34results about How to "Improve troubleshooting performance" patented technology

The invention provides a transformer partial discharge fault diagnosis method based on hierarchical threshold synchrosqueezed wavelet. The method is characterized by comprising the following steps of: 1, subjecting a transformer partial discharge signal f(t) to SWT decomposition to obtain an SWT decomposition coefficient S<y><a(i)>(t), wherein the i = 1, 2, L, I; 2, estimating a noise standard deviation [sigma]n by using a formula described in the specification; 3, assigning values to ca(i) step by step at a fixed step length for the ith coefficient of the SWT, calculating the minimum mean square error by a mean square error formula, and determining the optimal value of the ca(i); 4, calculating the hierarchical threshold [gamma]a(i) = ca(i) [sigma]n, and achieving the ith SWT coefficient noise reduction through the ith SWT coefficient formula; 5, using a noise reduction formula to reconstruct de-noised layer coefficients to obtain a de-noised de-noising transformer local discharge signal y(t); and 6, performing feature extraction on the de-noised de-noising transformer local discharge signal y(t), and performing fault diagnosis according to extracted features.

The invention discloses a fault diagnosis method, system and device for a permanent magnet synchronous motor as well as a readable medium. The fault diagnosis method for the permanent magnet synchronous motor comprises the following steps: carrying out calculation based on an analytic relationship model from stator current to stator flux linkage to obtain a first set of stator flux linkage values;carrying out calculation based on an analytic relationship model from stator voltage to the stator flux linkage to obtain a second set of stator flux linkage values; solving the difference between the second set of stator flux linkage values and the first set of stator flux linkage values to serve as an initial flux linkage residual; carrying out negative sequence primary and negative sequence secondary synchronous coordinate transformation on the initial flux linkage residual to generate a primary flux linkage residual and a secondary flux linkage residual; and extracting direct current components of the initial flux linkage residual, the primary flux linkage residual and the secondary flux linkage residual, and comparing the direct current components with preset thresholds respectivelyto determine relevant information about permanent magnet faults or sensor faults. Through the application of the above scheme, multiple faults existing in a phase current sensor, a rotor position sensor, a permanent magnet and the like of the permanent magnet synchronous motor can be detected online.

The invention discloses a transformer fault type diagnosis method based on a semi-supervised DBNC (deep belief network classifier). The method comprises the following steps: selecting a sample data set; dividing sample data into an unlabelled pre-training set, a labelled set, a test set 1 and a test set 2; performing state coding on a fault type; building a DBNC-based transformer fault diagnosis model; initializing parameters of each layer of the model; training each RBM at the bottom layer by layer by utilizing comparison divergence; optimizing the whole network parameter through back propogation, so as to enable network classification performance to be globally optimal; and storing a trained network, and verifying classification performance of a network by utilizing sample data of the test set 1. The method disclosed by the invention solves the problems that only a small amount of complete data samples can be obtained under normal conditions when deep learning network fault data is adopted for performing analysis and processing in transformer fault diagnosis, a large number of labelled complete data samples are very difficult to acquire and a large amount of manpower and resources are required.

The invention discloses an analog circuit fault test diagnosis method, which comprises the following steps of: acquiring fault information at each test node of a circuit to be diagnosed by adopting different types of power supply signals as test excitation signals; Determining an optimal test excitation signal and an optimal test node set by using a fuzzy clustering FCM algorithm according to thecollected fault information; Respectively acquiring fault feature information of the circuit in the determined optimal test node set; Carrying out dimension reduction on the fault feature informationby adopting a sparse random projection method, and carrying out normalization processing to obtain a preprocessed fault sample; And carrying out fault diagnosis on the fault sample by using a naive Bayes classifier. According to the invention, the optimal test excitation signal and the optimal test node set can be automatically selected, and the effect of analog circuit fault diagnosis is greatlyimproved. By adopting the sparse random projection method, redundancy and interference components of fault characteristic attribute types are removed, and the data preprocessing time can be saved.

The invention discloses an OLTC mechanical fault diagnosis method based on sample entropy and SVM. The method comprises the steps that step 1, an acceleration vibration sensor is placed on an OLTC topcover to collect vibration signals in various states; step 2, performing EEMD decomposition on the original vibration signal to obtain components IMF, and further processing the first four IMF components; step 3, calculating and selecting a sample entropy of the IMF component; step 4, for the training data set, using the sample entropy obtained through calculation as a feature vector, inputting the feature vector into an SVM for training to obtain an SVM classifier, inputting the SampEn value of the IMF component of the test sample into the SVM classifier, and outputting the SampEn value of the IMF component of the test sample through the SVM classifier to obtain the operation state of the test sample. The working state of the transformer on-load tap-changer can be monitored in real time,the real-time fault diagnosis requirement of the transformer on-load tap-changer is met, data support and theoretical basis are provided for targeted maintenance, and waste of manpower, material resources and time is avoided.

The invention provides a fault diagnosis method and device based on a knowledge graph, equipment and a medium. The method comprises the steps that when a device breaks down, the feature vector of a current fault device is determined based on a device fault knowledge graph, and the device fault knowledge graph is constructed based on fault diagnosis data of historical fault devices; determining the similarity between the feature vector of the current fault device and the feature vector of each historical fault device in the device fault knowledge graph; and determining the diagnosis result of the historical fault equipment corresponding to the highest similarity as the diagnosis result of the current fault equipment, and pushing a solution corresponding to the diagnosis result to the current fault equipment. The accuracy of equipment fault diagnosis can be improved, and the equipment fault diagnosis effect is improved.

The invention discloses an FFCNN-SVM transfer learning fault diagnosis method based on feature fusion under a small sample. The method comprises the following steps: migrating a mature model in a source domain to a target domain through a model migration method in migration learning to form a preliminary model of the target domain; then, by utilizing a convolutional layer and extracting characteristics of picture features, adding the convolutional layer on the preliminary model, then training the preliminary model by utilizing a small amount of scarce sample data provided by the target domain, and forming a target domain shallow model after fitting; replacing a full connection layer of the CNN with the SVM to achieve a classification effect. The method is advantageous in that: through a bearing fault data set, the new fault diagnosis performance of the method can be well verified; experimental results show that compared with other transfer learning methods, the method provided by the invention has a better fault diagnosis effect.

The invention discloses a rolling bearing fault diagnosis method based on an adaptive termination criterion OMP, and the method comprises the steps of calculating the theoretical fault characteristic frequency of a diagnosed bearing through combining the geometric parameters and rotating speed of the diagnosed bearing; analyzing the frequency spectrum of the fault signal, and observing the approximate interval of the high-frequency natural vibration frequency of the bearing; setting a Laplace wavelet parameter set of the initial dictionary; obtaining wavelet frequency and damping which are most matched with fault impact components in the fault signals through a correlation filtering method, and building a needed complete dictionary; using an adaptive termination criterion OMP algorithm to decompose the fault signal, searching the optimal algorithm iteration frequency, and rebuilding a fault impact component in the signal; and analyzing the envelope spectrum of the reconstructed signal, and extracting the fault characteristic frequency in the spectrum to complete fault mode diagnosis. According to the invention, the fault impact component can be reconstructed from the fault signal more accurately, the fault characteristic frequency is extracted, and fault diagnosis of the rolling bearing is realized.

The invention discloses a novel deep feature learning method for planetary gear fault diagnosis. The method comprises the following steps: a, detecting an original vibration signal generated in the operation process of a planetary gear box of electromechanical equipment by using a vibration sensor; b, introducing a sparsity penalty term and a contractibility limit term on the basis of the loss function of the automatic coding machine; c, optimizing specific positions and key parameters of each sparse automatic coding machine and each contraction automatic coding machine in the deep learning architecture by using a quantum ant colony optimization algorithm; d, determining the initial depth of the deep learning architecture and the initial width of each layer by taking the acquired originalvibration signal of the planetary gear box as the input of the novel deep learning architecture. According to the novel deep feature learning method for planet gear fault diagnosis provided by the invention, the data learning capability and the feature extraction robustness can be exerted to the optimal at the same time, and the positions of a sparse automatic coding machine and a contraction automatic coding machine in a deep learning architecture can be actively adjusted.

The invention provides a fault diagnosis method and device for a sensor, and the method comprises the steps: constructing a fault diagnosis model based on a coding and decoding network, adding a data attention mechanism into a coder of the fault diagnosis model, and adding a time attention mechanism and a fusion module into a decoder; and training the fault diagnosis model, determining a fault threshold value of the sensor, and performing fault diagnosis of the sensor based on the fault threshold value. On the basis of a traditional encoder-decoder network (EDN), a novel data-time attention network (DAN) for air conditioning unit sensor fault diagnosis is provided, and the problem that reading of a sensor shows dynamic data-time dependence due to an operation mechanism of an air conditioning unit system, and is prone to being influenced by external factors and control parameters is solved When fault diagnosis is carried out based on the model, the method is more sensitive to deviation faults, and the fault diagnosis performance of the sensor is greatly improved.

The invention discloses a fan fault transferable diagnosis method based on data enhancement and a capsule neural network. The method comprises the following steps: preprocessing collected fan vibration signal data, and detecting and eliminating abnormal values; extracting an optimal characteristic frequency band of the fault based on the average power spectral density; calculating the average power spectral density intensity value of the fan vibration signal on the optimal fault characteristic frequency band, and taking the average power spectral density intensity value as the input of a first-class support vector machine to carry out fault degradation detection so as to determine the initial failure occurrence point of the fault; the vibration signals are re-divided into fault data and normal data according to failure points, the data are labeled, and a training data set is constructed; initializing network hyper-parameters of the capsule neural network, and training the network hyper-parameters; and inputting a new vibration data signal into the trained network to obtain a diagnosis result. According to the method, the fault samples are effectively expanded through data enhancement, and the accuracy and mobility of model fan fault diagnosis are improved based on the multi-dimensional rich features extracted by the capsule neural network.