Energy efficiency and length modification of stilling basins with variable Baffle and chute block designs: A case study of the Fewa hydroelectric project

1 Introduction

Hydropower, a reliable renewable energy source, faces challenges in construction and operation, especially during floods. Hydraulic structures, particularly dam headworks, are vulnerable to high flood events. A spillway is crucial for safely releasing excess water from a reservoir or river to downstream areas [1]. The velocity of water passing through a spillway is so high that it is liable to cause erosion of the bed channel immediately below the top of the structure [2]. To stabilize riverbed sediment transport, mitigate the impacts of hydropower sustainability [3], and prevent riverbed erosion at downstream and dam undermining, the incorporation of an energy dissipator at the base of structures is needed to dissipate high-velocity flow energy and establish safe flow conditions in the outlet [4]. The stilling basin is crucial in dissipating the excess energy retained by dams, ensuring the integrity of headworks [5,6]. Before entering the river, the flow downstream of the spillway structure should return to its normal state. To reduce the risk of structural instability and to maximize the energy dissipator’s hydraulic performance, the energy dissipation must be accurately assessed [7,8]. A hydraulic jump is a crucial technique used by hydraulic engineers to design stilling basins, acting as an energy dissipater [9,10]. The dimensions of hydraulic jump stilling basins depend on the jump length and sequent depth [11]. Many efforts have been made to reduce the size of hydraulic jump stilling basins by using baffle blocks and end sills for modifying the jump characteristics to obtain comparable or better performance in shorter lengths [12]. For the dam structure to be both economical and efficient, the stilling basin design must be optimized [13,14]. The addition of chutes and basin blocks hinders and elevates water flow, reducing the basin length and concrete usage, thus lowering construction costs [15]. These blocks stabilize the jump under normal conditions, ensuring that it remains steady and not drained or washed out [16]. The impacts of these blocks are determined by their placement, height, and the gap that separates them [17].

During the past decades, many attempts have been made to reduce the size of stilling basins and enhance their energy dissipation efficiency by forcing the jump to occur within a short distance of the apron using various types of basin appurtenances [18,19]. These appurtenances act as hydraulic structures that facilitate the dissipation of potential energy retained by the dam or barrage [20]. Hayder [21] studied the stilling basins and found that rough beds with semi-circular shapes reduced the hydraulic jump length by 56% and the tail water depth by 25% with the same Froude number, compared to various basins like USBR-I, USBR-II, USBR-III, USBR-IV, and SAF. Abbas et al. [22] studied hydraulic jump characteristics in a stilling basin with adverse slope change. They used baffle blocks with different configurations to alter the basin's dimensions and test their effects on hydraulic jump characteristics. The results showed that baffle blocks reduced the sequent depth ratio, length of jump ratio, and roller length, but increased the energy dissipation ratio. According to Elnikhely [23], water dissipates more energy when it forms over staggered cylinder blocks that are fixed to a spillway; the least amount of energy is lost in a diagonal arrangement. Bestawy [24] compared the hydraulic performance of 14 different shapes of baffle blocks in a stilling basin, including semi-circular, triangular, trapezoidal, and rectangular shapes in different orientations setting inflow as Froude numbers 5–9, where the semi-circular shape was found to be the most efficient, as it significantly reduced parameters like the sequent depth ratio and hydraulic jump length compared to other shapes. Kang [25] conducted an experimental study to assess the impact of different shapes of baffle blocks on flow dissipation downstream of a weir. The study involved five different baffle block shapes, installed on a fixed bed, under two discharge values (0.140 and 0.325 m³/s). The flow rate downstream of the weir was measured with and without baffles, and the results indicated that square baffle blocks provided a significant flow dissipation effect. Gubashi et al. [26] analyzed how stepped spillway configurations influence flow turbulence and energy loss, offering insights into the impact of geometric variations. A study by Zaffar and Hassan [27] examined the Taunsa Barrage in Pakistan, focusing on the remodeling of a stilling basin built in 1958 that was later remodeled in 2008, modifying chute blocks and baffle blocks. Various hydraulic parameters were compared and analyzed using Flow-3D, and the new basin had minimal energy dissipation and significant scouring. Djunur et al. [28] further examined how baffle block angles affect jump behavior in porous rectangular designs, offering a new perspective on geometric optimization. Bakhtyar et al. [29] used a dynamic programming procedure to optimize the cascade stilling basin design of Tehri Dam to minimize excavation and concrete work costs, which shows a 31% improvement in minimizing concrete and excavation volumes and construction costs. These studies underscore the impact of shape, orientation, and flow interaction in stilling basin performance and form the conceptual foundation for the novel block designs explored in this study.

Although there have been several studies on the stilling basin design and efficiency of energy dissipation, there remains limited exploration of innovative geometries in a comparative modeling framework. Most research work focuses on traditional designs and misses out on modification strategies, balancing energy dissipation with the compactness of structure. Advances in computer technology have made it possible to examine the hydraulic modeling of full-scale spillways and barrage bays with the use of more effective computational fluid dynamics (CFD) techniques [30]. Yamini et al. [31] demonstrated the effectiveness of CFD modeling in evaluating the hydraulic performance of seawater intake systems, showcasing the adaptability of numerical simulations in complex flow systems and underlining the growing reliance on CFD tools for advancing hydraulic design. Using a FLOW-3D model, the energy dissipation efficiency of a stilling basin is examined in this study. The objective of this study is to numerically evaluate the performance of modified stilling basin geometries using FLOW-3D, focusing on improving energy dissipation and minimizing basin length. This research work is unique in that it modifies stilling basin appurtenances, including curved, triangular, and honeycomb structures, with the regular types recommended by the USBR by altering the geometry of the baffle and chute blocks, which aids in improving the flow characteristics within the basin and increasing energy dissipation close to the floor. By conducting simulations using Flow3D Hydro and analyzing the results, the research determines the most efficient shape for energy dissipation. These findings will contribute to the modification of stilling basin design and enhance its hydraulic performance. Through this approach, the study lends momentum toward more efficient, economical, and compact stilling basin design insights with practical solutions for modern hydropower development.

2 Materials and methods

2.3 Methodology

The methodology adopted in this research modifies the length of the stilling basin for energy dissipation and hydraulic performance with reference to Model 1, which replicates the hydraulic structure of the stilling basin of the Fewa Hydroelectric Project at field conditions. The different arbitrary geometries of the chutes and baffles blocks were chosen as different models, which aid in maximizing energy dissipation. The approach was systematic in site selection, data collection, 3D model creation, and CFD simulations using Flow3D Hydro. This multivariate process was a composite of model comparison, modification, and performance evaluation to determine the efficient flow characteristic optimal basin length. Field data and advanced modeling techniques were combined to simulate the complex hydraulic phenomenon within the basin of unsteady flow conditions accurately. It was done by varying the length across different models within a specified range based on the preliminary design information for iterative simulations to be carried out on the impacts against hydraulic parameters toward maximizing energy dissipation operational efficiency. The length reduction of every model (M_n) continued until the total energy (E ₂) and velocity (v ₂) at 35 m downstream from the dam heel exceeded the threshold values established by the baseline Model 1 (M1). This modification process gives well in advance recommendations on how to design and build effective stilling basins, thereby giving valuable insights for future study. A flow chart outlining the methodological procedure was developed prior to conducting the study and analysis, as shown in Figure 1.

Figure 1

Methodological flowchart.

2.4 Construction of a 3D model

Data from the site were processed using AutoCAD to plot all the required dimensions and features for the three-dimensional model. The AutoCAD file was then exported into the SketchUp, where the final 3D model was generated to replicate the hydraulic parameters of the Stilling basin at field conditions, measuring a length of 21.23 m, a width of 28 m, and a 1.57 m high end sills comprising chute and baffle blocks, as shown in Figure 2(a). Different block configurations were designed to enhance energy dissipation and improve hydraulic performance in the stilling basin to replace the existing chute and baffle blocks, as described in Table 1. Model 1 represents the actual configuration of the Fewa Hydropower Project and served as the baseline. Models 3–5 (curved blocks on horizontal/vertical surfaces and triangular blocks) were inspired by various previous studies on alternative baffle shapes aiming to enhance energy dissipation [21,24,25], where different geometries such as trapezoidal, semicircular, and triangular blocks showed improved turbulence and flow resistance. Models 2 and 6–8, which incorporate honeycomb-shaped blocks, represent novel design propositions. These designs were developed based on the hypothesis that their increased surface complexity and turbulence-inducing cavities could improve dissipation and reduce required basin length. The uniformity in block size across all models ensured a comparable blockage ratio, allowing fair performance assessment among geometries. In this respect, the 3D model was exported to Flow3D Hydro in order to allow for enhanced hydraulic behavior assessment. This functionality allowed for an effective evaluation of the stilling basin geometry, supporting the modification process. The approach provided herein ensures testing of various configurations of blocks and identification of the most feasible length of the stilling basin for each model.

Figure 2

(a) Model 1 confirming as per existing field conditions, (b) Model 2 with honeycomb baffle blocks, (c) Model 3 with curved (on horizontal surface) blocks as baffle blocks and chute blocks, (d) Model 4 with curved shape (on vertical surface) baffle blocks and chute blocks, (e) Model 5 with triangular baffle blocks and chute blocks, (f) Model 6 with honeycomb baffle blocks and chute blocks, (g) Model 7 with honeycomb staggered baffles blocks and chute blocks, and (h) Model 8 with two series of honeycomb baffle blocks and no chute blocks.

Table 1

Description of different model block configurations

Model	Description	Chute blocks	Baffle blocks
Model 1	Model confirming as per existing field condition (Figure 2(a))	2.47 m long, 2.15 m high, and 1.73 m wide	2.57 m long, 2.15 m high, and 1.61 m wide
			Thickness at top = 0.43 m
Model 2	Model with honeycomb baffle blocks (Figure 2(b))	2.47 m long, 2.15 m high, and 1.73 m wide	2.15 m × 2.15 m × 2.15 m honeycomb blocks, featuring 1.42 m diameter cylindrical slots with a 2.03 m diameter spherical cut at the center
Model 3	Model with curved (on horizontal surface) blocks as baffle blocks and chute blocks (Figure 2(c))	Curved block with 2.15 m long, 2.15 m wide, and 2.15 m high, and horizontal surfaces cut in a curved manner by a cylindrical surface with a diameter of 1.65 m. It shows a trapezoidal shape from side view, with a width of 2.15 m at the bottom and a reduced width of 1.07 m at the top	Curved block with 2.15 m long, 2.15 m wide, and 2.15 m high, and horizontal surfaces cut in a curved manner by a cylindrical surface with a diameter of 1.65 m. It shows a trapezoidal shape from side view, with a width of 2.15 m at the bottom and a reduced width of 1.07 m at the top
Model 4	Model with curved shape (on vertical surface), baffle blocks, and chute blocks (Figure 2(d))	Curved block with 2.15 m long, 2.15 m wide, and 2.15 m high, and vertical surfaces cut in a curved manner by a cylindrical surface with a diameter of 1.65 m. It shows a trapezoidal shape from side view, with a width of 2.15 m at the bottom and a reduced width of 1.07 m at the top	Curved block with 2.15 m long, 2.15 m wide, and 2.15 m high, and vertical surfaces cut in a curved manner by a cylindrical surface with a diameter of 1.65 m. It shows a trapezoidal shape from side view, with a width of 2.15 m at the bottom and a reduced width of 1.07 m at the top
Model 5	Model with triangular baffle blocks and chute blocks (Figure 2(e))	Triangular blocks of dimensions of 2.15 m for each face, forming an equilateral triangle with 2.15 m high	Triangular blocks of dimensions of 2.15 m for each face, forming an equilateral triangle with 2.15 m high
Model 6	Model with honeycomb baffle blocks and chute blocks (Figure 2(f))	2.15 m × 2.15 m × 2.15 m honeycomb blocks, featuring 1.42 m diameter cylindrical slots with a 2.03 m diameter spherical cut at the center	2.15 m × 2.15 m × 2.15 m honeycomb blocks, featuring 1.42 m diameter cylindrical slots with a 2.03 m diameter spherical cut at the center
Model 7	Model with honeycomb staggered baffles, blocks, and chute blocks (Figure 2(g))	2.15 m × 2.15 m × 2.15 m honeycomb blocks, featuring 1.42 m diameter cylindrical slots with a 2.03 m diameter spherical cut at the center	2.15 m × 2.15 m × 2.15 m honeycomb blocks were placed in staggered, featuring 1.42 m diameter cylindrical slots with a 2.03 m diameter spherical cut at the center
Model 8	Model with two series of honeycomb baffle blocks and no chute blocks (Figure 2(h))	No chute blocks	Two series of 2.15 m × 2.15 m × 2.15 m honeycomb blocks separated by twice the width of the blocks, featuring 1.42 m diameter cylindrical slots with a 2.03 m diameter spherical cut at the center

2.5 Simulation setup

The imported model from the SketchUp was defined in the Flow3D Hydro by assigning different factors such as the mesh size and boundary conditions. To ensure the accuracy and reliability of the FLOW-3D simulation results, model calibration and validation were performed by comparing the simulated water depth with field-measured depth at a reference section 35 m downstream of the dam heel. Using the same discharge conditions as those measured in the field, numerical simulations were conducted in FLOW-3D. Calibration consisted of changing mesh size, turbulence model parameters, and boundary conditions to minimize errors between simulated and observed values. A mesh size of 0.15 m was defined in the region containing the basin appurtenance, and it was varied up to 2 m in the upstream, where the flow does not need much analysis. The RNG (renormalization group) k−ε turbulent model is employed in this study among the other turbulent models that are accessible. The averaged equations for the turbulence quantities used by the k−ε model are derived using statistical techniques, demonstrating an improved capacity to describe flows with severe shear effects [34] and provide better results in the free surface modeling [35]. The volume of fluid (VOF) method was defined as the upstream boundary condition, where the downstream condition was set as the floor bottom as a wall and the fraction of the fluid set as zero that allows splashing of the fluid, as demonstrated in Figure 3. The initial condition of the fluid was defined with the Global and Regions, where the Global condition of 0.83 m from the bottom was set to define the presence of water in all of the selected area, and 11.5 m from the floor at the upstream of the dam was set as the Regional condition as a stagnation pressure. After the construction of the model, it was FAVORized (Fractional Area Volume Obstacle Representation) to embed the geometry of the model into the described mesh. FAVORizing allows for visualizing the deviation of the shape of the components of the model that has been embedded in the defined mesh and, hence, the size of the mesh can be changed as per the requirement [36,37].

Figure 3

Simulation setup.

The calibration yielded a coefficient of determination (R ²) value of 0.9316, indicating good agreement, while validation yielded an R ² value of 0.9847, indicating strong model performance and generalizability. These comparisons, plotted in Figure 4, confirm that the model is a faithful representation of the stilling basin hydraulic performance for a range of operational discharges. The model was thus deemed to be sufficiently robust to be utilized to evaluate the hydraulic performance of alternative block configurations and basin geometries. After the model calibration and validation, VOF with the discharge of 588.44 m³/s, a 1,000-year flood discharge, considered as the design discharge for the stilling basin, was defined as the upstream boundary condition for the analysis. After the model run, further trials were done varying the length of the stilling basin by shifting the end sill. The length was reduced by 2.5% of the length of the stilling basin and then by 5%, 7.5%, and so on up to 12% to compare the results with Model 1.

Figure 4

Model: (a) calibration and (b) validation.

3 Results and discussion

The use of FLOW-3D 2023R2 in this study, coupled with the RNG k–ε turbulence model and VOF method, provides a high degree of reliability and robustness in simulating complex hydraulic performance for each of the eight stilling basin models under a 1,000-year design discharge. The total energy at the crest of the spillway (E ₁) reaches a value of 14.177 m. The energy dissipated along the spillway is very small and is neglected for the convenience of the study. As we move downstream, at 35 m from the heel of the dam, where the actual river bed starts, the total hydraulic head (E ₂) and depth-averaged velocity (v ₂) were considered and analyzed for predicting the best model to enhance energy dissipation. A total energy (E ₂) of 5.636 m and a depth-averaged velocity (v ₂) of 7.205 m/s of Model 1(Base Model) were computed at section 35 m from the heel of the dam when the flow reached a steady state, which was considered as the threshold value for comparing all of the remaining models. Among the study of seven different models studied and comparing each of them with the base model, five of the models performed best in terms of all the prior mentioned hydraulic parameters.

The simulation results for Model 1, which replicates the conventional stilling basin at the Fewa Hydropower Project based on USBR-type design, showed an energy dissipation efficiency of 60%. This value aligns closely with typical empirical estimates for conventional stilling basins, which generally fall in the range of 55–70% energy dissipation, depending on the basin geometry and block arrangements [16]. A comparative analysis of total energy dissipation for each model, as illustrated in Figure 5, reveals that Models 3 and 6 achieved the highest energy dissipation rate of 67%, followed closely by Models 4 and 2, with dissipation rates of 66 and 64%, respectively. Additionally, Figure 6 indicates that Model 6 exhibited the lowest depth-averaged velocity of 6.337 m/s, followed by Model 3 of 6.379 m/s. Similarly, as illustrated in Figure 7, the total energy was lowest for Model 6, measuring 4.696 m, with Model 3 closely following 4.742 m. In contrast, Models 5 and 7 exceeded acceptable limits for both total energy and depth-averaged velocity. Consequently, Model 6, incorporating honeycomb basin blocks and chute blocks, demonstrated the highest efficiency in energy dissipation with the least depth average velocity under the given conditions.

Figure 5

Energy dissipation computed at section 35 m from the heel of the dam.

Figure 6

Depth-averaged velocity computed at section 35 m from the heel of the dam.

Figure 7

Total energy computed at section 35 m from the heel of the dam.

A series of trials was done with a reduction in length of the stilling basin for each model, except model 1, until the threshold limit for the parameter was crossed. Therefore, the number of trials required for each model was varied accordingly, giving the reduced length of the stilling basin for each model, as described in Tables 2 and 3. Overall, six trials were done, with the length of the stilling basin reducing to 12% less than the initial length of the stilling basin of length 21.23 m. Among all the different models, Model 2 was identified as the optimized model for reducing the length of the stilling basin. For Model 2, the total energy at a distance of 35 m was found to be 5.647 m, whereas the depth average velocity was 7.233 m/s when the length was reduced by 12%. Both parameters surpassed their respective threshold limits, suggesting that further reduction in the stilling basin length was not possible. The length at which the total energy and depth average velocity of the model corresponded to the threshold limit was determined when the initial length of the stilling basin was reduced by 11.8%. As a result, the final reduced length of the stilling basin was determined to be 18.752 m with the model incorporated with honeycomb baffle blocks (Model 2). This modified design demonstrates that the stilling basin can effectively dissipate energy with a shorter length, providing a more economical design.

Though honeycomb structures have demonstrated significant hydraulic benefits in both energy dissipation and basin length saving, their long-term efficiency would be threatened by maintenance problems. For this purpose, in particular, the recessed cavities and internal channels of the honeycomb structures would be potential sites of sedimentation or debris accumulation and thus a reduction in flow efficiency or increased need for maintenance activities. These factors were not addressed in the present CFD-based analysis. While the proposed honeycomb-based configurations have demonstrated significant improvements in energy dissipation and potential for basin length reduction, several limitations must be acknowledged. The present study evaluated stilling basin performance under a single, extreme design discharge (1,000-year flood), which reflects a critical operational scenario for structural safety. However, hydropower plants typically operate under a wide range of flow rates, including low-flow (partial load) and high-flow (flood release) conditions. The sensitivity of energy dissipation performance to such variations was not assessed in this study and remains a limitation. Despite such limitations, the proposed approach offers a helpful framework for evaluating the geometry of unconventional energy dissipators with CFD techniques. Further investigations must be conducted to test the configurations under varying flow conditions, physical model verification, and maintainability under sediment transportation conditions. Aids in these aspects would enhance the overall versatility and practical application of honeycomb block systems across various hydraulic settings.

4 Conclusions

This study has shown the effectiveness of various stilling basin layouts in optimizing energy dissipation and enhancing hydraulic performance. In doing so, the 3-D Reynolds-averaged Navier–Stokes equations were solved, including the RNG k–ε turbulence model and a VOF method, to capture the free surface. Of the eight models examined, Model 6, with honeycomb structures as the basin’s blocks, achieved the highest efficiency of 67% reduction in flow energy. This performance is due to the increased turbulence and frictional losses of honeycomb structures because of the increase in drag by the honeycomb cavity shape, thus effectively dissipating energy. Among the eight models, Model 2 incorporated with honeycomb baffle blocks was able to reduce the stilling length by 11.8% of its original length, bringing it down to 18.572 m without surpassing the threshold limits. Furthermore, Model 3 showed comparable potential with a 7.5% reduction in basin length but beyond the velocity barrier, and Models 4, 6, and 8 had intermediate success, attaining a 5% reduction in length but above the total energy limitations. On the other hand, Models 5 and 7 were less effective, failing to adhere to the requisite hydraulic conditions and exhibiting the lowest energy dissipation. This combination of hydraulic efficiency and structural compactness demonstrates not only improved flow control but also economic benefits through potential reductions in the construction material and land usage.

The study's findings highlight the importance of honeycomb structures for enhancing the efficiency and stability of stilling basins. The complicated architecture of these structures promotes complex flow interactions that considerably lower energy levels, allowing for a more compact basin design while maintaining hydraulic performance. This is especially useful if space or economic restrictions require a more efficient approach. The ability to maintain excellent energy dissipation within a reduced basin length emphasizes the practical benefits of using honeycomb structures in future hydraulic systems. However, the study scope was limited by excluding the evaluation of long-term structural stability, sediment transport implications, and scouring resistance of honeycomb block designs under sediment loads. Moreover, exploring alternative materials, sustainability assessments, and cost–benefit analyses for implementing such innovative geometries in real-world hydraulic structures would contribute significantly to advancing stilling basin design and application. Future work could incorporate machine learning-based surrogate models, automated parametric design, or evolutionary optimization algorithms to explore a much larger design space, which could help identify non-intuitive geometries that yield superior hydraulic performance and efficiency. Overall, this work provides an understanding of stilling basin modification, opening the way for future research focused on improving hydraulic engineering procedures.