Method, device and equipment for predicting heavy haul train coupler force

By constructing a coupler force prediction model and utilizing a one-dimensional convolutional neural network, a long short-term memory network, and a Transformer encoder, efficient and accurate prediction of the coupler force distribution of the entire train from a small amount of discrete parking space data is achieved. This solves the problems of high cost and poor real-time performance in existing technologies and provides an efficient safety monitoring solution.

CN122241055APending Publication Date: 2026-06-19SHENHUA RAIL & FREIGHT WAGONS TRANSPORT
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENHUA RAIL & FREIGHT WAGONS TRANSPORT
Filing Date
2026-02-12
Publication Date
2026-06-19

AI Technical Summary

⚠Technical Problem

When acquiring coupler force in heavy-haul trains, existing technologies are inadequate. Direct measurement is costly and incomplete, while dynamic simulation is computationally complex and lacks real-time performance, failing to meet the safety monitoring requirements of heavy-haul trains.

⚗Method used

A coupler force prediction model composed of a one-dimensional convolutional neural network, a long short-term memory network, a Transformer encoder, and an attention mechanism is adopted to achieve high-precision prediction of the coupler force distribution of the entire train by acquiring real-time coupler force data of a small number of discrete car positions.

🎯Benefits of technology

It reduces hardware costs, improves prediction accuracy and real-time performance, and can meet the safety monitoring and early warning needs of heavy-haul trains.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

This disclosure provides a method, apparatus, and device for predicting coupler forces in heavy-haul trains. The method includes: acquiring sampled coupler force data from multiple discrete positions in a target heavy-haul train; inputting the sampled coupler force data into a pre-trained coupler force prediction model, and outputting global coupler force data via the coupler force prediction model; wherein the global coupler force data represents the predicted coupler force values ​​for all positions in the heavy-haul train at the current moment; wherein the coupler force prediction model is trained based on at least one set of training data, each set of training data including sampled coupler force data and corresponding global coupler force ground truth data. Using the method of this disclosure, the urgent need for real-time monitoring and early warning of coupler forces during train operation can be met, providing efficient and reliable technical support for the safe operation of heavy-haul trains.
Need to check novelty before this filing date? Find Prior Art