|
| 基于数字孪生的大型调水泵站水力机组内部流动状态预测 |
| Prediction of the internal flow state of the hydraulic units in large pumping stations based on digital twins |
| 投稿时间：2025-08-08 修订日期：2026-03-03 |
| DOI： |
| 中文关键词: 轴流泵站 数字孪生 降阶模型 奇异值分解 流动状态预测 |
| 英文关键词: Axial-flow pump station Reduced Order Model Singular Value Decomposition Digital twin Reduced-order model Flow state prediction |
| 基金项目:江苏省水利科技项目(2024026)；国家重点研发项目课题(2022YFC3204603)； |
|
| 摘要点击次数: 38 |
| 全文下载次数: 0 |
| 中文摘要: |
| 针对大型调水工程传统流体动力学分析方法存在计算周期长、算力成本高以及难以实时捕捉动态特征的局限性，本文以低扬程泵站轴流式水力机组为研究对象，提出了一种基于 Simulink 仿真环境、由降阶模型驱动的数字孪生流动状态预测方法。该方法通过构建物理场降阶模型与虚实动态交互机制，实现机组在复杂工况下对流动状态进行毫秒级预测，并搭建物理模型试验台进行了系统验证。结果表明：经引入 RRMSE 及 LOOCV 方法评估，采用奇异值分解所建立的静压、速度及总压模型预测结果平均相对误差控制在 5% 以内，湍流强度平均相对误差为 8.5%，证明了模型在预测未知工况时的泛化能力。同时，该方法在保证仿真精度的前提下，单工况点平均仿真耗时仅约 0.1 s，计算效率显著提升，为智慧水利系统的高效建模与实时决策提供了技术性支撑。 |
| 英文摘要: |
| To address the limitations of traditional hydrodynamic analysis methods in large-scale water diversion projects—specifically long computation cycles, high computational costs, and the inability to capture dynamic characteristics in real-time—this paper proposes a digital twin flow field state prediction method driven by a Reduced Order Model (ROM) within the Simulink environment. This study focuses on axial-flow pump units in low-head pump stations. By constructing physical field ROMs and establishing a virtual-real dynamic interaction mechanism, the proposed method achieves millisecond-level prediction of flow field states under complex operating conditions. Furthermore, a physical model test rig was constructed to conduct systematic verification. The results indicate that, as evaluated by the RRMSE and LOOCV methods, the average relative errors of the static pressure, velocity, and total pressure models established via Singular Value Decomposition (SVD) are controlled within 5%, while the average relative error for turbulence intensity is 8.5%. These findings demonstrate the model's generalization capability for unknown operating conditions. Concurrently, while maintaining simulation accuracy, the average simulation time per operating point is approximately 0.1 seconds. This significant improvement in computational efficiency provides robust technical support for efficient modeling and real-time decision-making in smart water conservancy systems. |
|
查看/发表评论 下载PDF阅读器 |
| 关闭 |