Four-stage low pressure model steam turbine tests are carried out under the low load conditions of 0% to 20% load. In such low load conditions, the reverse flow is generated from turbine exit. Steady pressure measurements using multi-hole pneumatic probes are made to specify the outer boundary of the reverse flow region. The reverse flow regions are determined from the flow angles measured by the multi-hole pneumatic probes, traversing in the radial direction which rotates 360 deg around the longitudinal axis. The outer boundary of the reverse flow regions varies depending on turbine loads and has good agreement with the results of the numerical analyses. The pressure fluctuations are measured using unsteady pressure transducers installed on both the inner and outer side walls of the outlet stage and on the next-stage stationary blade surfaces to investigate the relation between pressure fluctuation and volumetric flow. It is found that the pressure fluctuations, which are defined by the standard deviation of unsteady pressure, become larger with decreased volumetric flow at the outer side as well as the inner side which is the same as the tendency seen for blade dynamic stress characteristics. The authors have previously reported good agreement between the experimental and numerical results. The unsteady pressure probe as another measurement technique is employed to investigate the spanwise pressure fluctuations at the outlet of the moving blade. The results show that as the load decreases, large pressure fluctuations are observed in the vicinity of the outer side after the stages where the reverse flow is observed. This is the same tendency as the results of wall pressure measurements. The generation of large pressure fluctuations, detected by the two different measurement techniques, might have a relationship with the effects of not only the vortex motion in the reverse flow region but also the overall flow field (including main forward flow) oscillated by the multiple vortex motions in the reverse flow region as seen in both experiments and computations. The large pressure fluctuations in the vicinity of the outer side after the blade lead to the increase of exciting force and vibration stress on moving blades. Detailed aerodynamic investigations of these part-load conditions are needed to analyze a blade excitation for further improvement of reliability and availability of steam turbines. The complicated flow structures at low load conditions in a steam turbine can be understood with the aid of both the steady and unsteady flow measurements and calculations.
A computational technique for multistage steam turbines, which can allow for thermodynamic properties of steam, is presented. Conventional three-dimensional multistage calculations for unsteady flows have two main problems. One is the long computation time and the other is how to include the thermodynamic properties of steam. Ideal gas is assumed in most computational techniques for compressible flows. To shorten the computational time, a quasi-three-dimensional flow calculation technique is developed. In the analysis, conservation laws for compressible fluid in axisymmetric cylindrical coordinates are solved using a finite volume method based on an approximate Riemann solver. Blade forces are calculated from the camber and lean angles of blades with momentum equations. The axisymmetric assumption and the blade force model enable the effective calculation for multistage flows, even when the flow is strongly unsteady under off-design conditions. To take into account .steam properties including effects of the gas-liquid phase change and two-phase flow, a flux-splitting procedure of compressible flow is generalized for real fluid. Density and intemal energy per unit volume are selected as independent thermodynamic variables. Pressure and temperature in a superheated region or wetness mass fraction in a wet region are calculated by using a steam table. To improve computational efficiency, a discretized steam table matrix is made in which the density and specific intemal energy are independent variables. For accuracy and continuity of steam properties, the second order Taylor expansion and linear interpolation are introduced. The computed results of the last four-stage low-pressure steam turbine at low load conditions show that there is a reverse flow near the hub region of the last .stage bucket and the flow concentrates in the tip region due to the centrifugal force. At a very low load condition, the reverse flow region extends to the former stages and the unsteadiness of flow gets larger due to many vortices. Four-stage low-pressure steam turbine tests are also carried out at low load. The radial distributions of flow direction down.stream from each stage are measured by traversing pneumatic probes. Additionally, pressure transducers are installed in the side wall to measure unsteady pressure. The regions of reverse flow are compared between computations and experiments at different load conditions, and their agreement is good. Further, the computation can follow the trends of standard deviation of unsteadv pressure on the wail to volumetric flow rate of experiments.
A computational technique for compressive fluid in multistage steam turbines which can allow for thermodynamic properties of steam is presented. The understanding and prediction of flow field not only at design conditions but also at off-design conditions are important for realizing high-performance and high-reliability steam turbines. Computational fluid dynamics is useful for estimations of flows. However, current three-dimensional multi-stage calculations for unsteady flows have two main problems. One is the long computation time and the other is how to include the thermodynamic properties of steam. Properties of the ideal gas, such as equations of state and enthalpy formula, are assumed in most computational techniques for compressible flows. In order to shorten the computation time, a quasi-three-dimensional flow calculation technique is developed. In the analysis, system equations of conservation laws for compressible fluid in axisymmetric cylindrical coordinates are solved by using a finite volume method based on an approximate Riemann solver. Blade forces are calculated from the camber and lean angles of blades using momentum equations. The axisymmetric assumption and the blade force model enable the effective calculation for multi-stage flows, even when the flow is strongly unsteady under off-design conditions. In order to take into account steam properties including effects of the gas-liquid phase change and two-phase flow, a flux-splitting procedure of compressible flow is generalized for real fluid. Density and internal energy per unit volume are selected as independent thermodynamic variables. Pressure and temperature in a superheated region or wetness mass fraction in a wet region are calculated by using a steam table. To improve computational efficiency, a discretized steam table matrix is made in which the density and specific internal energy are independent variables. For accuracy and continuity of steam properties, the second order Taylor expansion and linear interpolation are introduced. The computed results of last four-stage low-pressure steam turbine at low load conditions show that there is a reverse flow near the hub region of the last (fourth stage bucket and the flow concentrates in the tip region due to the centrifugal force. At a very low load condition, the reverse flow region extends to the former (i.e. the first to third) stages and the unsteadiness of flow gets larger due to many vortices. Four-stage low pressure steam turbine tests are also carried out at low load or even zero load. The radial distributions of flow direction downstream from each stage are measured by traversing pneumatic probes. Additionally pressure transducers are installed in the side wall to measure the unsteady pressure. The regions of reverse flow are compared between computations and experiments at different load conditions, and their agreement is good. Further, the computation can follow the trends of standard deviation of unsteady pressure on the wall to volumetric flow rate of experiments. The validity of the analysis method is verified.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.