Component risk prediction method and system based on time sequence prediction and storage medium

By deconstructing and separating the multidimensional feature data of open-source components over time, and combining the trend inertia index and the instantaneous impulse index, a long short-term memory network model is used for risk assessment. This solves the problem of feature interference in existing technologies and enables accurate early warning and assessment of risks to open-source components.

CN121599494AActive Publication Date: 2026-03-03SUZHOU BONA XUNDONG SOFTWARE CO LTD
View PDF 8 Cites 1 Cited by

Patent Information

Application Number
CN202610121886.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-29
Publication Date
2026-03-03
Estimated Expiration
2046-01-29

AI Technical Summary

Technical Problem

In the risk prediction of open-source components, existing technologies suffer from feature interference caused by the differences in the time evolution characteristics of indicators of different dimensions. This makes it difficult for the model to accurately capture the decline trend of long-term indicators or the abnormal change of short-term indicators, and thus cannot meet the requirements for accurate early warning in high-security scenarios.

Method used

By collecting multidimensional feature data of open-source components, deconstructing the data on a time scale based on the frequency of indicator changes, the feature data is divided into slow evolution sequences and fast mutation sequences. The trend inertia index and instantaneous impulse index are calculated respectively, and the long short-term memory network model is used to perform time-series dependency analysis to generate component risk scores. Finally, graded early warning or automatic interception is carried out.

Benefits of technology

It achieves accurate early warning of risks to open source components, eliminates mutual interference between features of different frequencies, ensures that the model responds sensitively to long-term hidden dangers and sudden anomalies, and outputs interpretable risk assessment results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121599494A_ABST
    Figure CN121599494A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of computers, in particular to a time sequence prediction-based component risk prediction method and system and a storage medium, and the method comprises the steps: collecting multi-dimensional feature data of an open source component, and carrying out the time scale deconstruction based on the index change frequency of each feature, constructing a heterogeneous feature sequence set containing a slow evolution sequence and a rapid mutation sequence; respectively calculating a trend inertia index reflecting a long-term decline trend and an instantaneous pulse index reflecting short-term abnormal fluctuation, and generating a time sequence prediction input vector; inputting the time sequence prediction input vector into a long short-term memory network model, and outputting a basic risk probability of the component in a future preset time window; calculating and generating a component risk score based on the basic risk probability, the trend inertia index and the instantaneous pulse index; and comparing the component risk score with a preset safety threshold, and executing graded early warning or automatic interception operation. According to the invention, an accurate and interpretable risk early warning result can be output.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Rolling bearing residual life prediction method based on space-time degradation characteristic decoupling

    CN120724412A

  • Building engineering interaction method and system based on BIM model, and medium

    CN120930856A

  • Nuclear power plant radiation environment monitoring method and system

    CN121142603A

  • Dual-energy fusion imaging and target identification method and system

    CN121208030A

  • Photovoltaic support adaptive tracking control method and system

    CN121254903A

Cited By

  • Visual identification and trend prediction method and system for wrinkles of continuous strip-shaped material

    CN121884008A