Multi-faults diagnosis of rolling bearings via adaptive customization of flexible analytical wavelet bases

Multi-faults detection is a challenge for rolling bearings due to the mode mixture and coupling of multiple fault features, as well as its easy burying in the complex, non-stationary structural vibrations and strong background noises. In this paper, a method based on the flexible analytical wavelet transform (FAWT) possessing fractional scaling and translation factors is proposed to identify multiple faults occurred in different components of rolling bearings. During the route of the proposed method, the proper FAWT bases are constructed via genetic optimization algorithm (GA) based on maximizing the spectral correlated kurtosis (SCK) which is firstly presented and proved to be efficient and effective in indicating interested fault mode. Via using the customized FAWT bases for each interested fault mode, the original vibration measurements are decomposed into fine frequency subbands, and the sensitive subband which enhances the signal-to-noise ratio (SNR) is selected to exhibit the fault signature on its envelope spectrum. The proposed method is tested via simulated signals, and applied to analyze the experimental vibration measurements from the running roller bearings subjected to outrace, inner-race and roller defects. The analysis results validate the effectiveness of the proposed method in identifying multi-faults occurred in different components of rolling bearings.     

基于CMPSO算法的永磁同步电机转动惯量识别研究

永磁同步电机(PMSM)是一种具有很强非线性的动态系统,在工业驱动应用中发挥着重要作用。实时辨识转动惯量对于高精度PMSM系统控制和稳定性状态监测具有重要意义,但传统转动惯量识别方法精度较低。在传统离散模型参考自适应理论基础之上,用柯西变异粒子群优化(CMPSO)算法代替待辨识参数自适律设计环节,以实现PMSM转动惯量识别。该转动惯量辨识方法充分利用了CMPSO算法的快速高效收敛性,仿真结果和试验结果表明了可行性、正确性和准确性。     

A new bearing fault diagnosis method based on modified convolutional neural networks

Fault diagnosis is vital in manufacturing system. However, the first step of the traditional fault diagnosis method is to process the signal, extract the features and then put the features into a selected classifier for classification. The process of feature extraction depends on the experimenters’ experience, and the classification rate of the shallow diagnostic model does not achieve satisfactory results. In view of these problems, this paper proposes a method of converting raw signals into two-dimensional images. This method can extract the features of the converted two-dimensional images and eliminate the impact of expert’s experience on the feature extraction process. And it follows by proposing an intelligent diagnosis algorithm based on Convolution Neural Network (CNN), which can automatically accomplish the process of the feature extraction and fault diagnosis. The effect of this method is verified by bearing data. The influence of different sample sizes and different load conditions on the diagnostic capability of this method is analyzed. The results show that the proposed method is effective and can meet the timeliness requirements of fault diagnosis.     

Inverse identification of constitutive parameters of Ti2AlNb intermetallic alloys based on cooperative particle swarm optimization

Ti₂AlNb intermetallic alloy is a relatively newly developed high-temperature-resistant structural material, which is expected to replace nickel-based super alloys for thermally and mechanically stressed components in aeronautic and automotive engines due to its excellent mechanical properties and high strength retention at elevated temperature. The aim of this work is to present a fast and reliable methodology of inverse identification of constitutive model parameters directly from cutting experiments. FE-machining simulations implemented with a modified Johnson-Cook (TANH) constitutive model are performed to establish the robust link between observables and constitutive parameters. A series of orthogonal cutting experiments with varied cutting parameters is carried out to allow an exact comparison to the 2D FE-simulations. A cooperative particle swarm optimization algorithm is developed and implemented into the Matlab programs to identify the enormous constitutive parameters. Results show that the simulation observables (i.e., cutting forces, chip morphologies, cutting temperature) implemented with the identified optimal material constants have high consistency with those obtained from experiments, which illustrates that the FE-machining models using the identified parameters obtained from the proposed methodology could be predicted in a close agreement to the experiments. Considering the wide range of the applied unknown parameters number, the proposed inverse methodology of identifying constitutive equations shows excellent prospect, and it can be used for other newly developed metal materials.     

A drilling tool design and in situ identification of planetary regolith mechanical parameters

The physical and mechanical properties as well as the heat flux of regolith are critical evidence in the study of planetary origin and evolution. Moreover, the mechanical properties of planetary regolith have great value for guiding future human planetary activities. For planetary subsurface exploration, an inchworm boring robot (IBR) has been proposed to penetrate the regolith, and the mechanical properties of the regolith are expected to be simultaneously investigated during the penetration process using the drilling tool on the IBR. This paper provides a preliminary study of an in situ method for measuring planetary regolith mechanical parameters using a drilling tool on a test bed. A conical-screw drilling tool was designed, and its drilling load characteristics were experimentally analyzed. Based on the drilling tool-regolith interaction model, two identification methods for determining the planetary regolith bearing and shearing parameters are proposed. The bearing and shearing parameters of lunar regolith simulant were successfully determined according to the pressure-sinkage tests and shear tests conducted on the test bed. The effects of the operating parameters on the identification results were also analyzed. The results indicate a feasible scheme for future planetary subsurface exploration.     

智能方法在惯性系统误差参数辨识中的应用

总结了目前我国提高惯性系统导航精度的技术途径,阐述了国内外惯性系统误差参数辨识方法的研究现状,介绍了当前滤波算法、智能优化算法和人工神经网络方法在惯性系统导航领域的应用情况以及存在的不足。最后,分析了空间飞行器惯性系统误差参数辨识技术的未来研究方向,即智能优化算法和人工神经网络方法等智能方法将在惯性系统误差参数辨识中发挥越来越重要的作用,通过将误差系数标定问题转换为参数辨识问题,采用智能方法在庞大的解空间内实现对惯性系统误差参数的快速辨识。     

基于Hilbert-Huang变换的航天器发射段力学环境分析

文章基于Hilbert-Huang变换,对某载人航天器发射过程中整流罩分离期间测量的力学环境参数进行处理分析,利用HHT的基函数自适应性特点,得到振动时间历程信号的时频域Hilbert谱,准确识别出信号主频率及对应时间范围。此应用案例可为地面力学试验条件制定以及运载接口条件确定提供参考。     

月球表层采样样品智能确认方法

表层采样是月球采样探测的重要方式，样品智能确认有助于提升工作效率与复杂问题处理能力。结合月球表层采样铲挖工作过程，分析了铲挖过程中臂载相机图像的特点，模仿有人参与识别过程，提出了层次解耦的月球样品智能识别流程，利用深度学习方法构建了一类深度卷积识别网络，完整地描述了图像、特征、标记在网络中的正反传递关系，并在月球表层采样地面试验中进行了验证，结果表明该方法对不同光照、不同背景、不同过程、不同形态的样品，具有较好的泛化识别能力，误识别率优于8.1%，平均单幅识别时间约0.7 s。     

基于叶尖定时的叶片耦合振动参数辨识与仿真

为了实时掌握叶片振动情况、预防故障发生,目前最常用的非接触振动测量方法是基于单自由度模型的叶尖定时法,但当叶片耦合振动时2个振动峰相隔很近,该方法无法同时识别出其振动参数。采用基于2自由度模型的曲线拟合方法,得到叶片耦合振动2个峰的初始相位等参数,用相位遍历法辨识出叶片的振动阶次和频率。结果表明:采用基于2自由度模型的曲线拟合方法提高了叶片耦合振动参数辨识的精度,辨识了相邻2个峰的振动参数。辨识振动与仿真参数设置基本一致。     

未知时延系统的全局优化辅助变量辨识

对于未知时延系统,借助分离性原理,推导出迭代的可分离的非线性最小二乘（SNLS）辨识方法。为了降低收敛于局部最小的可能性,消除强观测噪声所引起的参数估计的偏差,利用全局优化理论,引进辅助变量,推导了全局优化的辅助变量（GOIV）辨识方法。仿真试验验证了算法的有效性。