Experimental and Numerical Investigations of Velocity Profiles for Unsteady Open Channel Flow

Abstract

Open channel flows, particularly under unsteady conditions, present a complex interplay between fluid dynamics and channel geometry. This complexity offers both challenges and opportunities for researchers in the field. Understanding unsteady open channel flow behaviour is crucial for various engineering applications, including flood forecasting, effective river management strategies, and accurate environmental impact assessments. However, capturing the dynamic nature of unsteady flow phenomena in experiments requires special considerations, as these can deviate significantly from simpler steady flow conditions. Unsteady flow phenomena, such as flood waves, are prevalent in natural water bodies and can have significant implications for flood forecasting, river management, and environmental assessments. Understanding these phenomena is essential for developing effective strategies to mitigate flood risks and manage water resources sustainably. This study addresses these challenges through a two-pronged approach: experimental investigations and numerical modelling. The experimental component meticulously examines the variation of longitudinal velocity profiles in an open channel as compared to steady flow. Two single repeated hydrographs with distinct unsteady parameters are introduced into a smooth rectangular channel, and velocity profiles are measured at specific cross-sections for two distinct flow depths (representing one low flow depth with aspect ratio 4.29 and one higher flow depth case with aspect ratio 2.73) within each hydrograph. Comparative analysis with steady flow conditions reveals that steady flow velocity profiles typically reside between the rising and falling limb profiles for corresponding flow depths. Depth-averaged velocities (DAV) are calculated by integrating local point velocities, providing further insights into the discrepancies between the two unsteady flow depths and the steady flow case. It is observed that for both the Hydrographs, the steady flow velocity profiles are placed within the velocity profiles of same depth of rising and falling limb cases. This is due the reason that in case of rising limb, there is acceleration in flow but for the same flow depth in falling limb, there is deceleration in the flow. These fluctuations in flow create dynamic flow patterns and can lead to challenges in predicting flow behaviour accurately. Computational Fluid Dynamics (CFD) simulations using ANSYS Fluent, utilizing established turbulence models like k-ω and k-ε, compare simulated results with experimental data for depth-averaged velocity and boundary shear stress profiles. The analysis reveals that the RNG k-ε model exhibits greater suitability for predicting depth- averaged velocity, while the SST k-ω model demonstrates superior accuracy for boundary shear stress profiles. These findings highlight the critical role of turbulence models in accurately representing flow properties influenced by complex roughness conditions, offering valuable insights for engineers and researchers tackling real-world fluid mechanics problems. The two-equation turbulence models are applied to simulate the flow in both smooth and rough (grass bed) rectangular channel to check the effectiveness of the turbulence models. From the validations of longitudinal velocity profiles and error analysis, it is observed that the Std k-ε model predicts the velocity more accurately in smooth case than other discussed turbulence models. To test and compare this numerical model with unsteady flow condition, OpenFOAM software is used instead of ANSYS Fluent. The Fluent software has some limitations which makes it difficult to apply a hydrograph as the inlet condition for representing unsteady flow condition of a channel. Numerical solution for the experimental unsteady flow conditions for the experimental smooth channel is conducted using a Computational Fluid Dynamics (CFD), OpenFOAM. This comprehensive approach provides a detailed understanding of open channel flow dynamics under unsteady conditions. By using this knowledge that for the smooth channel, the Std k-ε model is utilized for numerical simulation of unsteady flow for the same geometry and flow condition for both the hydrographs comprising unsteady flow. In addition to the experimental investigations, the study proposes a modified log-law approach for predicting velocity profiles in unsteady flow channels. The model's accuracy is rigorously validated against experimental data, demonstrating its effectiveness in predicting velocity profiles across a wide range of Reynolds numbers. The model's simplicity and efficiency make it a valuable tool for engineers and researchers alike, offering a means to predict velocity profiles not only in steady flow but also potentially in unsteady open channel scenarios, particularly for channels of smooth surfaces.