TL;DR
This paper introduces a quasi-2D energy analysis framework for simulating transient flows in viscoelastic pipes, focusing on the interplay between frictional and viscoelastic effects across various Reynolds numbers (Re). The findings reveal that while both frictional and viscoelastic work increase with Re, friction dominates at higher Re, and a 1D model can underestimate these effects at lower Re.
What Problem Does This Paper Address?
Transient flows in pipe systems, particularly those involving viscoelastic pipes, can lead to significant pressure fluctuations, commonly known as water hammer. While quasi-2D models have been used to simulate these flows, a detailed understanding of how the Reynolds number (Re) influences the energy dissipation due to both pipe wall viscoelasticity and fluid friction remains a gap. Existing models often struggle to accurately capture peak pressure damping and the complex energy transfer dynamics in these systems. This research aims to bridge this gap by employing an integral total energy (ITE) method to analyze the energy contributions of friction and viscoelasticity under varying Re conditions.
Methods
The study develops a quasi-2D friction model for transient flow in viscoelastic pipes. The core of the methodology involves:
- Quasi-2D Governing Equations: The continuity and momentum equations are adapted for a quasi-2D flow, considering both axial and radial velocity components.
- Kelvin-Voigt Model: Viscoelastic pipe behavior is modeled using the Kelvin-Voigt model, which accounts for both instantaneous and retarded strain components.
- Integral Total Energy (ITE) Method: This method is applied to derive an energy equation for the quasi-2D system. It allows for the analysis of the total internal energy, kinetic energy, frictional dissipation, work done at pipe ends, and work done by the pipe wall.
- Numerical Scheme: The Method of Characteristics (MOC) is employed to discretize and solve the governing equations for both the quasi-2D and a simplified 1D model. A gradually encrypted radial grid is used to handle the steep radial velocity gradients.
- Energy Analysis: The paper focuses on analyzing the work done by frictional (Df) and viscoelastic (WP) terms, as well as the total energy dissipation, under different initial Reynolds numbers.
Datasets
The experimental data used for validation was obtained from the Water Engineering Laboratory at the University of Perugia, Italy. This dataset includes measurements from a high-density polyethylene (HDPE) pipe with specific dimensions (inner diameter 93.3 mm, wall thickness 8.1 mm, length 128.6 m). Five groups of experimental data, each with a different initial Reynolds number, flow rate, and pressure, were analyzed. The Kelvin-Voigt parameters for the pipe were calibrated based on literature values specific to this experimental setup.
Key Insights and Ideas
- Reynolds Number Dominance: The study highlights that as the Reynolds number increases, the frictional work (Df) becomes significantly larger than the viscoelastic work (WP), especially at higher Re values. This suggests that friction plays a more dominant role in energy dissipation in turbulent transient flows within viscoelastic pipes.
- 1D Model Underestimation: A crucial finding is that the 1D model underestimates the frictional work (Df) at lower initial Re values (Re < 3 × 10⁵). This underestimation is reduced as Re increases, with negligible differences between 1D and 2D models for Re > 3 × 10⁵.
- Viscoelastic Work Consistency: The work done by the viscoelastic term (WP) shows remarkable similarity between the 1D and 2D models across all tested Re values. This is attributed to the consistent calculation of retarded strain in both models.
- Energy Transfer Dynamics: The analysis of energy changes reveals that while friction leads to continuous energy dissipation, the viscoelastic term exhibits sinusoidal fluctuations, indicating a more complex interplay of energy transfer and dissipation, particularly in the early stages of transient flow.
- Proportionality Shift: As Re increases, the proportion of energy dissipation attributed to friction in the total energy change grows, while the proportion from viscoelasticity decreases.
Key Results
- Both frictional work (Df) and viscoelastic work (WP) increase with increasing Reynolds number (Re).
- For high initial Re values, Df is substantially larger than WP.
- At Re < 3 × 10⁵, the 1D model underestimates Df compared to the 2D model. This discrepancy diminishes for Re > 3 × 10⁵.
- WP values are largely consistent between the 1D and 2D models across different Re values.
- The energy dissipation of the friction term increases rapidly initially and then stabilizes.
- The energy change of the viscoelastic term shows sinusoidal fluctuations in the early transient phase, with a longer duration and larger fluctuation range at smaller initial Re.
- The proportion of energy dissipation from friction increases with Re, while the proportion from viscoelasticity decreases. For example, at Re = 69,680 and t ≈ 60 L/a, the viscoelastic term's contribution to total energy variation is about 10.1%.
Limitations and Future Work
- Model Simplification: While the quasi-2D model captures radial effects, it still relies on simplifications. The paper acknowledges that the 1D model can underestimate certain effects at lower Re.
- Experimental Validation: The study validates its models against experimental data from a specific HDPE pipe. Further validation with different pipe materials and geometries would be beneficial.
- Future Directions: The authors suggest that this energy analysis framework provides a theoretical basis for further investigation into transient flows in viscoelastic pipes. Future work could explore the impact of different pipe materials, more complex viscoelastic models, and a wider range of flow conditions. The detailed energy breakdown offers a pathway to optimize pipe system design and operation by better understanding the trade-offs between frictional and viscoelastic energy losses.
Summary updated