Icing prediction method and system based on terrain potential energy field and multi-scale meteorological coupling

Through the method of terrain potential tensor and multi-scale meteorological coupling, the problems of model fragmentation and lack of terrain coupling in ice cover prediction are solved, high-precision ice cover thickness prediction under complex terrain and extreme meteorological conditions is achieved, and the stability and timeliness of the prediction are improved.

CN120597513APending Publication Date: 2025-09-05STATE GRID SICHUAN ELECTRIC POWER CORP ELECTRIC POWER RES INST
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202510689900.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-27
Publication Date
2025-09-05

AI Technical Summary

Technical Problem

Existing ice cover prediction technologies have problems such as model fragmentation, lack of terrain coupling and inaccuracy in extreme scenarios, which lead to a significant increase in ice cover thickness prediction errors under complex terrain and extreme meteorological conditions.

Method used

By constructing a terrain potential energy tensor to quantify the dynamic influence of terrain elements on wind acceleration and humidity concentration and dispersion, combined with the meteorological field correction of multi-scale partial differential coupling equations, a physical-data dual-model collaborative prediction framework is established, integrating the ice growth mechanism and complex nonlinear relationships to achieve accurate prediction of ice thickness.

Benefits of technology

It significantly improves the accuracy of ice cover prediction in complex terrain and extreme meteorological conditions, enhances the timeliness and stability of prediction, and solves the problem that traditional static terrain parameters cannot represent dynamic terrain properties.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120597513A_ABST
    Figure CN120597513A_ABST
Patent Text Reader

Abstract

The invention discloses an icing prediction method and system based on terrain potential energy field and multi-scale meteorological coupling, relates to the technical field of power transmission line icing prediction, and solves the problems that a traditional icing thickness prediction model is split and traditional static terrain parameters cannot represent dynamic terrain potential energy. According to the technical scheme, the method comprises the following steps: acquiring meteorological data and high-resolution digital elevation data; calculating a topographic potential energy tensor; establishing a multi-meteorological coupling model, and correcting the multi-meteorological coupling model through the topographic potential energy tensor; carrying out meteorological element correction on the corrected multi-meteorological coupling model to obtain a corrected meteorological field; and obtaining a first icing thickness prediction value and a second icing thickness prediction value based on a pre-constructed icing growth physical model and a pre-trained machine learning model, and carrying out weighted calculation on the first icing thickness prediction value and the second icing thickness prediction value to obtain a final icing thickness prediction value. And the effect of improving the prediction precision under complex terrain and extreme meteorological conditions is achieved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Rotary kiln wall thickness prediction method based on mixing of TabNet model and mechanism model

    CN121051328A

  • Rotary kiln wall thickness prediction method based on tabnet model and mechanism model hybrid

    CN121051328B

  • Wind power plant icing disaster sensing and early warning method and system

    CN121330856A