375 results about "Bearing vibration" patented technology

The invention discloses an attention mechanism and LSTM fused industrial bearing vibration time sequence signal fault prediction method and system. The method comprises the following steps: collecting a bearing vibration signal and carrying out filtering, noise reduction and normalization preprocessing; constructing a deep learning model combining the bidirectional BiLSTM and a coordinate attention mechanism to extract bidirectional time sequence features and enhance key fault features; carrying out model training by adopting a multi-target composite loss function and an Adam optimizer, and introducing an early stop mechanism to prevent overfitting; performing fault type identification and degree evaluation on the real-time vibration signal by using the trained model, and performing quantitative analysis by fusing multi-scale spectrum kurtosis features and nonlinear kinetic parameters; and finally, outputting a fault diagnosis report, and triggering multi-stage early warning based on an adaptive threshold. The method can realize high-precision and high-reliability bearing fault prediction and health state evaluation, and is suitable for intelligent operation and maintenance of industrial equipment.

The invention belongs to the technical field of bearing fault diagnosis, and particularly relates to a transmission casing typical small bearing fault weak vibration signal extraction method, which comprises the following steps: reconstructing a bearing vibration signal to obtain a modulation impact signal, and calculating a weighted kurtosis value based on the modulation impact signal to improve a Protugram algorithm; noise interference is added to a bearing fault signal, and extraction of a small bearing fault weak impact signal is realized by optimizing a filtering frequency band based on a weighted kurtosis value improved Protugram algorithm; bearing impact characteristics can be enhanced based on a simulation sensor resonance enhancement algorithm (SPM); the impact quantification of the bearing can eliminate the influence of different rotating speeds and load working conditions, and realizes the quantitative diagnosis of bearing faults. And finally, the fault data of the rolling body, the inner ring and the outer ring of the flight attachment case small bearing part test bed are verified, and effective extraction and enhancement of bearing fault weak signals can be realized.

The invention relates to the technical field of electrical monitoring, provides a general generator electrical monitoring system with a fault self-diagnosis function, and aims to deeply excavate potential correlation between electrical and mechanical parameters by judging a coherence coefficient and a phase difference between a current harmonic component and a bearing vibration frequency band and marking a fault coupling identifier by using a coupling mode library. Electrical and mechanical coupling faults can be accurately identified, and the identification capability of complex faults can be greatly improved; meanwhile, the diagnosis threshold is dynamically adjusted in combination with the load rate and the winding temperature, so that the system can better adapt to different operation conditions of the generator, and the diagnosis accuracy and reliability are improved; the fault causal chain is analyzed through the causal inference algorithm, and the fault source is positioned, so that compared with the existing fault tracing mode lacking systematicness, the fault generation reason and process can be analyzed more comprehensively and deeply, the fault source can be positioned quickly and accurately, operation and maintenance personnel can take targeted measures in time, and the fault tracing efficiency is improved. And the operation safety and reliability of the generator are improved.

The invention relates to a rolling bearing early fault early warning and diagnosis method, which comprises the steps of extracting an envelope component from a bearing vibration signal, constructing a time-delay feedback stochastic resonance optimal model by taking an improved signal-to-noise ratio INSR as an optimization objective function, obtaining an output signal, obtaining a first reconstruction signal through CEEMDAN adaptive decomposition and IMF component screening, and obtaining a second reconstruction signal through the CEEMDAN adaptive decomposition and IMF component screening. Calculating the signal-to-noise ratio ISNR of the vibration signal at the fault characteristic frequency, comparing the signal-to-noise ratio ISNR with a preset initial threshold value, judging whether an early warning is given out or not, and if the early warning is given out, processing the vibration signal by using a multi-wavelet adjacent coefficient adaptive threshold value method to obtain a denoised signal; processing the denoised signal by combining a fast spectral kurtosis method and an ensemble empirical mode decomposition method to obtain a second reconstructed signal; and processing by using improved fast spectrum correlation to obtain a corresponding enhanced envelope spectrum, and comparing the enhanced envelope spectrum with a fault characteristic frequency for identification. Compared with the prior art, accurate early warning and diagnosis can be carried out on early weak faults of the rolling bearing.

The invention relates to the technical field of bearing fault diagnosis, in particular to a multi-bearing fault positioning and diagnosing method, which comprises the following steps of: acquiring bearing vibration signals of a plurality of sensing points, extracting peak time information to complete synchronous calibration, extracting high-frequency band energy and constructing a feature vector, screening deviation features and clustering and grouping, and identifying an energy abnormal channel, positioning a bearing position, extracting time domain and frequency domain features, comparing fault modes, and outputting a fault type diagnosis result. According to the method, the consistency of multi-channel data synchronization is realized by extracting the peak point of the vibration signal and calibrating the time offset, the high-frequency energy change rate is utilized to match the weight coefficient, the detection sensitivity of tiny fault features is enhanced, and the separability of a composite fault state is improved by combining feature relative deviation rate screening and density trend clustering. A frequency section energy proportion change recognition channel is adopted, accurate positioning of a fault bearing is achieved, and accurate judgment of a fault type is achieved in combination with a multi-feature comparison mode.

The invention discloses an online monitoring method and system for the wear state of a bearing of a flour mill, particularly relates to the technical field of state monitoring of industrial equipment, and is used for solving the problem of misjudgment caused by signal characteristic deviation under a dynamic load working condition in an existing flour mill bearing vibration signal monitoring method. A grinding pressure parameter and a bearing vibration signal are collected in real time, the dynamic incidence relation of the grinding pressure parameter and the bearing vibration signal is analyzed, interference frequency band components are separated, the time-frequency ridge line track and curvature sudden change density of a characteristic frequency band are extracted, a dynamic reference standard is generated to adjust an amplitude-frequency threshold value in a self-adaptive mode, and finally a wear state evaluation result is output through amplitude-frequency joint analysis. Accurate monitoring of the wear state of the bearing under the complex working condition is achieved, and the reliability of an online monitoring system and the scientificity of maintenance decision are remarkably improved.

The invention discloses a bearing defect intelligent detection method and system based on deep learning, and relates to the field of bearing defect detection. Multi-modal data such as bearing vibration, acoustics, thermal imaging and the like are acquired by using various sensors, and a four-dimensional feature tensor is constructed through preprocessing such as noise reduction and feature extraction; features are fused through a reconfigurable multi-branch convolutional neural network, and defects are identified and a development trend is predicted in combination with a meta-learning twin network; a decision threshold is optimized by adopting a quantum heuristic algorithm, and multi-level early warning is realized; and continuous evolution of the model is completed through edge-cloud collaboration and federated learning, the functions of data calibration compensation, model dynamic optimization and the like are achieved, and efficient and accurate detection of bearing defects is achieved. The detection time is remarkably shortened, and the positioning precision is high; the novel defect response speed is high, and faults can be predicted in advance; system energy consumption is reduced, model updating is improved, stable operation of equipment is effectively guaranteed, and cost reduction and efficiency improvement of industrial intelligent operation and maintenance are facilitated.

The invention belongs to the technical field of vibration analysis and testing of bearings, and particularly relates to a cam driven bearing wear degree monitoring method and system, and the method comprises the steps: carrying out the equal-angle resampling processing of a vibration signal through a rotating speed signal, decomposing an obtained angular domain vibration signal into a plurality of mode components through a variational mode decomposition algorithm, and carrying out the measurement of the vibration signal; according to the kurtosis value of each modal component and the correlation coefficient of each modal component and the original vibration signal, evaluating the impact saliency weight of each modal component, and performing weighted summation on the energy of each modal component to obtain comprehensive impact energy; calculating to obtain a speed decoupling wear index without the influence of the rotating speed by utilizing the comprehensive impact energy and the vibration energy calculated by the physical mapping model; and the speed decoupling wear index is compared with a preset self-adaptive alarm threshold value, and the wear state of the cam driven bearing is judged according to a comparison result. According to the invention, the problems of false alarm and missing alarm under the variable-speed working condition are solved.

The invention relates to the technical field of lubricating oil monitoring, in particular to a compressor lubricating oil online monitoring vibration diagnosis method and system, and the method comprises the steps: obtaining vibration signal data of a compressor bearing, a connecting rod mechanism and a piston assembly and lubricating oil medium parameters, arranging the data according to a time sequence, and generating a lubrication working condition time sequence signal data set; according to the method, through time sequence matching of vibration signal data and lubricating oil state parameters, a lubricating working condition time sequence signal data set is established, and in combination with time window division, the bearing vibration frequency offset, the piston transverse impact amplitude and the connecting rod torsional vibration change rate are extracted; and calculating the viscosity drop rate, the temperature change range and the pollutant particle growth rate of the lubricating oil, screening the correlation trend of the bearing vibration frequency offset before and after the change of the lubricating state, and enhancing the lubricating state recognition capability.

The invention relates to the technical field of fault diagnosis, and discloses a fault diagnosis method and system based on variational mode decomposition. The method comprises the steps of obtaining an original bearing vibration signal; introducing a sparsity penalty term into the mathematical model of the variational mode decomposition algorithm to obtain an improved variational mode decomposition algorithm; decomposing the original bearing vibration signal based on an improved variational mode decomposition algorithm to obtain a decomposition result; and obtaining an envelope spectrum of each mode according to a decomposition result, and determining a fault frequency according to a frequency corresponding to a maximum peak value of the envelope spectrum. According to the method, the sparsity penalty term is introduced into the mathematical model of the variational mode decomposition algorithm, the bearing vibration signal is decomposed through the improved variational mode decomposition algorithm, the envelope spectrum is solved for each mode obtained through decomposition, the frequency corresponding to the maximum peak value of the envelope spectrum is the fault frequency, and the fault frequency can be calculated under the noise condition. Accurate extraction of fault frequency is realized, and noise interference is overcome.

The invention discloses a mining ventilator bearing fault diagnosis method based on a CEEMDAN and a multi-scale spatio-temporal information fusion map neural network. The method comprises the steps of collecting a mining ventilator bearing vibration signal; the vibration signals of the mining ventilator bearing are preprocessed; cEEMDAN is carried out on the preprocessed mining ventilator bearing vibration signal, and the signal is decomposed into a plurality of IMF components through adaptive multi-scale decomposition; constructing a multi-scale space-time convolution module, and adopting convolution kernels of different scales to extract time sequence characteristics of each IMF component in parallel; introducing a channel attention mechanism to perform adaptive weighting on the multi-scale spatial-temporal features, and dynamically strengthening the characterization intensity of the key fault mode by calculating the importance weight of the channel features; and aggregating the multi-scale spatial-temporal features through GCN to generate global feature representation, and inputting the global feature representation into a classifier to output a fault diagnosis result. Through combination of the CEEMDAN and the multi-scale spatio-temporal information fusion graph neural network, the technical bottlenecks of noise covering, spatio-temporal correlation deficiency and insufficient generalization ability are broken through.

The invention belongs to the technical field of fault diagnosis, and particularly relates to a fault feature extraction method and system for bearing data, and the method comprises the following steps: S1, carrying out the multi-frequency-band division of a collected bearing vibration signal, calculating the envelope spectrum of each frequency band, selecting the frequency band with the maximum peak amplitude in the envelope spectrums, and carrying out the multi-frequency-band division; determining a main peak frequency of the envelope spectrum as a fault characteristic frequency, and performing conversion to obtain a fault impact period T; s2, calculating an autocorrelation function of the bearing vibration signal, and determining the length L of an FIR filter according to the attenuation characteristic of the autocorrelation function and the fault impact period T; and based on the fault impact period T, carrying out period synchronous averaging on the bearing vibration signal. According to the method, the problem that interference is easy to amplify or effective information is easy to lose under strong noise is solved, so that the extracted fault features are purer and more prominent on an envelope spectrum, and the recognition capability of early weak faults is enhanced.

The invention relates to the field of rotating machinery fault diagnosis, and discloses a bearing fault diagnosis method based on a physical perception KAM network, and the method comprises the steps: carrying out the discretization of a bearing vibration signal through a Gabor filter group based on physical prior initialization, and generating modal feature lexical elements with physical frequency band meanings; and inputting the lexical elements into a PC-KAM backbone network, calculating a hidden state vector by using a state space model branch, and dynamically adjusting the position of a primary function node of a Kolmogov-Arnod network branch to realize collaborative dynamic feature extraction. In the training stage, an orthogonal subspace constraint and physical perception low-rank adaptation fine tuning mechanism is introduced. And finally, searching a historical fault case, performing multi-modal fusion on the historical fault case and the deep feature sequence, mapping a fusion representation into a soft prompt, and inputting the soft prompt into a large language model to generate a diagnosis report. According to the method, the problems of poor physical interpretability of characteristics and few-sample diagnosis under variable working conditions are effectively solved, and the generalization and decision-making ability of a diagnosis system are improved.

The invention discloses a numerical control machine tool power tool apron noise reduction method based on UUSGSR, and belongs to the technical field of numerical control machine tool servo tool rest power tool apron monitoring. The method comprises the specific steps that Hilbert transform and elliptic filtering preprocessing are conducted on monitored bearing vibration signals in a tool rest power tool apron; constructing an unsaturated scaling type Gaussian potential stochastic resonance model processing signal; a high-order weighted harmonic noise ratio HWHNR is proposed as a target function to reflect signal nonlinear characteristics; optimizing the structural parameters and the damping factors by using a grey wolf optimization algorithm to obtain an optimal under-damping unsaturated scaling type Gaussian potential stochastic resonance model UUSGSR containing the model structural parameters and the damping factors, and performing stochastic resonance processing on bearing vibration signals in the power tool apron of the tool rest by using the model. Noise contained in a bearing vibration signal in the power tool apron of the tool rest is weakened, and the signal-to-noise ratio is improved. In conclusion, the method can realize noise reduction of the vibration signal of the bearing in the power tool apron of the tool rest.

The invention discloses a trailer and a trailer damping method.The trailer is applied to a wheel track type belt conveyor and used for supporting and dragging a conveying belt on a track of the wheel track type belt conveyor and walking along with the conveying belt, the trailer comprises a frame and a plurality of wheel assemblies distributed on the two sides of the frame, and the frame is provided with mounting seats corresponding to the wheel assemblies one to one; the wheel assembly comprises a wheel body, a wheel axle and a bearing, the wheel axle is in transmission connection with the wheel body through the bearing, at least part of the area of the wheel axle penetrates through a mounting hole formed in the mounting base and is in clearance fit with the mounting hole, and an elastic structure located between the mounting base and the wheel axle is arranged in the mounting hole. When the trailer transports a conveying belt and materials on the track, material loads and wheel-rail force are subjected to energy storage and energy consumption when passing through the elastic structure, so that the overall vibration of the trailer is attenuated, the vibration reduction effect of the trailer during operation on the track can be greatly improved, the peak value of a bearing vibration curve can be reduced, the peak clipping effect is achieved, and the service life of the bearing is prolonged.

The invention relates to the technical field of nuclear power plant equipment safety, in particular to a bearing fault diagnosis method and device, electronic equipment and a storage medium. According to the bearing fault diagnosis method provided by the embodiment of the invention, target bearing vibration data is acquired; performing multi-scale feature extraction on the target bearing vibration data through the bearing fault diagnosis model to obtain multi-scale bearing vibration features; carrying out bearing state detection on the target bearing based on the multi-scale bearing vibration characteristics to obtain a bearing vibration state; performing feature filtering processing on the multi-scale bearing vibration features to obtain filtered bearing vibration features; and carrying out bearing fault diagnosis on the target bearing according to the multi-scale bearing vibration characteristics, the bearing vibration state and the filtered bearing vibration characteristics to obtain a bearing fault category. Therefore, the accuracy of bearing fault diagnosis can be improved.

The invention relates to the technical field of cooling water abnormity monitoring, in particular to a steel slag pressure hot stuffy rolling cooling water abnormity alarm system and method.Roll gap cooling water flow, injection temperature, hot stuffy tank pressure and bearing vibration data are collected, a dynamic mapping relation between position coordinates of a spraying opening and a water inlet valve is established, temperature and pressure time sequence offset is calculated, and the temperature and pressure time sequence offset is calculated; the temperature judgment boundary is corrected by combining the pressure bearing limit value of the sealing ring and the real-time flow, and an abnormal alarm is triggered according to the temperature over-limit frequency in the continuous time period. A multi-source analysis framework is constructed by integrating temperature, flow, pressure and vibration parameters, single-point monitoring limitation is eliminated, nonlinear influence of the flow on tank pressure is quantified by adopting a dynamic weight model, a self-adaptive temperature threshold value is generated in combination with a sealing pressure bearing limit, dynamic coupling of an alarm mechanism is realized, an abnormal scene is positioned based on time sequence analysis and coordinate mapping, and the accuracy of monitoring is improved. The sealing hidden danger is pre-judged through the thermal unbalance coefficient, and the pressure rate deviation is intercepted in real time.

The invention belongs to the technical field of rail transit equipment, and provides a vibration damping device for rail transit, which comprises a vibration damping seat, the top of the vibration damping seat is vertically connected with a rail in a sliding manner, and a bearing vibration damping assembly is arranged between the rail and the vibration damping seat; the two sets of main vibration reduction assemblies are arranged on the two sides of the rail correspondingly, each main vibration reduction assembly comprises an eccentric block arranged in a vibration reduction base, a rotary driving mechanism is arranged between the eccentric block and the bearing vibration reduction assembly, and the rotary driving mechanism drives the eccentric block to swing in a reciprocating mode when the rail vibrates up and down; the hydraulic vibration reduction mechanism is in transmission connection with the rotary driving mechanism, a secondary vibration reduction part is arranged in the hydraulic vibration reduction mechanism and corresponds to the eccentric block, and when the amplitude of the track is too large, the eccentric block abuts against the secondary vibration reduction part to increase damping of the hydraulic vibration reduction mechanism. The damping device can adapt to different vibration intensities, can automatically enhance damping force under the large-amplitude working condition, and provides a better vibration reduction effect.

The invention discloses a self-control maintenance method and system for an intelligent clean air conditioner, and belongs to the technical field of clean air conditioners. According to the method, a multi-source heterogeneous data sensing layer is constructed to collect fan bearing vibration data, filter monitoring data and surface air cooler monitoring data; and constructing a fault prediction model according to the fan bearing vibration data, adopting a machine to learn vibration spectrum time-varying characteristics, predicting the residual life of the fan bearing, and performing self-control maintenance. A filter blockage dynamic model is further constructed according to filter monitoring data, maintenance and air volume compensation are carried out according to the blockage condition and the service life of a bearing, meanwhile, the frosting condition of a surface air cooler is monitored, hot gas bypass defrosting is carried out in time, the clean air conditioner can find and process potential faults in time, and the stability of the system is improved; the intelligent self-control maintenance greatly reduces the frequency and cost of manual maintenance, and improves the management efficiency of the equipment at the same time.

This invention relates to the field of data processing, and more particularly to an oscilloscope signal data processing method and system for bearing testing. The method involves acquiring bearing vibration data using an oscilloscope, extracting data from a first time moment for frequency domain analysis, obtaining its power spectrum, and dividing it into frequencies to obtain a first frequency sequence. Similarly, a second frequency sequence is obtained from the data at a second time moment. Aligning the two frequency sequences yields frequency matching pairs. Utilizing the frequency variations of neighboring frequencies within a single matching pair, the noise figure of each frequency in the matching pair is calculated, resulting in the noise figure for any frequency in the first frequency sequence. Based on these noise figures, a first mean sequence is constructed. The distribution consistency of each frequency point in the first mean sequence is obtained, and the segmentation effectiveness of each frequency point is calculated. Combined with a noise segmentation threshold, oscilloscope signal processing is completed. This invention denoises bearing vibration data acquired by an oscilloscope, improving the accuracy of the denoised oscilloscope signal data.

The invention discloses a generator stator foot load adjustment test method based on vibration control, and the method comprises the steps: obtaining the shaft vibration and bearing vibration data of a front bearing and a rear bearing, and judging whether a supporting system has a structure resonance phenomenon or not; measuring bottom foot vibration data, and judging whether a structure resonance phenomenon caused by unreasonable stator bottom foot load distribution possibly exists or not; performing an overspeed test on the unit, and performing a stator bottom load adjustment test and judging a frequency modulation direction if a formant exists and a working rotating speed avoidance frequency margin meets a judgment condition; performing a modal test on the stator during shutdown to obtain cold-state initial modal frequency data, correcting the cold-state initial modal frequency data with the hot-state modal frequency data, and verifying the frequency modulation direction; performing a stator foot load uniform distribution adjustment test and modal test verification, and if a modal frequency avoidance degree requirement is met, ending the test; otherwise, performing a stator foot load fine adjustment test, calculating to obtain elevation fine adjustment data of each area of the stator foot meeting conditions according to the influence coefficient, and ending the test.

The invention discloses a few-sample fault diagnosis method based on a multi-scale physical information neural network, and belongs to the technical field of fault diagnosis. The method comprises the following steps: step 1, constructing a bearing vibration signal data set containing multiple working conditions, and preprocessing an original vibration signal; 2, inputting the vibration signal samples in the training set into a CCSWGAN network, generating a high-quality synthetic sample, and carrying out normalization processing on the high-quality synthetic sample; step 3, mixing the original sample and the synthetic sample and randomly disrupting the original sample and the synthetic sample to form training data, inputting the training data into an MCB-PII module, and performing multi-scale feature extraction and fault diagnosis; and 4, deploying the trained model to production equipment, and carrying out fault detection and identification on vibration signals collected in real time. According to the method, the problems of unstable performance and insufficient adaptability of the model under complex working conditions are effectively avoided, so that higher accuracy and reliability are provided for bearing fault diagnosis.

The invention relates to the technical field of bearing vibration detection, in particular to a vibration detection device based on artificial intelligence, which comprises a detection table, a control panel, a detection mechanism, a driving mechanism, a feeding and discharging mechanism and a mounting mechanism, and is characterized in that the control panel is fixedly mounted on the detection table, and the detection mechanism is fixedly mounted on the detection table; the detection mechanism is used for detecting the bearing, the driving mechanism is used for driving the mounting mechanism to move, the mounting mechanism is fixedly mounted on the detection table and used for fixing the bearing, and the feeding and discharging mechanism is fixedly mounted on the detection table and used for feeding and discharging the bearing. According to the vibration detection device based on artificial intelligence, through arrangement of the feeding and discharging mechanism and the conveying mechanism, the moving block can sequentially drive the multiple different mounting blocks and the bearing table to move, the working efficiency and the using effect of the vibration detection device are improved, and the production and maintenance efficiency of new energy automobiles is improved.

Multi-source data of a rotor surface temperature value Tzz, a temperature value Dz of a stator coil, a resistance value of the stator coil, a voltage value I of the stator coil, a bearing vibration frequency and a bearing rotation speed of the permanent magnet synchronous motor are acquired, calculation accuracy is improved, a magnetic coefficient Tcd, a thermal short-circuit coefficient Dxs of the stator coil and a bearing wear vibration coefficient Zdxs are calculated respectively, and the stability of the permanent magnet synchronous motor is improved. And the fault of the permanent magnet synchronous motor is diagnosed through evaluation and analysis. A fault diagnosis recognition coefficient Sbx is obtained through comprehensive analysis and calculation, key fault parameters of the permanent magnet synchronous motor are comprehensively analyzed, and the fault diagnosis precision is remarkably improved; the fault diagnosis identification coefficient Sbx is analyzed and evaluated and is finally substituted into the unsupervised learning model; an efficient and reliable solution is provided for deep learning of fault diagnosis, operation monitoring and maintenance of the permanent magnet synchronous motor, and the method has important engineering application value and popularization significance.

The invention belongs to the technical field of mechanical fault diagnosis, and discloses a rolling bearing small sample fault diagnosis method and system based on an online soft label Gaussian prototype network. The method comprises the following steps: collecting bearing vibration signals under different working conditions to construct a data set, dividing the data set into a meta-training set and a meta-test set, and splitting the data set into a support set and a query set; constructing a Gaussian prototype network model containing an embedding module, a prototype calculation module and a classification module; in the meta-training stage, soft labels are dynamically generated by adopting an online soft label strategy, and multi-task training is carried out in combination with hard label loss to obtain optimal parameters; in the meta-test stage, a Gaussian prototype is constructed based on a support set, classification is achieved by calculating the Euclidean distance between a sample and the prototype, and prototype parameters are finely adjusted to adapt to cross-working-condition diagnosis when feature distribution drifts. According to the method, label noise interference is effectively relieved, the diagnosis precision can still be ensured under a small number of labeled samples, and the model generalization ability and the diagnosis stability are remarkably improved.

The invention relates to a bearing fault diagnosis method based on wavelet transform and a mixed attention mechanism, and the method specifically comprises the steps: obtaining a bearing vibration signal through a sensor, and obtaining original fault data; sequentially carrying out wavelet transformation, Gaussian noise addition and block sampling on the original fault data; generating a category label and a position label for the sampled data; constructing a fault diagnosis model based on a mixed attention mechanism, a residual structure and learnable convolution; the sampled fault data is divided into a training set and a test set, the training set and the test set serve as input data of the model for training, a training result is monitored in real time for parameter optimization, and meanwhile a fault diagnosis result is visualized; and performing a contrast experiment with the existing model under various noise conditions, and checking the performance of the model. The method provided by the invention solves the problem that a traditional bearing fault diagnosis method is relatively low in fault recognition rate under a variable noise condition.

The invention belongs to the technical field of bearing residual life prediction, and discloses a pyramid attention-based rolling bearing residual life prediction method and system. According to the method, through fast Fourier transform, wavelet transform and time domain statistical feature extraction, bearing vibration signals are processed, and a multi-feature set is constructed. A relationship between the multiple feature set and rolling bearing life prediction is captured. A Weibull distribution loss function is introduced, and convergence of the rolling bearing residual life prediction model is accelerated. Kalman filtering is introduced to carry out smoothing and noise reduction processing on a rolling bearing residual life prediction sequence, and the stability of a prediction result is further improved. The method solves the problems that an existing bearing residual life prediction method is high in operation complexity when the sequence long-time dependency relationship is captured, and the dependency relationships of different time scale ranges are difficult to capture.

The invention relates to the technical field of wind power technologies, in particular to a bearing vibration diagnosis method and system based on a large language model. The method comprises the following steps: acquiring bearing vibration signal data including a fault frequency spectrum signal x, a reference bearing vibration signal and a reference frequency spectrum signal; performing data preprocessing on the acquired bearing vibration signal data; constructing a fault classification model, wherein the step of constructing a feature recognition network and constructing an alignment network to train the constructed fault classification model; and performing fault prediction by using the trained fault classification model. According to the invention, the fault classification network, the feature alignment module and the large language model are utilized to combine the vibration time domain signal and the cue word together, so that the large language model can indirectly process the ultra-long data text.

The invention discloses a bearing fault diagnosis method based on a one-dimensional convolution attention mechanism, and particularly relates to the field of bearing fault diagnosis. The method comprises the following steps: firstly, acquiring original vibration signal data of a rolling bearing, preprocessing the original vibration signal, and carrying out standardization and signal cutting; a bearing vibration signal data set is established, obtained bearing vibration signals are labeled, and bearing vibration signal data are divided into a training set and a test set; then constructing a one-dimensional convolution attention mechanism network model which is formed by connecting a one-dimensional convolution layer, a pooling layer, an efficient additive attention layer, a full connection layer and a Softmax classifier in sequence; training the bearing vibration signal data set by using the constructed network model, and storing the network model with trained parameters; and finally, carrying out fault detection on the bearing vibration signal test set by using the stored network model so as to obtain a fault classification result of the vibration signal. The method is suitable for bearing fault diagnosis.

The invention relates to the technical field of bearing life prediction, in particular to a bearing residual service life prediction method based on a hybrid model, which comprises the following steps: collecting bearing vibration signals under different working conditions; taking the first features as input, constructing a fusion voting mechanism based on a feature screening model, and selecting the first features to obtain second features; sequentially inputting the second feature and the actual life value into a BiGRU network, a GAT network and a full-connection network to obtain a full-connection feature vector; and inputting the full connection feature vector, the feature vector of the BiGRU network and the last time step feature of the bearing vibration signal time sequence into a super network, and inputting the output feature vector of the super network into a full connection layer to obtain a predicted life value. According to the method, the problem that the requirements of precision, generalization and interpretability are difficult to balance by an existing method is solved.