3results about How to "Effective alignment" patented technology

The invention discloses a feeding mechanism of an automatic cold header for machining piston pins. The feeding mechanism comprises a main machine (1), a feeding mechanism body (20) is arranged on the front portion of the main machine (1), the feeding mechanism is connected with an iron wire frame (2) through an iron wire, and a discharging mechanism (3) is arranged on one side of the main machine (1). The feeding mechanism body comprises a base (4), fixed wheels (5) are arranged on one side of the base (4), and moving wheels (6) are arranged on the other side of the base (4). Side bases (7) are arranged on two sides of the base (4), sliding bases (8) are arranged on the inner sides of the side bases (7), and the contact surfaces of the sliding bases (8) and the side bases (7) are in sliding fit. The moving wheels (6) are arranged on the upper sides of the sliding bases (8), a threaded barrel (9) is fixed below the sliding bases (8), a threaded column (10) is arranged in the threaded barrel (9), a bearing pedestal (11) is arranged at one end of the threaded column (10), a rotating ring (12) is fixed to the other end of the threaded column (10), the rotating ring (12) is provided with a circle of small holes (13), and a rotating handle (14) is arranged in one small hole (13) in an embedded mode.

The invention belongs to the technical field of bearing lubrication state monitoring, and particularly relates to a bearing lubrication state real-time monitoring method based on hydrodynamic model correction, and the method comprises the steps: obtaining an original vibration acceleration signal and operation state data, carrying out the variational mode decomposition of the original vibration acceleration signal, and extracting a cavitation feature component; calculating a cavitation kinetic energy impact index according to the extracted cavitation characteristic component and the operation state data; dynamically correcting the equivalent fluid density of the lubricating oil according to the cavitation kinetic energy impact index and the contact area pressure of the bearing estimated by the running state data; and solving the minimum oil film thickness of a lubricating oil film by utilizing the corrected equivalent fluid density, and determining a lubrication safety margin index by combining the temperature rise rate of the bearing. According to the invention, the defect of monitoring result distortion caused by single-phase fluid hypothesis is overcome, and the accuracy of bearing lubrication state monitoring and early risk early warning is improved.

This invention relates to a feature extraction model training and feature extraction method and apparatus, comprising: inputting a training image set into an initial model; extracting several local features from the training images by a local feature extraction layer; enhancing the several local features using an attention mechanism by a feature enhancement layer to obtain several enhanced local features; a global feature extraction layer aggregating features based on the several enhanced local features to obtain global features; and a classifier obtaining predicted classification parameters for the training images based on the global features. The model parameters of the initial model are adjusted according to the label parameters and predicted classification parameters of the training images to obtain a trained classification model. A feature extraction model for extracting global features is constructed based on the local feature extraction layer, feature enhancement layer, and global feature extraction layer in the classification model. This provides an end-to-end global feature extraction scheme, reducing algorithm complexity.