A method for predicting the hydraulic line of a slurry pipeline by fusing a physical model and a two-channel network

By integrating physical models with dual-channel networks, the accuracy and consistency issues in hydraulic line prediction for high-concentration slurry pipelines were addressed. This approach achieved the coordinated integration of global operating conditions and local terrain features, constructed a custom neural network model, improved prediction accuracy and stability, and supported the optimization and energy management of slurry pipeline systems.

CN122132734APending Publication Date: 2026-06-02KUNMING UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
KUNMING UNIV OF SCI & TECH
Filing Date
2026-02-06
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

In the prediction of hydraulic lines in high-concentration slurry pipelines, existing technologies suffer from insufficient accuracy and consistency. Traditional empirical formulas cannot accurately describe the flow characteristics, numerical simulation methods are highly dependent on parameters and difficult to measure in real time, and machine learning methods ignore physical laws and have serious data quality problems.

Method used

By employing a method that integrates physical models and dual-channel networks, a neural network model with data preprocessing, adaptive pipeline segmentation, and quadruple physical constraints is constructed. This model combines Bernoulli's equation and Darcy-Weisbach's formula to achieve the coordinated integration of global working conditions and local terrain features, eliminate abnormal data, and introduce physical constraints to build a custom dual-channel deep neural network.

Benefits of technology

It achieves high-precision and stable hydraulic line prediction with an average relative error of less than 1.19%, maintaining prediction consistency under unknown operating conditions and long-term operation, and supporting the optimized operation and energy consumption control of slurry pipeline systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122132734A_ABST
    Figure CN122132734A_ABST
Patent Text Reader

Abstract

This invention discloses a method for predicting hydraulic lines in slurry pipelines by integrating a physical model and a dual-channel network, belonging to the field of slurry transportation technology. The invention collects slurry pipeline data, performs preprocessing on the collected data, and dynamically merges adjacent approximate slope segments using an adaptive pipeline segmentation algorithm. Based on the preprocessed data and the merged pipeline segments, a dual-channel deep neural network is constructed to process global operating conditions and local terrain features respectively. Finally, the prediction results are analyzed and visualized. This invention integrates a physical model into the dual-channel neural network and incorporates local terrain features to predict the distribution of hydraulic lines. By segmenting the entire pipeline, the model's prediction results closely match the actual results, providing an effective technical solution for the safe and efficient operation of slurry pipelines.
Need to check novelty before this filing date? Find Prior Art