264results about How to "Improve the accuracy of fault diagnosis" patented technology

A dynamic layer diagnostic device for smart power grid fault comprises a data collection and monitoring unit, a data processing unit, a data base unit, a communication unit and a man-machine interaction unit. A dynamic layer diagnostic method includes: when a smart power grid breaks down, calculating fault diagnosis starting conditions to conform a diagnosis strategy, wherein the fault diagnosis strategy comprises switch layer diagnosis, feeder layer diagnosis and transformer substation layer diagnosis; starting the switch layer diagnosis when changes of switch motion information are remarkable before and after the fault; starting the transformer substation diagnosis when changes of electricity amount information are remarkable before and after the fault; stopping the diagnosis when a fault element is the only one element during the diagnosis of the switch layer; otherwise, retrieving the switch historical action recording, and starting the transformer substation diagnosis when recording matched with the current switch action exists; and otherwise, starting the feeder layer diagnosis. The dynamic layer diagnostic method performs layering analysis on the fault, fully utilizes various fault information and improves fault diagnosis accuracy according to different characteristics of multisource information after the power grid fault and difficulty layer in obtaining and processing of various information.

The invention discloses a rolling bearing probabilistic intelligent fault diagnosis method based on adaptive MRVM. The method comprises the steps that the original fault data of a rolling bearing are measured through an acceleration sensor; a vibration signal is segmented, and wavelet packet energy characteristics are extracted; principal component analysis and dimension reduction are used for normalization simultaneously; a training sample set and a test sample set are processed and divided; an algorithm is used to adaptively select nuclear parameters; the training sample set is used to train and test a multi-class correlation vector machine; and the test result is compared with the actual fault type to acquire the validity of a diagnosis model. According to the invention, the method overcomes the defect that a traditional intelligent fault diagnosis method cannot output the fault probability value; the fault diagnosis accuracy of the rolling bearing is improved; more fault type determining information of the rolling bearing can be provided; through the fault type probability value provided by the invention, the state of the rolling bearing can be further assessed; and method has the advantages of good engineering value and application prospect.

The invention provides network fault diagnosis methods and systems. A gradient boosting decision tree classifier prediction model is trained through adoption of a symptom data set and a fault data setin network history data. Network fault diagnosis is carried out through utilization of the trained gradient boosting decision tree classifier prediction model. Network fault diagnosis precision can be effectively improved. Network fault diagnosis time can be effectively reduced. The method and the system are applicable to diversified production scenes.

The invention discloses a noise diagnosis algorithm for rolling bearing faults of rotary equipment. Firstly, a sound pick-up device collects running noise signals of a rolling bearing, and the signalsare subjected to preliminary fault judgment through a bearing normality and anomaly pre-classification model based on an anomaly detection algorithm; secondly, according to a fault pre-judgment result, the abnormal signals (the faults occur) pass through a neural network filter to filter normal components in the signals of the bearing, the output net abnormal signals are connected to a subsequentfeature extraction module, and the normal signals (no faults occur) are directly connected to the feature extraction module; the feature extraction module extracts Mel-cepstrum coefficients (MFCC) ofthe signals to serve as eigenvectors, feature reconstruction is carried out by utilizing a gradient boosted decision tree (GBDT) to form composite eigenvectors, and principal component analysis (PCA)is used for carrying out dimensionality reduction on features; and finally, feature signals are input into an improved two-stage support vector machine (SVM) ensemble classifier for training and testing, and at last, high-accuracy fault type diagnosis is achieved. According to the algorithm, the bearing faults can be effectively detected and relatively high fault identification accuracy is kept;and the algorithm has relatively high effectiveness and robustness for detection and classification of the bearing faults.

The invention discloses a LightGBM fault diagnosis method based on improved Bayesian optimization. The LightGBM fault diagnosis method comprises the following steps: 1) determining hyper-parameters needing to be optimized by a LightGBM model and a hyper-parameter value range; 2) improving the Bayesian optimization algorithm to obtain an improved Bayesian optimization algorithm GP-ProbHedge; 3) selecting an optimal hyper-parameter combination of the fault diagnosis model by using the method in the step 2) in combination with a five-fold cross validation mode; and 4) constructing an improved Bayesian optimization LightGBM fault diagnosis model, and giving a model iteration process and an optimization result. By adopting the technology, compared with the prior art, according to the invention,an improved Bayesian optimization algorithm is provided to carry out optimization selection on parameters of a fault model; by improving an acquisition function of a traditional Bayesian optimizationalgorithm and a covariance function of a Gaussian process of the traditional Bayesian optimization algorithm, an improved Bayesian optimization LightGBM fault diagnosis method is provided, and equipment faults are diagnosed and predicted.

The invention provides a portable bearing fault diagnosis device and method based on vibration detection. The portable bearing fault diagnosis device comprises a vibrator sensor and a portable bearing fault diagnosis instrument, wherein the portable bearing fault diagnosis instrument comprises a shell and a circuit device; the circuit device comprises an A/D (Analogue/Digital) conversion module, a DSP (Digital Signal Processor) data analyzing module, an ARM (Advanced RISC Machines) data processing module, a storage module, a display module and a power supply module; the A/D conversion module is used for uploading an acquired and processed signal of a vibration sensor to the DSP data analyzing module for analysis and fault judgment; the ARM data processing module is used for receiving a processing result and distributing storage and display. The diagnosis method mainly comprises the following steps: a vibration signal is obtained by the vibration sensor; the signal is acquired by the A/D conversion module; the DSP data analyzing module is used for analyzing and processing; a fault analyzing result is stored and displayed. The portable bearing fault diagnosis device is simple and rapid to operate, and has high diagnosis precision and high automation degree; the real-time monitoring and the fault diagnosis of a rolling bearing fault can be effectively realized.

The invention discloses a turbo generator set vibration fault diagnosis method based on forward reasoning. After a diagnosed turbo generator set starts, a parameter detection device is employed to carry out real-time detection on related work parameters in a start and operation process of the diagnosed turbo generator set and synchronously sends the detection information to a vibration fault diagnosis device for automatic vibration fault diagnosis, vibration faults of the diagnosed turbo generator set are diagnosed level by level by the vibration fault diagnosis device according to the detection result transmitted by the parameter detection device in combination with the active power value of the diagnosed turbo generator set, and the fault diagnosis process comprises steps of 1, shaft vibration swinging value diagnosis; 2, start process vibration fault diagnosis; 3, zero load operation vibration fault diagnosis; and 4, loaded operation vibration fault diagnosis. The method is advantaged in that the steps are simple, and the method is reasonable in design, is convenient to realize, has good use effects, can conveniently and rapidly accomplish the online steam turbine vibration fault diagnosis process and can further realize accurate and reliable diagnosis results.

The invention discloses a locomotive motor bearing automatic diagnostic method based on a vibration acceleration signal, comprising the steps of: acquiring a vibration acceleration signal and a speed key phase signal of a motor bearing; performing order tracking and constant-angle incremental sampling, and converting the vibration acceleration signal into an angular-domain vibration acceleration signal; calculating an FFT spectrum and an EEVS enhanced envelope spectrum; setting effective frequency bands in the FFT spectrum and the EEVS enhanced envelope spectrum according to a bearing fault frequency; calculating the fault probability of each component of the bearing according to the band energy in the FFT spectrum and the EEVS enhanced envelope spectrum; setting weights and calculating the weighted probability of each component of the bearing; comparing a preset fault reference probability with the weighted probability of the current bearing so as to automatically determine a fault type and representing the severity of the fault by the weighted probability.

The invention provides a multi-fault diagnosis method of a spacecraft attitude control system. The multi-fault diagnosis method comprises the steps: establishing a spacecraft attitude dynamic model and an attitude kinematics model; establishing a spacecraft augmented attitude dynamics model and an augmented attitude kinematics model based on the above models; designing an RBF neural network interference observer based on the augmented attitude dynamics model and the augmented attitude kinematics model, and estimating an interference value in the system; designing a robust fault detection observer, calculating a residual evaluation value, and selecting a fault detection threshold; according to the residual evaluation value and a fault detection threshold value, determining fault occurrenceparts comprising an actuator, a gyroscope, an attitude sensor or a plurality of parts; designing an RBF neural network fault isolation observer, reconstructing actuator faults, gyroscope faults and attitude sensor faults on the three axes, and selecting a fault isolation threshold; and finally determining a specific part with a fault according to the reconstructed fault value and the fault isolation threshold, thereby finishing multi-fault diagnosis.

The invention discloses an intelligent fault diagnosis method under a small sample based on an attention mechanism element learning model. According to the intelligent fault diagnosis method, an attention mechanism and a meta-learning method are used for establishing an association network model; short-time Fourier transform is carried out on mechanical signals to obtain a time-frequency spectrogram of the mechanical signals; feature extraction and operation state recognition are further carried out on the time-frequency spectrogram; and rich fault information contained in the mechanical signals can be effectively mined. According to the intelligent fault diagnosis method, a pseudo distance can be trained adaptively to evaluate the similarity between related data; clear mathematical formula definition is not needed; and high mechanical equipment fault diagnosis accuracy can be obtained. Therefore, the dependence of a feature extraction process on artificial experience and the dependence of an existing intelligent fault diagnosis algorithm on a large amount of training data in a traditional diagnosis method are eliminated, and the problem of mechanical equipment fault diagnosis under the condition of small sample data is practically solved.

The invention discloses a mechanical fault diagnosis method and system based on TJM transfer learning. According to the method, CEEMDAN decomposition is introduced. The algorithm calculation amount isreduced while the mode mixing problem is solved, and meanwhile the problems that according to a traditional machine learning method, when training and testing data distribution has a certain degree of difference, the established classification model is poor in popularization capacity, and even sometimes the classification model cannot be universally used are solved through a transfer learning method. And meanwhile, the problem of low fault diagnosis efficiency caused by data difference between different working conditions of the rotary machine is solved, and the problems that the fault stateis incomplete and the fault diagnosis cannot be correctly and completely carried out due to insufficient data acquisition quantity of the rotary machine in some working states are also solved. According to the method, the characteristic that cross-domain feature matching and instance reweighting are jointly executed in the TJM transfer learning method is utilized, the problem that the recognitionand diagnosis rate is not high due to the fact that the data difference between the source domain and the target domain is large is solved to the maximum degree, and the fault diagnosis precision is greatly improved.

The invention discloses a rolling bearing health assessment and fault diagnosis method and a monitoring system. The problem that in the prior art, a large amount of known data is needed or too much human experience intervention is needed to guarantee the monitoring effect is solved, and the effect of accurately detecting and recognizing bearing faults through online real-time analysis on bearing vibration signals is achieved. According to the technical scheme, the method comprises the following steps: acquiring a vibration signal of a bearing, and processing the vibration signal to obtain a spectrogram; establishing a graph model for the spectrogram; performing similarity comparison on adjacent matrixes generated by the graph model to calculate an abnormity degree, and making a decision onan abnormity degree index; setting a threshold value to carry out hypothesis inspection, and carrying out fault inspection on the bearing; And carrying out fault diagnosis when a bearing signal has afault.

The invention discloses a gear case fault diagnosis method based on blind source separation and relates to a mechanical fault diagnosis method, the method is applied to the blind source separation technology for solving the gear case fault diagnosis problem. The blind source separation technology is applied to the gear case diagnosis based on vibration analysis as the main tool for pre-processing the signal and extracting the fault feature and is capable of greatly strengthening the fault information, changing the conventional fault information strengthening thought by taking the noise reducing process as the main part, raising the diagnosis precision and solving the problems that the fault is difficult to locate and the early fault diagnosis rate is low. The blind source separation technology is applied to the gear case diagnosis based on vibration analysis as the main tool for pre-processing the signal and extracting the fault feature and is capable of greatly strengthening the fault information. The conventional fault information strengthening thought by taking the noise reducing process as the main part is changed for raising the diagnosis precision and solving the problems that the fault is difficult to locate and the early fault diagnosis rate is low.

The invention provides a rotary machine fault diagnosis method based on a depth automatic encoder, and aims to improve the fault diagnosis precision of a rotary machine, and the method comprises the steps: firstly collecting a vibration acceleration time domain signal of the rotary machine, and obtaining a training data set and a test data set; secondly, for each activation function, training a series of depth auto-encoders through different training sets by using a K-fold cross validation method; secondly, verifying the trained depth auto-encoder through a verification set, and obtaining theprecision of each fault label; thirdly, searching an optimal selection parameter by adopting a grid search method, screening the depth auto-encoder through the optimal selection parameter, and constructing an integrated depth auto-encoder model; and finally, obtaining a prediction label of the input sample, and mapping the prediction label back to the fault type of the rotating machine to realizefault diagnosis of the rotating machine.

The invention discloses an online fault diagnosing method for Fast RVM (relevance vector machine) sewage treatment. The method includes the steps of firstly, removing samples with incomplete attributes in sewage data, normalizing the samples into a [0, 1] interval, and determining a historical data set and an updating test set; secondly, using a relevance vector machine method based on clustering to compress the majority data of the historical data set; thirdly, using a virtual minority upward sampling method to extend the minority data of the historical data set; fourth, building a 'one-to-one' fast relevance vector machine multi-classification training model; fifthly, adding new samples from the updating test set into the model for testing, and updating the historical data set; sixthly, returning to the second step, reprocessing unbalanced historical data, training the model, and repeating the above process until online data testing is finished. By the online fault diagnosing method, the unbalance of the sewage data is lowered effectively, classification accuracy is increased, online updating speed is increased, operation faults can be diagnosed in real time, and the safety operation of a sewage treatment plant is guaranteed.

The invention discloses a bearing fault diagnosis method based on an improved ant lion algorithm and a support vector machine. The bearing fault diagnosis method comprises the following steps that vibration acceleration signals under a typical fault state are collected; the collected signals are extracted to obtain a data sample of a typical fault type; based on the improved escape mechanism ant lion optimization algorithm and the support vector machine, a bearing fault diagnosis model is established; the data sample is input into the bearing fault diagnosis model to optimize the bearing faultdiagnosis model; and bearing fault diagnosis is performed based on the optimized bearing fault diagnosis model. According to the bearing fault diagnosis method based on the improved ant lion algorithm and the support vector machine, the improved EALO algorithm is provided by introducing an escape mechanism and adaptive convergence conditions to improve the optimization performance of the algorithm, and the improved ant lion algorithm is combined with the support vector machine to realize the bearing fault diagnosis, so that important theoretical significance and practical value are achieved on improving the rolling bearing fault diagnosis accuracy, ensuring the safety of rolling bearings and stabilizing the operation.