6 results about "Multidimensional scaling" patented technology

The invention discloses a beam moving gantry crane winch fault diagnosis method, and relates to the technical field of fault diagnosis, and the method comprises the steps: carrying out the wavelet decomposition and reconstruction of a gantry crane winch bearing monitoring signal, and generating denoising data; respectively inputting the de-noised data into a first feature extraction path and a second feature extraction path, generating a symbolized embedded vector by the first feature extraction path through standardization, symbolization and phase-space reconstruction processing, and calculating a relative entropy feature value of mode distribution; the second feature extraction path calculates a yoke divergence matrix and generates a dependency feature value by constructing an original embedding vector and a delay embedding vector; constructing two-dimensional feature vectors, and calculating the Euclidean distance of the feature vectors between the samples to generate a distance matrix; performing multi-dimensional scale transformation on the distance matrix to generate low-dimensional space distribution coordinates of the sample; and determining a fault category based on distance matching between the low-dimensional space distribution coordinates and a known fault category space region. And accurate diagnosis of different faults is realized.

This application provides a method and apparatus for prototyping visualization based on local singularities. The method includes: acquiring a detrital zircon U-Pb age dataset of a target region; preprocessing the dataset and dividing it into age intervals; performing local singularity analysis on each age interval to calculate the singularity index of each age interval and constructing a local singularity spectrum; performing multidimensional scaling (MDS) analysis based on the singularity index to obtain visualization results in a low-dimensional space; and interpreting the visualization results in conjunction with geological background information to determine prototyping relationships and material migration paths. This application can accurately identify fine prototyping paths missed by traditional methods, significantly improving the precision and reliability of prototyping.

The invention provides a multi-dimensional scale ambiguity-free distance measurement positioning method for measuring incoming wave time difference, and belongs to the technical field of wireless positioning, and the method comprises the steps: measuring the time difference of arriving at different nodes by using a radio signal in an environment, constructing a matrix and a vector for removing rotation and mirror image ambiguity, converting a nonlinear constraint into a solvable model, and carrying out the measurement of the incoming wave time difference. A double fitting error search mechanism is designed, through rotation angle lattice point search, the minimum fitting error of an original image and a mirror image is calculated and compared, the mirror image and rotation blur are judged and eliminated, and therefore the node positioning result after deblurring is determined. According to the scheme, the absolute coordinates of the nodes can be effectively solved without depending on satellite positioning, the positioning error is small, the self-positioning problem of the nodes of the wireless sensor network in the complex shielding environment is remarkably solved, and the self-positioning level of the nodes in the complex environment is improved.

The invention provides a track circuit compensation capacitor feature extraction method based on multi-scale yoke difference correlation, and the method comprises the steps: carrying out the phase-space reconstruction of monitoring signal data of a track circuit compensation capacitor monitoring point, obtaining an embedded vector data set, selecting a sample pair from the embedded vector data set, calculating a yoke difference divergence matrix of the sample pair, and carrying out the extraction of a track circuit compensation capacitor feature. Based on a yoke divergence matrix, calculating a synchronism index, traversing all monitoring signal samples, calculating a synchronism index between each pair of samples, generating a correlation matrix, inputting the correlation matrix to a multi-dimensional scaling algorithm, calculating a corresponding scalar product matrix, performing eigenvalue decomposition on the scalar product matrix, and obtaining a multi-dimensional matrix; and obtaining a spatial distribution matrix of all the monitoring signal samples, and performing comparison and matching based on the spatial distribution matrix of the labeled monitoring data samples and the spatial distribution matrix of the non-labeled monitoring data samples to realize fault identification and diagnosis of the track circuit compensation capacitor. The method has the advantages of high early fault feature extraction identification degree and high fault identification accuracy.

The application provides a multi-dimensional scale ambiguity-free ranging positioning method for measuring the time difference of incoming waves, and belongs to the technical field of wireless positioning. The method utilizes radio signals in the environment, measures the time difference of the radio signals arriving at different nodes, constructs a matrix and a vector for removing rotation and mirror image ambiguity, converts nonlinear constraints into a solvable model, and designs a double fitting error search mechanism. Through rotation angle grid search, the minimum fitting errors of the original image and the mirror image are calculated and compared, the mirror image and the rotation ambiguity are determined and eliminated, and the node positioning result after removing the ambiguity is determined. The scheme of the application can effectively solve the absolute coordinates of the nodes without relying on satellite positioning, has small positioning error, significantly solves the self-positioning problem of the wireless sensor network nodes in a complex shielding environment, and improves the self-positioning level of the nodes in the complex environment.

The invention provides a source analysis visualization method and device based on local singularity, and the method comprises the steps: obtaining a chipping zircon U-Pb age data set of a target region, carrying out the preprocessing of the data set, and dividing the data set into age intervals; performing local singularity analysis on each age interval, calculating to obtain a singularity index of each age interval, and constructing a local singularity spectrum; performing multi-dimensional scale analysis (MDS) based on the singularity index to obtain a visualization result in a low-dimensional space; and in combination with geological background information, interpreting the visualization result to determine a material source relationship and a material migration path. According to the method, a fine object source path omitted by a traditional method can be accurately identified, and the fineness and reliability of object source analysis are greatly improved.