4 results about "Cloud atlas" patented technology

The application provides a wireless charging coil type and mutual inductance identification method, establishes a circular transmitting coil and a circular receiving coil model, a square transmitting coil and a square receiving coil model, a circular transmitting coil and a square receiving coil model, a square transmitting coil and a circular receiving coil model, obtains a magnetic field cloud atlas and mutual inductance values of the receiving coil and the transmitting coil, and constructs a training data set; taking pixel values of the coil magnetic field cloud atlas as input and taking the coil type as output, a coil type identification model is constructed; taking the pixel values of the coil magnetic field cloud atlas as input and taking the transmitting coil and the receiving coil as output, a coil mutual inductance identification model is constructed; a loss function is set, and the coil type identification model and the coil mutual inductance identification model are trained, so as to identify the wireless charging coil type and mutual inductance. The application realizes identification of the wireless charging coil type and mutual inductance, and provides a new idea for wireless power transmission method research and engineering application.

The application discloses a mechanism decoupling transfer learning driven composite bolt connection failure efficient prediction method, relates to the technical field of composite material failure research, and comprises the following steps: a high-throughput simulation database is established, unstructured grids are converted into structured tensors by using a physical coordinate mapping strategy, and a generative network containing coordinate injection and gradient consistency constraints is constructed, so that the second-level high-precision prediction of a composite material connection damage cloud atlas under any geometric parameters is realized. The method is used to effectively fill the blank of the prior art, and provides a new failure prediction mode for composite material bolts with different layer designs, width-diameter ratios and end-diameter ratios.

The application relates to the technical field of image processing, in particular to a frozen wall closure state detection method and system based on temperature cloud atlas analysis. The method comprises the following steps: acquiring an anti-permeability strength image of a horizontal section of a detection depth; acquiring a binary image of the anti-permeability strength image; extracting a heat conduction skeleton of a frozen wall from the binary image by using a center axis transformation algorithm; performing multiple sampling on the heat conduction skeleton to obtain a discrete sampling point sequence; acquiring effective thermal resistance thickness at each discrete sampling point; acquiring the geometric curvature of each discrete sampling point; taking the product of the geometric curvature and a water flow scouring intensity factor as an adjustment coefficient; taking the ratio of the effective thermal resistance thickness and the adjustment coefficient as a buckling stability index; and determining the closure state of the frozen wall according to the comparison result of the global minimum value of the buckling stability index and a preset safety threshold. The application has the effect of improving the detection accuracy of the frozen wall state.

The application discloses a data post-processing method, device and equipment. The method comprises the following steps: obtaining cloud atlas data of a target part, and establishing a target node set based on the cloud atlas data; obtaining screening information set by a target object, determining a plurality of target nodes from the target node set based on the screening information, and determining associated nodes of each target node; determining a view angle vector and a characteristic size corresponding to each target node based on the associated nodes of each target node, wherein the view angle vector is used for determining a view angle direction of a local cloud atlas corresponding to the target node, and the characteristic size is used for determining a scaling ratio of the local cloud atlas corresponding to the target node; generating a local cloud atlas corresponding to each target node based on the view angle vector and the characteristic size, saving the local cloud atlas, and saving information of the target node and the associated nodes. The application solves the technical problem that the post-processing result of the cloud atlas data in the calculation result cannot meet the demand in the finite element simulation analysis process.