20 results about "Recurrence plot" patented technology

The invention discloses an online intelligent monitoring and analyzing system for wear of a circuit breaker contact, which relates to the field of contact wear monitoring and comprises a visual module, an acoustic module, a current module, an acquisition module, an analysis module, an alarm module, a processing module and a display module. The method comprises the following steps of: analyzing and establishing a quantum driving feature selection model; constructing a contact motion mechanical control model; setting parameter prior distribution; fusing a likelihood function and prior by utilizing a Bayesian theorem; a vibration chaos non-linear loss value is calculated based on recurrence plot line segment probability distribution analysis, a jump diffusion model is constructed, the number of jumps is counted to calculate Poisson intensity, corresponding display and alarm are carried out, the accuracy of contact abrasion state evaluation is guaranteed, and early abrasion abnormity is found in time.

The invention discloses a cable partial discharge mode identification method and device based on multi-mode fusion, electronic equipment and a storage medium, and belongs to the technical field of cable partial discharge online monitoring. And converting the original signal into a multi-modal spectrum through phase resolution analysis, wavelet packet energy analysis, skewness and kurtosis statistics and recurrence plot analysis. The group of maps are input into a multi-modal fusion recognition model, the model analyzes the spatial distribution characteristics and the time sequence characteristics of the maps at the same time through parallel vision and time sequence coding modules, and two analysis results are subjected to weighted fusion, so that final classification recognition of the discharge mode is realized. Through the implementation of the method and the device, the problems that key information is lost due to dependence on single-dimensional feature representation and the identification accuracy is limited due to the fact that an existing identification model cannot analyze the spatial distribution characteristic and the time sequence evolution characteristic of the signal at the same time in the prior art can be solved.

The present invention discloses a method for generating a rotor fault diagnosis model based on multiple transform domains and a diagnosis method. The former includes: collecting vibration signal data of a modular rotor fault of a gas turbine in multiple categories to construct a training and test data set; transforming each data to the short-time Fourier time-frequency domain using the STFT transform domain, transforming it to the Morlet wavelet time-frequency domain using the Morlet wavelet transform domain, and transforming it to the complex network phase space domain using the complex network recurrence plot method; inputting the training data set transformed by the three transform domains into a preset model, and using the CNN network layer for feature extraction; using the LSTM network to perform feature structure learning on the output features, and training to obtain a rotor fault model; using the test data set transformed by the three transform domains to verify and adjust the rotor fault model to obtain a rotor fault diagnosis model for fault detection. The present invention can realize automatic data processing, reduce manual input, and improve the efficiency and accuracy of fault diagnosis.

The invention discloses a marine diesel engine fault diagnosis method and system, and the method comprises the steps: collecting state signals of a marine diesel engine in different states, and carrying out the grouping of the state signals; obtaining a multi-modal signal matrix based on the grouped state signals; obtaining a normal data set and an abnormal data set based on the multi-modal signal matrix; and completing fault diagnosis based on the normal data set and the abnormal data set. According to the method, the targets of denoising, dimensionality reduction and sufficient fault feature extraction are achieved. Specifically, the symplectic geometric decomposition provides deep understanding of a data structure, the RPCA removes noise and realizes dimension reduction, and the recurrence plot analysis helps to identify and extract fault features through visualization and analysis of dynamic characteristics of data. The comprehensive analysis process has remarkable advantages in processing complex time sequence data.

The invention provides a multi-feature fusion asymmetric time series data depth clustering method, which comprises the following steps of: inputting time series data into an encoder to obtain a first feature, converting the time series data into a recursion plot RP, inputting the recursion plot RP into a convolutional layer, converting the dimension of an output value into the dimension which is the same as the dimension of the first feature to obtain a second feature; fusing the first feature and the second feature through a fusion layer; sampling a fusion layer to obtain a mean value and a variance, randomly sampling from N (0, 1) to obtain epsilon, obtaining embedded layer data, constructing a loss function of an embedded module, predicting the probability that each sample belongs to each cluster, obtaining model prediction, and constructing a target variable based on a prediction model; empirical distribution of the target variables is defined, and KL divergence between the empirical distribution and uniform distribution is minimized; constructing a loss function of a regularization clustering layer, and combining the loss function of the embedded module to obtain a model loss function; training a clustering model based on the loss function and the target variable; and constructing a clustering model based on the new target variable.

The invention discloses a running-in and wear dynamic behavior characterization method based on a gray texture recurrence plot, and relates to the technical field of complex nonlinear signal processing, and the method comprises the steps: S1, carrying out the preprocessing of a friction signal outputted in a running-in and wear process, obtaining a pure friction signal, and storing the pure friction signal in a computer in a time sequence format; s2, setting a window length, and decomposing all the friction signal time sequences into a plurality of continuous and non-overlapped calculation windows; s3, time delay and embedded dimension parameters are calculated for the friction signals in each window, the frequency of two sets of parameters in all the windows is counted, the parameter with the highest occurrence frequency is selected as the same parameter, and phase-space reconstruction is conducted on the friction signal time sequence in all the windows. According to the macroscopic mode and quantitative parameters, the dynamic characterization research of the running-in wear behavior is carried out from the tribology-dynamics coupling angle, and a new perspective is provided for running-in state identification.

The invention provides a structural damage positioning method and system based on wavelet-PCA, and relates to the technical field of structural damage positioning. The method comprises the following steps: acquiring a structure vibration response signal through a sensor to obtain an original signal data set; de-noising the signal by adopting wavelet transform, and extracting a principal component characteristic matrix through principal component analysis; performing phase-space reconstruction on the principal component time sequence to generate a high-dimensional power system track; calculating quantization factors such as recursion rate, certainty and average diagonal length based on recursion plot analysis, and forming a recursion quantization feature matrix; generating a spatial correlation distribution diagram by analyzing the correlation between the recursion characteristics of different sensor positions; and realizing damage positioning according to the correlation abnormal region. The early weak damage can be effectively detected, the defect that a traditional linear analysis method is insensitive to nonlinear characteristics is overcome, and the method has the advantages of being high in anti-interference performance and accurate and reliable in positioning.

The application discloses a kind of based on phase space reconstruction's control loop regulating valve stickiness detection method, it is related to valve fault diagnosis field.The method of the present application is the nonlinear signal analysis method of unsupervised, the method combines phase space reconstruction, recursive analysis and power spectrum analysis and other technologies, proposes stickiness statistical characteristic index and stickiness distribution characteristic index, to characterize the nonlinear and periodicity law brought by stickiness, only by loop parameter realizes the unsupervised regulating valve stickiness fault detection.The method proposed effectively extracts the stickiness characteristics in recurrence plot, not only simple and convenient to calculate, but also has high interpretability and accuracy, shows the great potential of recursive analysis in valve stickiness detection, provides a method basis for the online monitoring and performance evaluation of industrial loop.

The invention relates to the technical field of mine safety detection, in particular to a mine safety detection method and system based on the Internet of Things, and the method comprises the steps: obtaining one-dimensional data during a mining period, carrying out time dimension compression, extracting local time sequence features, generating an initial time sequence code, and converting the initial time sequence code into a two-dimensional similarity image through recurrence plot transformation; time features are extracted from the initial time sequence codes, and space features are extracted from the two-dimensional similarity images; the weight corresponding to the spatial feature is expanded to the same shape as the time feature through time sequence dimension expansion, the time feature is enhanced through element-by-element multiplication, and the time feature and the original time feature are added to obtain a cross enhanced time feature; the weight corresponding to the time feature is subjected to spatial dimension expansion to be in the same shape as the spatial feature, the spatial feature is enhanced through element-by-element multiplication and added with the original spatial feature, the cross-enhanced spatial feature is output, and a corresponding detection result is further obtained by using a full-connection network. And through high-dimensional space-time modeling and dynamic feature fusion, balance between real-time performance and accuracy is realized.

The invention discloses a watermelon maturity nondestructive testing method and system based on a water immersion density method, particularly relates to the technical field of physical quantity precision measurement and nondestructive testing, and aims to solve the problem that the density calculation precision is influenced by inaccurate liquid level height data caused by dynamic process interference in existing automatic immersion measurement. The method comprises the following steps: firstly, acquiring basic weight and liquid level height data during no-load and watermelon bearing, then controlling watermelon immersion, synchronously acquiring a driving signal and a liquid level height change sequence, then identifying a dynamic stage of an immersion process, and analyzing and judging the consistency of driving and liquid level response dynamics by using a recurrence plot. Multi-scale complexity and long-range correlation correlation analysis is carried out on the data subsets tending to the balance stage, whether the liquid level reaches a physical stable state or not is intelligently judged, and finally the watermelon density is accurately calculated by combining stable liquid level height data with weight data only under the condition that it is judged that the watermelon density is stable. The immersion dynamic interference is effectively filtered out, and the density measurement accuracy is remarkably improved.

The invention provides a user type identification method and system based on multi-modal mixed features. The method comprises the following steps: deconstructing time sequence electric quantity data into a plurality of typical power consumption curves according to seasons and holidays; on the basis of the obtained typical power consumption curves, statistical features are extracted and include the electricity price sensitivity coefficient, the power consumption stability and the statistical features combined with the shape distance of the typical power consumption curves, and user type labels are generated through clustering; performing two-dimensional processing on the time sequence electric quantity data based on a Grubrum angle field and a recurrence plot, fusing statistical features, and constructing a user classification recognition model fused with multi-modal mixed features; and inputting the time sequence electric quantity data of the identified user into the constructed model to obtain a user type.

A structure estimation method (S1) comprising: a step (S12) for generating a recurrence plot from information about the structures of spatially adjacent moieties of a molecule by representing the primary sequence of the molecule on the time axis; a step (S13) for generating a distance matrix that provides the distance between any two vertices of the recurrence plot; a step (S14) for reconstructing time-series data by using multidimensional scaling with respect to the distance matrix; and a step (S15) for reconstructing the three-dimensional structure of the molecule from the reconstructed time-series data, wherein the distance matrix has, as a component, a post-transformation distance obtained by transforming a Jaccard distance related to the recurrence plot, and the post-transformation distance is obtained as a solution or an approximate solution of an equation in which said post-transformation distance is defined as a variable.

The invention discloses a water outlet control method of an intelligent faucet and a faucet body, particularly relates to the technical field of fluid temperature control and valve adjustment, and is used for solving the problem of water outlet temperature fluctuation caused by mechanical nonlinear characteristics of a mixing valve in existing constant temperature control. A displacement-flow response dead zone is accurately recognized by monitoring retention characteristics of a temperature deviation phase track in real time and combining symbolized statistical analysis and recurrence plot determinacy evaluation of a flow signal, and a targeted compensation instruction is generated according to the moving direction of a valve element and the range of the dead zone, so that accurate compensation adjustment of the opening degree of the mixing valve is achieved. Therefore, temperature fluctuation is effectively restrained, and the stability and control precision of the outlet water temperature are improved.

The invention discloses an on-line test method and system based on an electronic tag, particularly relates to the technical field of product identification analysis, and is used for solving the problem of space-time dislocation between an electronic tag reading event and a test data acquisition event caused by movement of a tested object in a continuous movement conveying scene. Motion state information of a measured object is acquired in real time, phase-space reconstruction is performed on the measured object, recurrence plot quantitative characteristics are calculated to evaluate a deterministic chaotic disturbance level, accordingly, a dominant oscillation mode is extracted, and an instantaneous phase of the mode at a label reading and data acquisition event moment is acquired; a phase-locked loop is configured by using a chaotic disturbance level, whether a phase locking state is reached is judged by analyzing phase difference evolution between two phases, and when the phase locking state is reached, a tag identifier is analyzed to call and execute a corresponding test program. And collecting test data, calculating an actual physical point location associated with the data based on the motion state information and the event moment, and finally generating a test data file.

This invention discloses a method and system for high-frequency transformer fault identification, belonging to the field of high-frequency transformer fault detection technology. The method comprises: acquiring a characteristic gas vector sequence of the high-frequency transformer and converting the characteristic gas vector sequence into an actual fault recurrence graph; generating an expanded dataset based on the actual fault recurrence graph, a pre-acquired random noise vector, and a preset initial depth convolutional generative adversarial network model; training a preset initial MobileNetV3 model based on the expanded dataset and the characteristic gas vector sequence and constructing a cross-entropy loss function, then updating the model with the goal of minimizing the cross-entropy loss function until a preset convergence condition is met, thus obtaining the MobileNetV3 model; inputting the fault recurrence graph to be tested into the MobileNetV3 model to obtain the fault detection result of the high-frequency transformer to be detected. Therefore, by implementing this invention, the low accuracy of high-frequency transformer fault identification and the insufficient ability to identify complex faults can be avoided, achieving high-precision and high-reliability identification of various types of high-frequency transformer faults.

The invention discloses an all-weather atmospheric temperature, humidity, wind and rain profile combined detection method and system, particularly relates to the technical field of meteorological radar signal processing, and is used for solving the problem of motion parameter inversion failure caused by phase nonlinear distortion of an existing coherent Doppler radar at a dielectric mutation interface. A radar original echo signal is obtained; extracting a phase change sequence and calculating a phase difference value of adjacent range gates; identifying a dielectric abrupt change interface location when the differential value exceeds a linear threshold range; analyzing the continuous period interface position recurrence plot, and extracting diagonal characteristics as a deterministic coefficient and vertical section distribution as a laminar flow parameter; determining a dynamic stability level based on the two, and executing path integral compensation when the dynamic stability level is lower than a threshold value; an original signal phase is corrected through a compensation strategy, high-precision motion parameters are output through coherent accumulation, dynamic compensation of phase distortion in a non-uniform medium is achieved, and robustness of atmospheric parameter inversion is guaranteed.

The application discloses a mechanical equipment machining tool feed speed control system and belongs to the technical field of machining control. The system comprises a signal acquisition module, a chaotic analysis module, a recursive adjustment module and an optimization control module. The signal acquisition module acquires cutting force signals and tool vibration signals in real time, performs cooperative phase space reconstruction on the cutting force signals and the tool vibration signals, and obtains high-dimensional phase trajectories. The chaotic analysis module calculates the correlation dimension and the Kolmogorov entropy of the high-dimensional phase trajectories, and judges the chaotic degree of a cutting state according to the change rates of the correlation dimension and the Kolmogorov entropy. When the chaotic degree exceeds a set threshold, the recursive adjustment module extracts the recurrence rate of the cutting force signals based on a recurrence plot, and dynamically adjusts a feed speed reference value. Meanwhile, the tool vibration signals are subjected to local mean decomposition, a cutting chatter characteristic vector is constructed, and the cutting chatter characteristic vector is input into an extreme learning machine model to output a chatter risk grade. The optimization control module updates a feed speed control increment by using an adaptive momentum algorithm, fuses speed smoothness constraints, solves an optimal feed speed by using a Lagrange multiplier method, and converts the optimal feed speed into a servo driving instruction output.

The present invention improves the versatility of a feature amount in time series data. A feature amount extraction device (20) uses an inference model to extract a feature amount from a non-binary recurrence plot. In the non-binary recurrence plot, for each of all combinations of two samples selected from a plurality of samples included in the time-series data, a distance scale between the two samples is set to two-dimensional image information at positions determined by the order of one of the two samples and the order of the other in the time-series data. The inference model is constructed by machine learning which uses, as learning data, a non-binary recurrence plot generated from time-series data for learning.

The invention belongs to the technical field of hydroelectric generating set operation monitoring, and particularly discloses a hydroelectric generating set vibration trend prediction method and device, electronic equipment and a medium. The method comprises the steps of performing data preprocessing on throw monitoring data of a to-be-detected hydroelectric generating set, and determining throw time sequence data representing state characteristics of the set; performing modal decomposition on the throw time sequence data to obtain a plurality of intrinsic mode function components; performing phase-space reconstruction and recursive analysis on each intrinsic mode function component to obtain recursive plot information of each intrinsic mode function component; and inputting the throw time sequence data, each intrinsic mode function component and the corresponding recurrence plot information into a unit runout trend prediction model to obtain a runout trend prediction result, output by the unit runout trend prediction model, of the hydroelectric generating set to be tested. Through the method and the device, the accuracy and the effect of hydroelectric generating set vibration trend prediction can be effectively improved, and accurate prediction of the hydroelectric generating set vibration trend is realized.

The application discloses an online testing method and system based on electronic tags, and particularly relates to the technical field of product identification analysis, which is used to solve the problem of time-space dislocation between electronic tag reading events and testing data collection events caused by the movement of the measured object in a continuous motion conveying scene; the motion state information of the measured object is acquired in real time, the motion state information is reconstructed in phase space, and the quantitative features of the recurrence plot are calculated to evaluate the level of deterministic chaos disturbance, according to which the dominant oscillation mode is extracted, the instantaneous phase of the mode at the time of the tag reading and data collection events is acquired, the phase-locked loop is configured by using the level of chaos disturbance, whether the phase-locked state is reached is determined by analyzing the phase difference evolution between the two phases, when the locked state is reached, the tag identification is analyzed to call and execute the corresponding testing program, the testing data is collected, and the actual physical point based on the motion state information and the event time is calculated, and finally, the testing data file is generated.