error of RBF neural network Compared to not optimize before a certain degree of reduction. Then the trained neural network is used to test the fault samples. The test results show that this method accelerates the training speed of the neural network and improves the diagnostic accuracy of the neural network.