The invention provides a bearing health assessment and prediction method and system The method comprises the following steps: acquiring a vibration signal in the running process of a bearing; processing the vibration signal to acquire time domain statistical characteristic, a time frequency domain energy characteristic and a trigonometric function characteristic of the vibration signal; selecting the characteristics which are mostly sensitive to the bearing degradation according to the characteristic index of the bearing degradation, so as to form characteristic vectors; selecting the characteristic vectors during the normal running of the bearing as input vectors to train a self-organizing characteristic mapping network and establishing a self-organizing characteristic mapping network model; solving the minimum quantization error corresponding to each characteristic vector in the self-organizing characteristic mapping network model; constructing health indexes which reflect the bearing degradation performance by utilizing the minimum quantization errors; and predicting the trend of the bearing degradation by utilizing an extreme learning machine prediction method and the health indexes According to the bearing health assessment and prediction method and system, the degradation degree in the bearing degradation process can be correctly described, and the trend of the bearing degradation can be accurately predicted

Bearing health assessment and prediction method and system

The invention discloses an intelligent diagnosis method for faults of a mining drilling machine gearbox. The method comprises the steps as follows: adopting an experiential mode decomposition method to decompose a mining drilling machine gearbox monitoring signal into a plurality of basic mode components, and extracting weak fault information in the signal; calculating time-domain statistical characteristic and frequency-domain statistical characteristic of each basic mode component, utilizing a distance assessment technique to select statistical characteristics sensitive to the faults, and constructing an input vector; applying a multi-classification transductive support vector machine to perform classification on the running state of the gearbox, and performing intelligent diagnosis on the various gear faults of the mining drilling machine gearbox. According to the invention, the diagnosis efficiency is high, and various weak faults of the mining drilling machine gearbox can be effectively diagnosed.

Intelligent diagnosis method for faults of mining drilling machine gearbox

The invention relates to a mechanical structure dynamic load identification method based on a cubic b-spline scaling function The method comprises the steps of (1) measuring a frequency-response function between a mechanical structure position point bearing a dynamic load and a mechanical structure response measurement point with the hammering method, obtaining a unit impulse response function through rapid Fourier inversion, and then obtaining a transfer matrix through deconvolution; (2) measuring an acceleration signal of the response point generated by the structure dynamic load through an accelerator sensor; (3) establishing scaling function matrixes, dual matrixes and coefficient matrixes of the cubic b-spline scaling function under different layer numbers, and selecting a coefficient matrix with the minimum conditional number; (4) calculating the weight coefficient of the cubic b-spline scaling function under the regularized layer number by means of an acceleration response and coefficient matrix; (5) obtaining a regularized mechanical structure dynamic load to be identified by multiplying the weight coefficient by the corresponding scaling function matrix

Mechanical structure dynamic load identification method based on cubic b-spline scaling function

The invention discloses a method for predicting the residual life of an aero-engine. The method comprises the steps of acquiring full-life-operation-cycle sensor measurement data of the aero-engine and selecting sensor parameters constructing failure propagation models from the data; fusing the selected sensor parameters into first health indexes describing a health state of the engine; according to a failure propagation modeling principle, fitting an exponential type failure propagation model by each set of health index sequences to construct a failure propagation model library; acquiring historical sensor measurement data of the on-service aero-engine and fusing the sensor parameters into second health indexes; performing similarity matching on each set of the health index sequences and the models in the failure propagation model library; and according to a matching result, predicting the residual life of the on-service aero-engine of the specific type. The method has the following advantages: an effective technical support is provided for prediction and health management of the aero-engine and an effective reference is provided for a subsequent repair decision of the engine.

Method for predicting residual life of aero-engine

The invention relates to a mechanical rubbing fault feature extraction method based on multi-wavelet kernel-support vector regression (SVR). The method comprises the following steps: 1) a vibration signal of a mechanical key component is measured and taken, and the signal is subjected to redundant lifting scheme wavelet (RLSW) decomposition; 2) each layer of detail signals and the last layer of approximation signals obtained by decomposition are subjected to adaptive singular value decomposition (ASVD) noise reduction; 3) a signal is reconstructed by redundant lifting wavelets, and a noise reduction signal is acquired; 4) with two interval three-time Hermite spline multi-wavelet scale functions as interpolation basis functions, a novel multi-wavelet kernel function is constructed; and 5) the vibration signals after noise reduction are divided into a training set and a prediction set, the training set is used to construct a prediction model, the prediction set is predicted, and a weak periodic impact component hidden in the original vibration signal is extracted by the prediction residual. Thus, the mechanical rubbing fault features can be extracted.

Mechanical rubbing fault diagnosis method based on multi-wavelet kernel-support vector regression (SVR)

The invention discloses a second generation small-wave support vector machine assessment method for damage and remaining life of a metal structure. According to the method, an intrinsic mode function is obtained through decomposition of an experience mode, a time-frequency domain statistic characteristic of the intrinsic mode function is extracted, a most sensitive characteristic is chosen according to a distance accessment principle to construct a best characteristic set, a minimum quantization error indicator which has an obvious performance degradation trend along with time changes is constructed by means of self-organization neural network characteristic fusion techniques, a biorthogonal small-wave support vector machine kernel function of second generation small-wave transform is provided, a service life prediction model of the second generation small-wave support vector machine is constructed, the minimum quantization error indicator serves as a prediction characteristic, and quantitative assessment for the damage and the remainig service life of a metal structural component of mechanical equipment under a small subsample is achieved.

He Zhengjia

Papers

Second generation small-wave support vector machine assessment method for damage and remaining life of metal structure