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.
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
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.
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.
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.
Smart Images

Figure CN122132734A_ABST