2results about How to "Enables real-time forecasting" patented technology

The invention discloses a data prediction method and device based on a time sequence model, and the method comprises the steps: collecting historical data in a preset time range, and carrying out the normalization processing of the historical data; predicting the normalized historical data by using each network model in the trained time sequence model to obtain prediction data corresponding to each network model; and according to the prediction data corresponding to each network model, utilizing a linear regression model to obtain prediction data of a predetermined time point.

The application discloses a pipeline erosion working condition prediction method based on POD-Kriging, and relates to the field of pipeline erosion prediction. The method comprises the following steps: determining an input parameter set and an output physical quantity of a reduced-order model based on a ROM database and in combination with variable ranges of industrial working conditions; selecting a plurality of design points based on the parameter value interval of the input parameter set, and obtaining the output physical quantity corresponding to each design point to construct a training data set; constructing an initial reduced-order model based on the training data set by performing an intrinsic orthogonal decomposition method and a spatial interpolation algorithm; and performing collaborative optimization on the number of intrinsic modes in the initial reduced-order model and the number of design points to obtain an optimized reduced-order model. The application establishes parameter mapping through the intrinsic orthogonal decomposition and the spatial interpolation algorithm, and realizes the collaborative improvement of the calculation efficiency and the accuracy in the industrial pipeline erosion field prediction by combining the collaborative optimization of the mode number and the design points.