Barrier dam outburst peak flow prediction and risk assessment method based on data enhancement and ensemble learning

CN120911247APending Publication Date: 2025-11-07POWER CHINA KUNMING ENG CORP LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510938626.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-08
Publication Date
2025-11-07

AI Technical Summary

Technical Problem

Existing models for predicting peak flow rates of landslide dammed lakes lack measured data, resulting in insufficient model representativeness and limited applicability, and making it difficult to accurately assess the risk of dam failure.

Method used

By employing data augmentation and ensemble learning methods, multi-source data is collected, preprocessed, and augmented to construct a fluid dynamics-machine learning hybrid model. This model is then combined with rock mass fracture propagation rate for hazard assessment, generating a more comprehensive augmented dataset and calculating hazard assessment indices using an ensemble learning framework.

Benefits of technology

It improves the accuracy of peak flow prediction and the generalization ability of the model, enabling more accurate assessment of the risk of landslide dam failure and providing timely prevention and control measures.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120911247A_ABST
    Figure CN120911247A_ABST
Patent Text Reader

Abstract

The invention discloses a barrier dam outburst peak flow prediction and risk assessment method based on data enhancement and ensemble learning. The method specifically comprises the following steps: S1, collecting multi-source data of a barrier lake; s2, enhancing the preprocessed data; s3, the capacity increasing rate of the barrier lake is obtained through calculation; s4, constructing a fluid mechanics-machine learning hybrid model; s5, risk assessment and prediction; by collecting multi-source data such as the upstream flow, the downstream flow, the storage capacity, the water level, the rainfall capacity and the soil water content, the dynamic change process of the barrier lake is comprehensively reflected, and the limitation of a single data source is avoided; the data is enhanced, and a more comprehensive and more stable enhanced data set is generated through time sequence interpolation and multi-source data fusion, so that the accuracy of peak flow prediction and risk assessment is improved; the input features of the risk assessment model are fused through dynamic weighting, and a linkage control algorithm is calculated through an integrated learning framework to generate risk assessment indexes.
Need to check novelty before this filing date? Find Prior Art