Insight -â€‚Mechanical Engineering

Based onK-SVDDictionary learning algorithm of sparse said vibration signal compression measurement Reconstruction Methods

Mingzhi Zhang

For the current mechanical vibration signal band more and more wide basis traditional Shannon-In quest sampling theorem data collection of an arcane will get big vibration data the storage, transmission and processing bring difficult of problem put forward. Based onK-SVDDictionary learning algorithm of sparse said vibration signal compression measurement reconstruction methods. First analysis the vibration signal in based onK-Singular Value Decomposition(K-Singular Value decomposition K-SVD)Dictionary learning algorithm get of over-complete dictionary on the approximate sparse of CAN compression; then use Gaussian random matrix of vibration signal the compression measurement; finally based on compression measurements the orthogonal Matching Pursuit algorithm the original vibration signal the reconstruction. Simulation Test results show that when vibration signal compression ratio in60%~90%When based onK-SVDDictionary learning algorithm structure of over-complete dictionary than based on discrete cosine over-complete Dictionary Compression sensing reconstruction relative error small. The methods not only can get is high signal compression ratio and has accurate of Signal Reconstruction performance in don't lost vibration information of situation under greatly reduce the original vibration data.

Fretting Fatigue Life of connecting rod under multi-axial variable amplitude Stress

Lingyan Hang

Summary:A fatigue life calculation model for multi-axial variable-amplitude fatigue stress is proposed.ParisThe fatigue crack propagation rate formula is suitable for multi-axial variable amplitude stress.ParisThe life-span calculation model is obtained by introducing the modified short crack size into the formula, the life of the dangerous nodes on the mating surface is calculated. The results show that the model can be used to calculate the life of structural surface cracks from any initial size to any design size under complex conditions, which is of great significance to guide the life design of key components and to determine the maintenance time.

Modeling analysis, experiments. a Three-dimensional Bridge-type mechanism. combined flexure Hinges

Lina Zhang

Solve."low amplification ratio. compliant bridge-type mechanism a three dimensional bridge-typeMechanism. combined flexure Hinges. proposed which not only remains. superiority. conventional bridge-type mechanisms e.g ., structure symmetry compact size, simple design but also achieves high amplification ratio.. analyze. performance. 3D bridge-type mechanism, statics model. established via.Compliance matrix method based. which a new eval Index,. relative amplification. proposed. measure. displacement loss.mechanism. several 3D bridge type mechanisms. three frequently used flexure Hinges. analyzed. established model.Show, 3D bridge-type mechanism reaches. optimal performance when V-shaped hinge, filled leaf hinge. self-employed. bridge 1, bridge 2. finally a bridge-type mechanism. amplIfication ratio 41, relative amplification. 0.9. confirmed by FEA simulation, experiments.

Windowed interpolation Fast Fourier Transform in Rolling Bearings Application of Fault Diagnosis

Fei Li

In order to accurately identify the Fault Characteristic Frequency of rolling bearings, The Minimum Entropy Deconvolution andTeagerOn the basis of energy operator demodulation,HanningA New Fault Diagnosis Method for Rolling Bearings Based on window interpolation and fast Fourier transform. Firstly, the Minimum Entropy Deconvolution is used to denoise the bearing fault signal.TeagerAfter de-noising, the fault vibration signal is demodulated by energy operator.TeagerDemodulation spectrum; andHanningFinally, the amplitude of the three discrete spectrum near the signal frequency points is interpolated to get the accurate fault characteristic frequency. The analysis results of bearing vibration signals show thatTeagerCompared with the Energy Operator demodulation method