A Method for Constructing a Monitoring Model for the Execution Progress of Marine Scientific Research Project Funding

By constructing a monitoring model for the execution progress of marine scientific research project funding based on three-dimensional tensor decomposition and multi-task learning, the problem of lack of anomaly detection and prediction in traditional methods is solved, and adaptive anomaly detection and prediction are achieved, improving the timeliness and accuracy of monitoring.

CN121836129BActive Publication Date: 2026-05-26BEIHAI FORECASTING CENT OF STATE OCEANIC ADMINISTRATION ((QINGDAO MARINE FORECASTING STATION OF STATE OCEANIC ADMINISTRATION) (QINGDAO MARINE ENVIRONMENT MONITORING CENT OF STATE OCEANIC ADMINISTRATION))
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIHAI FORECASTING CENT OF STATE OCEANIC ADMINISTRATION ((QINGDAO MARINE FORECASTING STATION OF STATE OCEANIC ADMINISTRATION) (QINGDAO MARINE ENVIRONMENT MONITORING CENT OF STATE OCEANIC ADMINISTRATION))
Filing Date
2026-03-13
Publication Date
2026-05-26

Smart Images

  • Figure CN121836129B_ABST
    Figure CN121836129B_ABST
Patent Text Reader

Abstract

This invention provides a method for constructing a monitoring model for the execution progress of marine scientific research project funding, belonging to the field of large model construction technology. This invention extracts potential expenditure patterns, uses a dynamic time warping algorithm to identify project execution pattern categories, constructs a multi-task learning neural network to simultaneously predict funding and progress, and launches a CUDA parallel computing framework to run cumulative sum control charts, Hotelling's T-squared statistic, and wavelet multi-scale decomposition tasks in three thread blocks to achieve anomaly detection. Anomaly confidence is calculated through sequential probability ratio testing to trigger an early warning mechanism. Multiple imputation methods are used for missing data, and robust M-estimation is used to reduce the impact weight of anomalous observations. The optimal estimation result is output by fusing model predictions and noisy observations through Kalman filtering. This invention solves the technical problem of lacking anomaly detection and prediction capabilities in monitoring the execution progress of marine scientific research project funding.
Need to check novelty before this filing date? Find Prior Art