Abstract
High-flow combined valves are critical regulating components in steam turbine systems; their flow capacity, pressure loss characteristics and flow stability across a wide range of operating conditions directly affect the economic efficiency and reliability of the unit. However, the integrated structure of combined valves complicates the throttling jet, separation recirculation and local secondary flow between the upper and lower valves, and the underlying flow mechanisms still require further elucidation. This paper employs a combined approach of numerical simulation and experimental testing to investigate high-flow combined valves under various valve opening and pressure ratio conditions. Given the complex nature of the actual filter screen structure and the difficulty of performing high-precision discretisation directly, a porous medium equivalent model is used to simulate the filter screen. The numerical model was validated using scaled experimental data and total pressure loss characteristics. On this basis, a systematic analysis was conducted of the internal flow patterns, flow regulation characteristics, vortex structure evolution, and energy dissipation patterns within the combined valve. The results indicate that, over a wide range of operating conditions, the Realizable k-ε turbulence model combined with the porous media model can predict the total pressure loss characteristics of the filter screen more accurately than the SST k-ω model. Furthermore, although the filter screen increases the total pressure loss to some extent, it improves the uniformity of the incoming flow, attenuates downstream unsteady fluctuations, and reduces flow entropy generation. This study provides a basis for optimising flow control and designing filter mesh structures in high-flow combined valves.
1 Introduction
Steam-turbine inlet valves control the steam flow entering the turbine by varying the valve lift and throat flow area and are important components for regulating the unit load. During part-load operation, the control valve reduces the turbine inlet pressure through throttling, thereby causing irreversible energy losses (Windemuth et al., 2023). The valve-head profile and variation in throat area determine the relationship between valve lift and flow capacity (Zhang and Engeda, 2003), whereas the valve chamber, filter screen, and downstream diffuser passage collectively affect pressure recovery and total pressure loss (Slama et al., 2022a). A high-flow combined valve integrates the main steam valve and control valve into a single valve casing. The upper and lower valves share a common valve seat and post-valve expansion chamber, resulting in a more compact flow path but stronger flow coupling between the two valve elements. Published studies directly addressing steam-turbine combined valves remain limited, with most previous investigations focusing on individual control valves, compact intercept valves, or valve assemblies consisting of multiple separate valves.
For individual steam-turbine control valves, Zhang et al. (2004) experimentally investigated the relationships among valve-head geometry, valve lift, and flow capacity and identified several internal flow patterns. Zanazzi et al. (2014) combined experimental measurements with numerical simulations and found that asymmetric jets, flow separation, and recirculation affect the aerodynamic performance of steam-turbine valves. Yonezawa et al. (2010) observed unsteady transonic flow states associated with valve opening and pressure ratio in a simplified steam control valve. Wang and Liu (2017a) further demonstrated that choked flow at an extremely small opening produces a local supersonic jet and pronounced downstream pressure fluctuations. A subsequent CFD investigation by Wang et al. (2018) showed that the valve-stem profile and the opening–pressure-ratio combination alter jet expansion, wall attachment, separation, and recirculation. These studies clarified the typical compressible-flow phenomena in individual control valves. However, most focused on specific small-opening, choked, or unstable operating conditions, and a systematic map of the flow-regulation and loss characteristics over wide ranges of valve opening and pressure ratio has not yet been established.
The predicted internal flow is also affected by the turbulence model and local flow-separation state. Zeng et al. (2015) compared the ability of different turbulence models to predict the flow patterns in control valves and indicated that turbulence-model selection for flows involving strong streamline curvature, jet separation, and recirculation should be evaluated against experimental data. Domnick et al. (2017) classified the flow in steam-turbine control valves into wall-attached and wall-detached jet states, showing that variations in valve opening and pressure ratio may cause transitions between different flow patterns. Clari et al. (2011) investigated three-dimensional flow separation in a control-valve diffuser and found that the non-uniform velocity distribution beneath the valve head affects diffuser flow and pressure recovery. Windemuth et al. (2021) applied spectral proper orthogonal decomposition to extract the dominant coherent flow structures and further demonstrated that the post-valve pressure-fluctuation characteristics vary with the operating condition. These studies mainly addressed flow patterns and flow instability, while the pressure losses generated in different flow regions and the migration of these losses with valve opening and pressure ratio have received less attention.
For valve assemblies and compact valves, Slama et al. (2021a) experimentally and numerically investigated the flow distribution and pressure loss in a through-flow valve chamber and demonstrated that valve-chamber geometry affects flow-field uniformity and pressure recovery. A pressure-loss model subsequently developed for an L-shaped steam-turbine inlet control-valve assembly covered multiple mass-flow rates and valve lifts and was validated using air-model experiments and numerical simulations (Slama et al., 2019). Mrozek et al. (2021) further divided the total control-valve loss into contributions from the filter screen, valve chamber, throat, and diffuser, showing that valve lift changes the contribution of each region to the total loss. The component-level analysis by Bednar et al. (2019) indicated that the filter-screen loss can account for a considerable proportion of the total pressure loss under fully open conditions and that the non-uniform diffuser-inlet velocity distribution caused by vortices beneath the valve head is another important loss source. Turecky et al. (2017) investigated pressure losses in a control-valve diffuser at different valve lifts and pressure ratios. Although these studies established relatively comprehensive component-level loss-analysis methods, they mainly considered prescribed turbine operating lines or a limited number of lift–pressure-ratio combinations and did not systematically characterize the loss mechanisms when valve opening and pressure ratio vary independently over wide ranges.
Studies of compact valves have further demonstrated that structural integration strengthens the interactions among different flow-path components. Slama et al. (2021b) used scaled experiments and CFD simulations to obtain pressure-loss and pressure-fluctuation maps of a compact valve at different mass-flow rates and valve lifts and identified stable and unstable operating regions. The overall pressure loss was subsequently reduced by modifying the valve-chamber volume and local geometry (Slama et al., 2022a). Slama et al. (2022b) also investigated the losses in the turbine inlet chamber downstream of a compact valve and found continuous flow interactions among the upstream valve, inlet chamber, and downstream nozzles. These results indicate that the specific loss sources cannot be identified solely from the pressure difference between the valve inlet and outlet; instead, the local flow fields in the valve chamber, throat, diffuser, and downstream chamber must also be considered. Such regional coupling is more pronounced in a high-flow combined valve in which the upper and lower valves share a common valve seat and expansion chamber, but the corresponding wide-operating-condition characteristics remain insufficiently understood.
The inlet filter screen further complicates the pressure-loss mechanism. Wang and Liu (2017b) found that a circular filter screen redistributes the inlet flow of the main steam valve and reduces the lateral aerodynamic force acting on the main-valve stem, but simultaneously introduces additional flow resistance and modifies the unsteady loading on the downstream control valve. Shi and Yao (2019) reported that the filter screen makes the radial inflow more uniform while redistributing the pressure losses between the main steam valve and control valve. Gao et al. (2019) determined the resistance characteristics of a main-steam-valve filter screen through wind-tunnel experiments and demonstrated that a porous-medium model can reproduce its overall pressure loss. Panuska et al. (2018) compared the theoretical resistance and actual losses of different filter screens and evaluated the contribution of filter-screen loss to the total valve loss. These studies support the use of an equivalent porous-medium model for geometrically complex filter screens. Nevertheless, existing validation has mainly focused on global pressure loss, while the influences of the filter screen on velocity uniformity, post-valve fluctuations, and local entropy generation have rarely been assessed together with its resistance penalty.
A similar trade-off between resistance and flow uniformity exists in other flow-conditioning components. Bargal et al. (2025a) found that a rotary mixer installed upstream of a multiphase pump improves oil–air mixing uniformity, but its power requirement is approximately 2.8 times that of the configuration without a mixer. In their subsequent study, pressure demand and a uniformity index were jointly used to evaluate different mixer geometries, demonstrating that improved flow uniformity may be accompanied by a considerable pressure-loss penalty (Bargal et al., 2025b). Although the oil–air two-phase flow and rotary mixer considered in those studies differ from the single-phase compressible steam and stationary filter screen investigated here, their evaluation method demonstrates that a flow-conditioning component should be assessed simultaneously in terms of its additional resistance and improvement in downstream flow quality.
Based on the literature reviewed above, several research gaps remain. First, published studies directly addressing high-flow combined valves are limited. Existing valve studies have mainly focused on flow coefficients and global flow fields and have not sufficiently clarified the coupled flow mechanism when the upper and lower valves share a common throttling passage. Second, studies of individual and compact valves have generally followed specified operating lines or considered only a small number of representative conditions. Therefore, the evolution from subsonic to transonic and locally supersonic flow when valve opening and pressure ratio vary independently remains insufficiently characterized. Third, existing pressure-loss models have mainly been established using inlet and outlet total parameters or empirical loss coefficients, while the physical mechanisms responsible for the migration of the dominant loss regions with valve opening and pressure ratio require further investigation. Finally, previous filter-screen studies have generally evaluated pressure loss and flow-straightening performance separately, and a comprehensive assessment combining additional resistance, velocity uniformity, unsteady fluctuations, and entropy generation remains lacking.
To address these research gaps, the present study combines scaled experiments with three-dimensional compressible CFD simulations to investigate the flow and pressure-loss characteristics of a high-flow combined valve over wide ranges of valve opening and pressure ratio. An equivalent porous-medium region is used to represent the double-layer composite-hole filter screen. The Realizable k-ε and SST k-ω models are evaluated using experimentally measured mass-flow rates and total-pressure losses. On this basis, the relationships among valve opening, pressure ratio, and mass-flow rate are established over a wide operating range. The evolution of local high-Mach-number flow, separation, recirculation, turbulent kinetic energy, and vortex structures is analyzed to identify the principal pressure-loss and energy-dissipation regions. The overall influence of the filter screen is further evaluated in terms of additional resistance, flow redistribution, velocity-fluctuation suppression, and entropy-generation characteristics.
2 Materials and methods
2.1 Geometric configuration and operating conditions
The high-flow combined valve investigated in this study integrates a main steam valve and a control valve within a common valve casing, as shown in Figure 1. Steam enters through the inlet on the right-hand side, passes through the filter screen, and then flows through the annular throttling passages formed by the main steam valve, control valve, and shared valve seat before being discharged through the outlet diffuser on the left-hand side. The main steam valve provides the primary isolation function, whereas the control valve regulates the mass flow rate by varying the effective throat area. The integrated configuration shortens the flow path between the two valve elements but also strengthens the aerodynamic interaction among the throttling jet, post-valve expansion region, and outlet diffuser.
FIGURE 1
Valve opening angle and pressure ratio are dimensionless parameters commonly used in the industry to describe the operating state of compound valves and reflect their performance, thereby enabling a comprehensive evaluation of the performance of high-flow compound valves. The specific definitions are given below:
The valve opening is defined as the ratio of the absolute stem travel to the maximum stem travel, as given in Equation 1:where h is the absolute stroke of the valve stem; H is the maximum stroke of the valve stem. The pressure ratio ε is defined as the ratio of the outlet static pressure to the inlet total pressure, as given in Equation 2:where P2 is the static pressure at the valve outlet; P01 is the total pressure at the valve inlet.
Based on the representative valve positions encountered during actual unit operation, valve openings of 23%, 35%, 47%, 65%, and 77% were selected for both the experiments and numerical simulations. These operating points cover the principal regulation stages of the combined valve, ranging from the small-opening condition dominated by strong throttling to the intermediate-opening range with high flow-regulation sensitivity and the large-opening condition approaching the maximum practical flow capacity. For each opening, the pressure ratio was varied from 0.1 to 0.9 to establish a wide operating-condition range covering different throttling intensities and downstream pressure conditions.
2.2 Numerical methodology
The internal flow was treated as three-dimensional, compressible, viscous and turbulent. The conservative form of the compressible Reynolds-averaged governing equations was solved together with the total energy equation. The continuity, momentum, and total-energy equations are given by Equation 3-5, respectively:where is the density, is the velocity vector, is the static pressure, T is the static temperature, E is the total energy per unit mass, is the effective stress tensor including molecular and turbulent contributions, is the effective thermal conductivity, is the momentum source term introduced by the porous screen, and is the energy source term.
The density variation was evaluated using the ideal-gas relation given in Equation 6:where R is the gas constant. The local Mach number was used to assess compressibility effects in the throttling and post-valve expansion zones and was calculated using Equation 7:where γ is the specific heat ratio. Therefore, the compressibility effects caused by local acceleration through the throttling gap and post-valve expansion were explicitly considered through the coupled solution of the momentum equation, energy equation and ideal-gas relation. By solving the coupled momentum, energy and ideal gas equations of state, this paper is able to account for density variations and compressibility effects caused by throttling acceleration and post-valve expansion.
This paper examines various flow conditions involving different flow areas and pressure ratios. Direct numerical simulation and large eddy simulation place high demands on grid resolution, time steps and computational resources, making them unsuitable for engineering-scale parametric calculations. Recent comparisons between continuous eddy simulation and resolution-imposing simulation strategies have further shown that the treatment of turbulent eddy resolution remains a central issue in high-Reynolds-number turbulent-flow simulations (Fagbade and Heinz, 2024). Consequently, for performance studies covering a wide range of operating conditions, it is necessary to employ engineering calculation methods that accurately characterise the effects of turbulence on flow and pressure loss. Predicting performance across a wide range of operating conditions requires statistically stable mass flow rates, average flow structures and total pressure losses to be obtained at each operating point. Accordingly, steady RANS calculations were used to obtain the mass-flow rate, mean flow field and total-pressure loss over the operating map, whereas unsteady RANS (URANS) calculations were conducted for the velocity-fluctuation analysis.
Among the two-equation RANS models commonly used in engineering applications, the SST k-ω and Realizable k-ε models employ different treatments of near-wall and free-shear flows. The SST k-ω model generally performs well in predicting near-wall flows involving adverse pressure gradients and flow separation (Menter, 1994). The Realizable k-ε model employs a variable eddy-viscosity coefficient and a reformulated dissipation-rate equation, improving its response to strong shear, rapid strain, streamline curvature, and recirculation (Shih et al., 1995). The combined valve contains both near-wall separation in the throttling and diffuser passages and free-shear flows associated with annular jets, separated shear layers, and post-valve mixing. To assess the applicability of the two turbulence models to the high-flow combined valve, their predictions were compared with the experimentally measured mass-flow rates and total-pressure losses. The model providing better overall agreement with the experimental results was subsequently adopted for the wide-operating-condition simulations.
The turbulent kinetic energy transport equation is given by Equation 8:
The dissipation-rate transport equation is given by Equation 9:where is the turbulent kinetic energy term generated by the mean velocity gradient, is the turbulent energy dissipation rate, and are the turbulent Prandtl numbers, C1 and C2 are model coefficients, is the kinematic viscosity, and is the mean strain rate modulus. The turbulent viscosity is calculated using Equation 10:
In the Realizable k−ε model, is a variable coefficient related to the local flow state, thereby enhancing the model’s ability to capture flows characterised by high shear, curved streamlines, rotation and recirculation.
The governing equations were solved using ANSYS Fluent 2023 R1 (ANSYS Inc, 2023) based on a cell-centred finite-volume method. A pressure-based solver was employed, and pressure–velocity coupling was achieved using the Coupled algorithm. Pressure was interpolated using a second-order scheme, whereas the momentum, energy, turbulent kinetic energy and turbulent dissipation-rate equations were discretised using second-order upwind schemes. Steady-state solutions were considered converged when all scaled residuals decreased below 1e-3 and the monitored mass-flow rate and total-pressure loss remained stable. The converged steady-state solution was subsequently used to initialise the transient calculations. A second-order implicit temporal scheme with a fixed time step of 5e-6 was adopted, and all scaled residuals were required to decrease below 1e-6 within each time step.
2.3 Equivalent models for porous media in filter screens
As the filter mesh comprises a complex structure of interlaced large and small pores, it is difficult to discretise the actual region appropriately to achieve high-precision calculations. Therefore, a porous media model is employed to replace the fine-pore structure of the filter mesh, thereby constructing a surrogate model capable of reflecting the flow characteristics and pressure loss features of the filter mesh across a wide range of flow velocities. The porous media model involves defining a geometric entity as a porous media region, specifying the porosity of that region, as well as the fluid’s flow resistance coefficient and inertial resistance coefficient, and setting the porosity. When using the porous media model in Fluent, the following assumptions and conditions apply:
The porous medium region is not modelled as an actual porous geometric body. During geometric modelling, a solid or a surface is constructed directly, rather than an actual porous body or surface containing multiple pore channels; when using the model, it is sufficient to define the porosity to indicate that the constructed solid or surface be-longs to the porous medium region.
Flow within the porous medium region is predominantly laminar; where turbulence is present, its effect on the turbulent flow field is approximated.
When defining the specific heat capacity Cp, Cp must be a constant.
The pressure drop generated by the filter screen was represented by adding the momentum source term given in Equation 11 to the governing momentum equation:where, Si represents the source term of the momentum equation in the i(x, y, z) direction, |v| denotes the velocity, D and C are specified matrices. The first term on the right-hand side represents the viscous loss term, whilst the second term represents the inertial loss term. For a homogeneous porous medium, Equation 11 reduces to Equation 12:where, represents the permeability, represents the inertial resistance coefficient. At this point, momentum acts on the fluid, generating a pressure gradient, ,where , is the thickness of the porous medium region.
There are various methods for calculating the viscous resistance coefficient and the inertial resistance coefficient; a common approach is to calculate the resistance coefficients using experimental data on the relationship between pressure drop and velocity.
The relationship between pressure drop and velocity was fitted using the quadratic expression given in Equation 13:
The viscous and inertial resistance coefficients were calculated using Equations 14, 15, respectively:
The parameters of the filter mesh studied in this paper are as follows: the wire diameter of the fine-mesh wire mesh is 0.5 mm, the center-to-center distance between wires is 1.83 mm, the aperture diameter is 4 mm, and the spacing between the upper and lower layers is 6.5 mm. The overall volume obstruction ratio of the filter mesh is 70%, which produces a significant flow-straightening effect.
To ensure that the equivalent resistance coefficient is well-suited to all operating conditions of the valve, the following 10 sets of calculation conditions have been established. The equivalent resistance coefficient in the porous medium model is derived from the flow velocity at the outlet and the pressure loss caused by the filter.
Table 1 presents the velocity–pressure drop relationship applied in this paper to the porous media model. The final fitted values are B1 = 0.3832 and B2 = 10.24.
TABLE 1
| Number | Inlet velocity (m/s) | Outlet velocity (m/s) | Pressure loss (Pa) |
|---|---|---|---|
| 1 | 1 | 1.82 | 36.84 |
| 2 | 3 | 5.23 | 286.69 |
| 3 | 5 | 8.63 | 772.63 |
| 4 | 7 | 12.09 | 1506.15 |
| 5 | 9 | 15.63 | 2507.41 |
| 6 | 11 | 19.27 | 3805.91 |
| 7 | 12 | 21.14 | 4579.30 |
| 8 | 13 | 23.04 | 5443.91 |
| 9 | 14 | 24.99 | 6407.79 |
| 10 | 15 | 26.99 | 7481.65 |
Velocity-pressure-drop data used for porous-medium resistance fitting.
To further assess the capability of the equivalent porous-medium model to predict the local flow response of the filter screen, a 1/16 periodic geometrically resolved filter-screen model retaining the actual hole diameter, hole spacing, and screen thickness was established, together with an equivalent porous-medium model using the resistance coefficients determined from the velocity–pressure-drop relationship. Identical inlet and outlet boundary conditions, fluid properties, and numerical discretization schemes were applied to both models. The resulting velocity distributions are compared in Figure 2. In the geometrically resolved filter-screen model, discrete high-velocity jets form at the outlets of the individual perforations and subsequently undergo expansion and mixing. Because the individual perforations are not explicitly resolved in the equivalent porous-medium model, these near-field jets are spatially averaged. Nevertheless, the equivalent model reproduces the principal macroscopic flow characteristics of the geometrically resolved structure, including flow acceleration through the filter-screen region, downstream expansion of the high-velocity flow, and the overall velocity level after jet mixing. The discrepancies between the two models are primarily confined to the immediate vicinity of the perforation outlets, while their velocity distributions become progressively similar farther downstream. This comparison indicates that the porous-medium model is suitable for predicting the overall flow resistance and large-scale flow redistribution of the filter screen in the complete combined-valve simulation.
FIGURE 2
2.4 Analysis of grid-independence and experimental validation
Polyhedral meshes were generated using ANSYS Fluent Meshing 2023 R1, and a mesh-independence study was performed to balance computational cost and numerical accuracy. Polyhedral meshes are well-suited to representing the complex geometry of the combined valve and are often the preferred choice in engineering applications. Based on the dimensions of the geometric model, the minimum mesh size was set at 0.42 mm, whilst the maximum mesh sizes were set at 4 mm, 6 mm, 8 mm, 10 mm, 12 mm and 13 mm respectively. The expansion ratio for the transition mesh was set to 1.2, and the angle of the curved features was set to 12°. The inlet, porous medium and outlet were meshed separately, with nodes shared at the junctions; the number of mesh elements for the six mesh configurations was 1 million, 1.2 million, 1.9 million, 2.2 million, 3.1 million and 3.6 million. The results of the mesh independence study are shown in Figure 3. For valve lifts of 2 mm, 8.5 mm and 22.8 mm, and inlet pressures ranging from 10 kPa to 145 kPa, the meshing strategy with a maximum mesh size of 8 mm provided a balance under different valve lift and pressure ratio conditions; as the mesh size decreased, the steady-state flow rate of the valve did not change significantly. Consequently, the maximum mesh size for subsequent calculations was set at 8 mm, with approximately 1.9 million mesh elements.
FIGURE 3
Guided by similarity theory, scaled-down tests can be conducted using reduced-scale test valves. Scaled-down tests allow for flexible adjustment of valve opening and the placement of measurement points as required, enabling comprehensive measurement of flow field parameters within the combined valve to characterize all aspects of the test valve’s performance. Current standard methods are all based on scaled-down testing. The principal measured quantities in the scaled combined-valve experiments included valve lift, volumetric flow rate, pressure, temperature, lifting force, vibration displacement and sound pressure. Static parameters were acquired using the IMP data-acquisition system, whereas the vibration and acoustic signals were recorded using the LMS Test. Lab dynamic data-acquisition system. The instruments, measurement ranges and corresponding accuracy or acquisition specifications are summarized in Table 2. As shown in the Figure 4, the scaled-down test system for high-flow combined valves comprises the test bench itself, the air supply system, the piping system, the lifting system and the measurement system.
TABLE 2
| Measured quantity/item | Instrument | Model and specification | Measurement accuracy/acquisition specification |
|---|---|---|---|
| Valve lift | Vernier caliper | Type II vernier instrument | Resolution: 0.02 mm |
| Valve opening | Digital dial indicator | B15D273104 electronic digital dial indicator | Resolution: 0.01 mm |
| Volumetric flow rate | Vortex flowmeter | Prowirl 200 7F2C1H | Accuracy: 0.75% |
| Temperature | T-type thermocouple | RTK-281; measurement range: 0 °C–350 °C | Resolution/accuracy: 0.1 °C |
| Pressure | Multi-channel pressure scanner | 16 channels; measurement ranges: 0–0.2 MPa and 0–1 MPa | Accuracy: 0.075% |
| Reference pressure | Floating-ball pressure calibrator | Y055; measurement range: 0.005–0.6 MPa | Accuracy: 0.05% |
| Vibration displacement | Eddy-current displacement sensor | CWY-DO-810508-02-05-10-02 | Accuracy: 0.7% |
| Sound pressure | 1/2-inch free-field microphone and preamplifier | PCB 378B02; microphone: 37702; preamplifier: 426E01 | Frequency response: 3.75 Hz–20 kHz (±2 dB); 7 Hz–10 kHz (±1 dB) |
| Acoustic calibration | Acoustic calibrator | Larson Davis CAL200; calibration levels: 94 and 114 dB | Harmonic distortion: 2% |
| Valve lifting force | Tension–compression load cell | LSR-2 | Accuracy: 0.05% |
| Static data acquisition | Static data acquisition system | IMP data-acquisition unit and interface board | Sampling interval: 0.1 s |
| Dynamic data acquisition | Dynamic data acquisition system | LMS Test.Lab | Maximum sampling frequency: 200 kHz; A/D resolution: 24 bit |
Instruments and measurement specifications used in the scaled combined-valve experiments.
FIGURE 4
A comparison was made between the total pressure loss obtained from tests on a 70-degree blind-end filter screen and the results calculated using the Realizable k-ε turbulence model combined with the porous media model, as well as the SST k-ω model. As shown in Figure 5, the combination of a realizable k-ε turbulence model and a porous medium model enables a more accurate reproduction of the total pressure loss across a range of lift heights and at higher pressure ratios. In contrast, the SST k-ω model combined with the porous media model yields slightly overestimated total pressure loss results under conditions of small head and low pressure ratio. Therefore, the Realizable k−ε model combined with the porous-media representation was adopted for the subsequent simulations because it reproduced the experimentally measured global pressure-loss characteristics more accurately than the SST k−ω model.
FIGURE 5
In this paper, the CFD simulations were performed using the internal fluid domain of the combined valve. The inlet boundary was specified as a pressure inlet, and the inlet total pressure was prescribed according to each operating condition. The outlet boundary was specified as a pressure outlet, and the outlet static pressure was determined from the pressure ratio ε. The investigated pressure-ratio range was 0.1–0.9. The main valve-opening conditions were 23%, 35%, 47%, 65% and 77%. Additional representative openings of 10%, 47% and 100% were used for internal-flow visualization.
The working fluid in the simulation was treated as a compressible ideal gas. The energy equation was enabled to account for density variation associated with throttling acceleration and post-valve expansion. All solid surfaces, including the valve body, valve head, valve stem, valve seat and outlet pipe wall, were treated as no-slip adiabatic walls. The filter screen was replaced by an equivalent porous-medium region, in which the viscous and inertial resistance coefficients were specified from the pressure-drop fitting of the screen. Enhanced Wall Treatment was employed for the Realizable k−ε calculations. The y+ values over the principal wetted surfaces were predominantly greater than 30, indicating that the wall-adjacent cells were mainly located within the logarithmic region and that the wall-function branch of the enhanced treatment was active over most of these surfaces.
3 Results and discussion
3.1 Analysis of flow control characteristics under various operating conditions
To obtain the flow characteristics of the valves across a wide pressure range, CFD methods were employed to calculate and analyze the flow patterns within the valves, with the aim of comprehensively analyzing the relationship between flow rate and valve opening, as well as pressure ratio. The upper valve, functioning as a control valve, was analyzed at multiple opening positions, whilst the lower valve, serving as a main steam valve, is primarily operated in fully open or fully closed conditions in practical applications; consequently, only three opening positions were analyzed for the lower valve, with experimental verification conducted at a 47% opening. As shown in Figure 6, the relative deviations between the CFD-predicted and experimentally measured mass-flow rates remained below 5% over the experimental pressure-ratio range, and the two datasets exhibited consistent trends. As shown in the figure, above a pressure ratio of 0.3, the flow rate exhibits a linear relationship with the pressure ratio; the relationship between flow rate and pressure ratio at high pressure ratios can be derived using the ideal gas equation and Bernoulli’s equation. Below a pressure ratio of 0.3, the gradient of flow rate variation increases as the pressure ratio decreases. This is due to the emergence of supersonic regions under low-pressure-ratio conditions, where the compressibility of the fluid cannot be neglected. Pressure variations in the flow become more complex, interrelated with velocity and density; furthermore, the derivation of the flow equation is com-plicated by the effects of shock waves and expansion waves.
FIGURE 6
As shown in Figure 7, the flow-opening curve of the combined valve across all operating conditions exhibits a pattern of rapid initial growth followed by a slower rate. At low opening levels, flow rates are low across all pressure ratios; the combined valve demonstrates good low-flow control capability. At medium opening levels, flow increases most rapidly under low-pressure-ratio conditions, with high control sensitivity, but this is prone to significant flow fluctuations; At high opening levels, as the valve opens further, the rate of flow increase decreases significantly, the valve’s flow capacity approaches its upper limit, and the control capability of the combined valve is further reduced.
FIGURE 7
To better guide the control of combined valves, this study identifies the linear, equal-percentage and quick-opening characteristic regions for such valves, whilst ensuring that actual operating conditions do not fall within the non-linear region. This paper analyses the valve flow characteristics and control capability from the perspective of the MAP diagram covering the full operating range. As shown in the Figure 8, within the primary operating zone, the flow varies almost linearly with the valve opening, indicating good control performance. Calculations of the determinant of the flow function coefficients using the Hessian matrix reveal values consistently below zero, indicating the absence of local extrema. By fitting a function, the trend of flow variation across a wide range of operating conditions can be clearly observed. A second-order response surface polynomial was established using regression analysis to fit the flow function, as shown in the equation; the R-squared value of the function is 0.97387.
FIGURE 8
3.2 Analysis of internal flow characteristics under various operating conditions
As a fluid flows through a valve, significant velocity gradients and pressure variations may occur in localised regions due to the valve’s geometry and opening. Regions with high velocity gradients typically correspond to locations with strong flow shear, and these areas are often the primary sites for the initiation and development of turbulence. Furthermore, regions with high pressure losses indicate significant energy dissipation, which may be accompanied by vortex shedding, flow separation and the generation of secondary flows. Moreover, to comprehensively evaluate the turbulence characteristics within the valve, turbulent kinetic energy and the turbulent dissipation rate are employed as key parameters for assessment.
For the upper valve of a high-flow combined valve shown in Figure 9, at a 10% opening, the gas flow exhibits wall-following flow after passing through the annular gap, detaches at the right-angled turning wall, and the lower end of the lower valve head is completely occupied by a low-static-pressure recirculation region. At 65% opening, secondary flow occurs within the internal bore of the lower valve plug; the squeeze flow between the upper and lower valves prevents the main jet in the gap from fully expanding, resulting in a secondary supersonic phenomenon. The separated low-pressure region at the lower end of the lower valve head is reduced, with a localized separated low-pressure region present at the bend in the outlet structure. At 100% opening, the flow develops fully; the separated low-pressure region at the lower end of the lower valve head no longer exists, and the high-speed jet, after being reflected by the lower valve seat into the outlet duct, remains supersonic.
FIGURE 9
For the lower valve of the high-flow combined valve shown in Figure 9, at a small opening of 10%, the gas flow exhibits wall-following characteristics after passing through the annular gap channel; however, separation occurs at the right-angled turning wall, resulting in the lower end of the lower valve head being completely occupied by a separated low-pressure region. At an opening of 47%, a secondary flow phenomenon is observed within the internal bore of the lower valve plug. Concurrently, the squeeze flow between the upper and lower valves prevents the main jet from fully expanding, resulting in secondary supersonic flow. Compared to the low-opening condition, the separated low-pressure region at the lower end of the lower valve head is significantly reduced, though localized separated low-pressure regions still persist at the bends in the outlet structure. At 100% opening, the flow develops fully, and the separated low-pressure region at the lower end of the lower valve head no longer exists. After being reflected by the lower valve seat and entering the outlet duct, the high-speed jet continues to maintain a supersonic state.
For a high-flow combined valve, the region of high turbulent kinetic energy at a small upper valve opening is concentrated between the upper and lower valve heads and the lower valve seat; at a medium valve opening, it is concentrated in the cavity; and at a large valve opening, it is concentrated at the inlet of the outlet pipe. This is be-cause, as the valve opening decreases, a high-velocity jet forms a strong shear layer be-tween the upper and lower valve heads and the lower valve seat. Here, the velocity gradient is greatest, inducing Kelvin–Helmholtz instability and generating periodic vortex shedding, resulting in peak turbulent kinetic energy in this region. The region of high turbulent kinetic energy at a small lower valve opening is concentrated within the inner bore of the lower valve stem; at a medium opening, the high turbulent kinetic energy propagates rapidly through the cavity region; and at a large opening, the high turbulent kinetic energy is concentrated at the inlet of the outlet pipe. Under different operating conditions, the turbulent kinetic energy distributions of the gradually varying throat valve and the high-flow combined valve are similar. The distinction lies in the fact that, at a small opening of the upper valve, the position of the lower valve in the high-flow combined valve is subject to disturbance, whereas at a small opening of the lower valve, the position of the lower valve stem in the gradually varying throat valve is subject to disturbance; this is attributable to the throat design and the coupling structure between the upper and lower valves.
The vortex structures within the combined valve were identified using the Q-criterion. As shown in Figure 10, the vortex cores are mainly concentrated in the unloading chamber and the post-valve separation cavity at the 10% opening. With increasing valve opening, the vortex structures gradually extend towards the outlet pipes and the upstream and downstream separation cavities. The area-averaged Q values on the selected cross-section are 4.22 e6,7.24 e6,8.87 e6at the 10%, 47%, and 100% openings, respectively. The value increases by 71.8% from the 10% to the 47% opening and by a further 22.5% from the 47% to the 100% opening, resulting in an overall increase of 110.4%. This variation indicates that rotational motion becomes increasingly dominant over strain on the selected cross-section as the flow passage opens. At small openings, the vortex structures are primarily associated with the high-speed throttling jet, flow separation, and wall shear near the valve throat. At the fully open condition, the weakening of the local throttling restriction allows the high-momentum flow to propagate farther downstream, where flow turning, jet impingement, and the merging of the two outlet streams promote the development of larger-scale vortex structures. Following the valve, the fluid exhibits both wall-following flow and central flow due to the Conda effect. The central flow maintains its velocity as it strikes the wall within the separation chamber, where multiple streams merge to form a vortex, which then exits through the two outlet pipes. The static pressure at the core of the vortex can drop to negative pressure. If the gas stream carries solid dust particles, these particles will accumulate within the separation chamber. The stronger the vortex, the faster the particles move, potentially causing impact on the lower valve seat and the walls of the separation chamber. However, the presence of a filter screen allows larger particles to be filtered out, ensuring the safe operation of the outlet pipes and enhancing the stability of the system.
FIGURE 10
The iso surface plot reveals that, at small valve openings, wall shear flow occurs within the combined valve, whilst vortex flow develops at the centre of the outlet duct. The wall shear flow is accompanied by the dissipation of shock waves and expansion waves resulting from the separation of high-velocity flow. The curved surface of the valve seat induces the generation of expansion waves, causing the flow direction to gradually deflect towards the wall. As the valve opening increases, the vortex flow in the core region is formed by the reflection of the post-valve flow entering the separation chamber, creating a strong secondary flow. At the junction between the separation chamber and the outlet duct, the flow cross-section constricts, causing the vortex flow to accelerate in the axial direction of the duct. The swirling flow is stretched, resulting in a slender, tornado-like low-pressure region in the core area.
3.3 Analysis of the function of filter meshes and energy loss characteristics
A filter screen can improve the uniformity of fluid flow at small valve openings. As shown in Figure 11, there is a certain difference in the distribution of flow lines within the valve with and without a filter screen. In the absence of a filter screen, the incoming steam flow around the valve stem resembles a cylindrical flow pattern; however, due to the constraints of the wall surface and the high velocity of the incoming flow, a concentration of flow lines occurs on the side opposite the incoming flow. The filter screen, on the other hand, causes the incoming flow to flow along the wall surface in a direction normal to the screen structure towards the valve stem wall; this flow pattern enhances the uniformity of the lateral aerodynamic forces acting on the valve stem.
FIGURE 11
Based on the flow velocity contour plots shown in Figure 12, the influence of the filter screen structure on the uniformity of the flow field was further analysed. When the incoming flow strikes the surface head-on, a low-velocity zone is formed. The design of this filter screen incorporates a non-perforated section directly opposite the inlet, shifting this low-velocity zone from the valve stem position to the closed end of the filter screen, thereby significantly reducing the frontal impact force of the incoming flow on the valve stem. With this filter screen structure in place, the flow velocity at the valve stem is lower and more uniform, with no severe turbulence or local disturbances.
FIGURE 12
In a configuration without a filter screen, the valve stem exhibits significant temperature and pressure non-uniformity; local variations in flow velocity and turbulence lead to uneven heat exchange, thereby creating localised areas of undercooling. Following the installation of the filter screen, localised vortices appear at the bend on the closed end of the screen, and the undercooled region shifts to the area between the closed end of the screen and the upper part of the valve body. The pressure distribution is broadly symmetrical between the upper and lower parts of the valve body; the filter screen effectively promotes uniform fluid distribution within the valve, reduces unstable flow, and results in a more uniform temperature distribution across the valve stem surface. It is evident that the filter screen structure exerts a controlling influence on the flow within the valve, thereby enhancing the valve’s stability and reliability.
As shown in Figure 13, the velocity-fluctuation spectra at the valve-seat monitoring point vary considerably with the filter configuration and valve opening. Under the fully open condition, the maximum power spectral density decreases from approximately 7.5 without the filter screen to 4.5 with the 70° blind-end filter, corresponding to a reduction of about 40%. At the 35% opening, the maximum value in the low-frequency range decreases from approximately 5.5 to 4.0, representing a reduction of about 27%. The configuration without the filter also exhibits a pronounced high-frequency peak of approximately 9.0 near 9 kHz, whereas this component is almost eliminated after the filter is installed. The redistribution of the incoming flow by the filter therefore reduces the overall fluctuation amplitude and weakens the high-frequency instability generated by the throttling jet under the small-opening condition.
FIGURE 13
The energy-loss composition in the filter-screen region differs from that of the complete combined-valve flow domain. As shown in Figure 14, both viscous dissipation and turbulence-production loss generally increase with the valve opening and pressure ratio. At a pressure ratio of 0.9, the viscous-dissipation proportion increases from approximately 1.0% at the 23% opening to 2.8%, 4.3%, 5.0%, and 5.5% at the 35%, 47%, 65%, and 77% openings, respectively, and reaches approximately 5.9% under the fully open condition. By comparison, the turbulence-production proportion remains below approximately 2.3% over all the investigated openings, with values of approximately 0.65%, 1.15%, 1.9%, 2.25%, and 1.9% at the 35%, 47%, 65%, 77%, and 100% openings, respectively. The ratio of turbulence production to viscous dissipation remains below unity over most operating conditions, indicating that the additional loss generated locally within the filter-screen region is primarily associated with viscous dissipation. At small openings, the filter accounts for only a limited proportion of the total loss because the dominant irreversible loss remains concentrated at the valve throat.
FIGURE 14
Figure 15 shows that the reduction in flow non-uniformity is accompanied by an additional pressure-loss penalty. At a representative pressure ratio of 0.5, the total pressure loss is approximately 0.25 kPa without the filter screen and increases to approximately 0.45 and 0.50 kPa for filter-screen thicknesses of 6 and 12 mm, respectively. The corresponding increases are approximately 80% and 100%. In contrast, the entropy generation decreases from approximately 1.50 W/K without the filter screen to 0.78 and 0.45 W/K for the 6 and 12 mm filter screens, representing reductions of approximately 48% and 70%, respectively. Increasing the screen thickness from 6 to 12 mm produces an additional pressure-loss increase of approximately 11%, while reducing the entropy generation by a further 42%. The thicker screen therefore strengthens the suppression of irreversible loss associated with large-scale flow non-uniformity, although the improvement is accompanied by a moderate increase in local flow resistance.
FIGURE 15
4 Conclusion
This study combined scaled experiments with three-dimensional compressible CFD simulations to investigate the flow-regulation characteristics, internal flow structures, vortex evolution, and filter-screen effects of a high-flow combined valve over wide ranges of valve opening and pressure ratio. The main conclusions are as follows:
The Realizable k-ε model combined with the equivalent porous-medium model reproduced the experimentally measured total-pressure-loss characteristics more accurately than the SST k-ω model, particularly at small openings and low pressure ratios. Comparison with the geometrically resolved filter-screen model further showed that the equivalent model reproduced the macroscopic flow acceleration, downstream expansion, and velocity distribution after jet mixing, supporting its application to the complete combined-valve simulation.
The mass flow rate increased rapidly at small and medium openings and approached its upper limit at large openings. When the pressure ratio exceeded approximately 0.3, the mass flow rate varied approximately linearly with the pressure ratio. Below 0.3, local supersonic regions, shock waves, and expansion waves developed around the throttling gap and post-valve expansion region, resulting in pronounced nonlinear flow behaviour. The response-surface model describing the flow characteristics over the investigated operating range achieved a coefficient of determination of R2 = 0.98
As the valve opening increased, the dominant high-turbulence and vortex regions migrated from the throttling gap and post-valve separation cavity towards the outlet pipes. The area-averaged Q values on the selected cross-section increased from 4.22e6 at 10% opening to 8.87e6 at 100% opening, corresponding to an overall increase of 110.4%. This indicates that rotational motion becomes progressively more dominant over strain as the flow passage opens.
The 70° blind-end filter screen redistributed the incoming flow and suppressed velocity fluctuations near the valve seat. The maximum velocity-fluctuation power spectral density decreased by approximately 40% under the fully open condition and by approximately 27% at 35% opening, while the high-frequency component near 9 kHz was almost eliminated. At a pressure ratio of 0.5, increasing the filter-screen thickness from 0 to 6 and 12 mm increased the total pressure loss from approximately 0.25 kPa to 0.45 and 0.50 kPa, respectively, but reduced entropy generation from approximately 1.50 W/K to 0.78 and 0.45 W/K. Therefore, the filter-screen thickness should be selected by balancing the additional resistance against the improvements in flow uniformity and stability.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Author contributions
FS: Methodology, Validation, Investigation, Conceptualization, Formal Analysis, Writing – original draft. ZF: Methodology, Software, Writing – review and editing, Validation. XuZ: Project administration, Funding acquisition, Writing – review and editing. ZZ: Writing – review and editing, Software. KB: Data curation, Writing – review and editing, Formal Analysis. XiZ: Resources, Writing – review and editing, Supervision. JS: Writing – review and editing, Data curation.
Funding
The author(s) declared that financial support was received for this work and/or its publication. The authors gratefully acknowledge the financial support from the Open Project of the State Key Laboratory of Clean and Efficient Turbomachinery Power Equipment (Grant No. DEC8300CG202419299-A1228110).
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
ANSYS Inc (2023). ANSYS Fluent Theory Guide, Release 2023 R1. Canonsburg, PA, USA: ANSYS Inc.
2
BargalM. H. S.Ben-MansourR.Al-SarkhiA.AlhemsL. M. (2025a). CFD modeling of two-phase flow (oil/air) with and without rotary mixer inside a vertical pipe for upstream of multiphase pump. Arabian J. Sci. Eng.50, 9611–9630. 10.1007/s13369-024-09789-7
3
BargalM. H. S.Ben-MansourR.Al-SarkhiA.AlhemsL. M. (2025b). Geometrical modifications of rotary mixers within a vertical pipe for improving oil-air flow mixing in ESP applications: a computational approach. Results Eng.27, 106796. 10.1016/j.rineng.2025.106796
4
BednarL.TajcL.MiczanM.HoznedlM. (2019). Pressure loss analysis in the steam turbine control valve. Eng. Mech., 25, 45–48. 10.21495/71-0-45
5
ClariM. B.PolklasT.JoosF. (2011). “Three-dimensional flow separations in the diffuser of a steam turbine control valve,” in Proceedings of ASME Turbo Expo 2011: Turbine Technical Conference and Exposition. Vancouver, Canada, GT2011–45617.
6
DomnickC. B.BenraF. K.BrillertD.DohmenH. J.MuschC. (2017). Clarifying the physics of flow separations in steam turbine inlet valves at part load operation and improved design considerations. J. Fluids Eng.139 (8), 081105. 10.1115/1.4036263
7
FagbadeA.HeinzS. (2024). Continuous eddy simulation vs. resolution-imposing simulation methods for turbulent flows. Fluids9, 22. 10.3390/fluids9010022
8
GaoQ.DingL.HuangD. (2019). Experimental and numerical study on loss characteristics of main steam valve strainer in steam turbine. Appl. Therm. Eng.147, 935–942. 10.1016/j.applthermaleng.2018.07.031
9
MenterF. R. (1994). Two-equation eddy-viscosity turbulence models for engineering applications. AIAA J.32 (8), 1598–1605. 10.2514/3.12149
10
MrozekL.SlamaV.SedlakK.TajcL. (2021). Analysis of losses in steam turbine control valves. AIP Conf. Proc.2323, 060003. 10.1063/5.0042097
11
PanuskaM.TajcL.KollrossP. (2018). The effect of a strainer in the control valve of the steam turbine. AIP Conf. Proc.2047, 020015. 10.1063/1.5081648
12
ShiL.YaoS. (2019). Influence of a circular filter on super-critical steam flow through main stop and control valves in a 1000 MW super-critical power plant. Arabian J. Sci. Eng.44, 959–970. 10.1007/s13369-018-3333-9
13
ShihT. H.LiouW. W.ShabbirA.YangZ.ZhuJ. (1995). A new k-ε eddy viscosity model for high reynolds number turbulent flows—Model development and validation. Comput. and Fluids24 (3), 227–238. 10.1016/0045-7930(94)00032-t
14
SlamaV.MrozekL.TajcL.SimurdaD.HalaJ. (2019). Creation and validation of pressure loss model for steam turbine control valves. AIP Conf. Proc.2189, 020020. 10.1063/1.5138632
15
SlamaV.MrozekL.TajcL.KlimkoM.ZitekP. (2021a). Flow analysis in a steam turbine control valve with through-flow valve chamber. J. Nucl. Eng. Radiat. Sci.7 (2), 021401. 10.1115/1.4049055
16
SlamaV.SimurdaD.MrozekL.TajčL.HálaJ.RadnicT. (2021b). Pressure losses and oscillations in a compact valve of a steam turbine. MATEC Web Conf.345, 00027. 10.1051/matecconf/202134500027
17
SlamaV.SimurdaD.TajcL.RudasB.HalaJ.LuxaM.et al (2022a). Pressure losses reduction in a steam turbine compact valve. EPJ Web Conf.264, 01042. 10.1051/epjconf/202226401042
18
SlamaV.SimurdaD.LenhardR. (2022b). Pressure losses downstream of a compact valve in the inlet chamber of an intermediate-pressure steam turbine. Energies15 (22), 8753. 10.3390/en15228753
19
TureckyP.MrozekL.TajcL.KolovratníkM. (2017). Analysis of pressure losses in the diffuser of a control valve. EPJ Web Conf.143, 02135. 10.1051/epjconf/201714302135
20
WangP.LiuY. (2017a). Unsteady flow behavior of a steam turbine control valve in the choked condition: field measurement, detached eddy simulation and Acoustic modal analysis. Appl. Therm. Eng.117, 725–739. 10.1016/j.applthermaleng.2017.02.087
21
WangP.LiuY. (2017b). Influence of a circular strainer on unsteady flow behavior in steam turbine control valves. Appl. Therm. Eng.115, 463–476. 10.1016/j.applthermaleng.2016.12.073
22
WangP.MaH.QuayB. D.SantaviccaD. A.LiuY. (2018). Computational fluid dynamics of steam flow in a turbine control valve with a bell-shaped spindle. Appl. Therm. Eng.129, 1333–1347. 10.1016/j.applthermaleng.2017.10.104
23
WindemuthC.LangeM.MailachR. (2021). Analysis of the unsteady flow field in a steam turbine control valve using spectral proper orthogonal decomposition. Int. J. Turbomach. Propuls. Power6 (2), 11. 10.3390/ijtpp6020011
24
WindemuthC.LangeM.MailachR. (2023). Comparison of steam turbine control valve geometries and their dynamic behavior at part load. Int. J. Turbomach. Propuls. Power8 (4), 55. 10.3390/ijtpp8040055
25
YonezawaK.ToyohiraY.NagashimaM.TsujimotoY.TezukaK.MoriM.et al (2010). An experimental study of unsteady transonic flow in a steam control valve with a simple model. J. Environ. Eng.5 (1), 134–143. 10.1299/jee.5.134
26
ZanazziG.RiceT.SellM.RidouttC. (2014). Experimental and numerical investigation into the aerodynamics of a novel steam turbine valve and its field application. J. Eng. Gas Turbines Power136 (9), 091601. 10.1115/1.4026860
27
ZengL. F.LiuG. W.MaoJ. R.YuanQ.WangS.WeiL.et al (2015). A novel numerical simulation method to verify turbulence models for predicting flow patterns in control valves. J. Fluid Sci. Technol.10 (1), JFST0007. 10.1299/jfst.2015jfst0007
28
ZhangD.EngedaA. (2003). Venturi valves for steam turbines and improved design considerations. Proc. Institution Mech. Eng. Part A J. Power Energy217 (2), 219–230. 10.1243/09576500360611254
29
ZhangD.EngedaA.HardinJ. R.AungierR. H. (2004). Experimental study of steam turbine control valves. Proc. Institution Mech. Eng. Part C J. Mech. Eng. Sci.218 (5), 493–507. 10.1243/095440604323052283
Summary
Keywords
filter screen, flow characteristics, high-flow combined valve, large-flow coefficient, variable operating conditions characteristics
Citation
Song F, Feng Z, Zhang X, Zhong Z, Bai K, Zhang X and Song J (2026) Research on flow and pressure drop characteristics of high-flow combined valves under wide operating conditions. Front. Mech. Eng. 12:1900885. doi: 10.3389/fmech.2026.1900885
Received
05 June 2026
Revised
04 August 2026
Accepted
05 August 2026
Published
31 August 2026
Volume
12 - 2026
Edited by
Amirul Khan, University of Leeds, United Kingdom
Reviewed by
Mohamed H. S. Bargal, Minia University, Egypt
Yachao Ma, Southwest Petroleum University, China
Updates
Copyright
© 2026 Song, Feng, Zhang, Zhong, Bai, Zhang and Song.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Xu Zhang, 977301567@qq.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.