3 results about "Mathematical process" patented technology

This invention discloses a method for multi-source data fusion and power prediction of wind turbine units, belonging to the field of power prediction technology. It addresses the technical problems of poor multi-source data fusion and power prediction analysis in existing solutions. By filtering and retaining key features through mutual information, redundant information can be effectively reduced. Dynamic weighting via an attention mechanism makes the fused features more adaptable to real-time operating conditions. Differential weight adjustments for wind speed, terrain, and equipment status ensure high relevance of the fused features even in complex scenarios. Spatial feature extraction captures local spatial correlations, temporal features capture temporal dependencies, and random forests handle nonlinear mappings, solving the problem of insufficient generalization ability of single models. The attention mechanism automatically assigns weights to different time steps and features, avoiding complex mathematical processes such as matrix operations and noise covariance estimation.

The invention discloses a production process classification method and system, a storage medium and a terminal, and the method comprises the steps: obtaining process text data of a plurality of production processes, each production process comprising a plurality of production steps, and the process text data comprising step text data of each production step; a spatial vector model is established, the spatial vector model comprises feature parameters of n dimensions, and the feature parameter of each dimension is digitally represented; inputting the plurality of flow text data into a spatial vector model for calculation processing, and obtaining n-dimensional spatial vector coordinates corresponding to each production flow; according to the spatial vector coordinate of each production process, obtaining the similarity between different production processes; and classifying the plurality of production processes according to the similarity between different production processes and a similarity threshold. The process text data which cannot be directly mathematized are converted into the spatial vector coordinates which can be directly mathematically calculated, and the production process can be accurately classified by performing calculation according to the digitalized spatial vector coordinates.

Embodiments of the present disclosure describe a method for operating a polishing system including determining a profile of one or more regions of a polishing pad of the polishing system during a polishing process. The contour is compared with a reference contour, the reference contour is a simulated contour of the polishing pad, and the comparison of the contour and the reference contour comprises comparison of the contour and the simulated contour generated through a mathematical process, a machine learning process, a big data mining process or a neural network process. One or more parameters of the grinding process are adjusted based on the comparison.