The Pressure Field at the Output From a Low Pressure Exhaust Hood and Condenser Neck of the 1090 MW Steam Turbine: Experimental and Numerical Research

Author(s):  
Michal Hoznedl ◽  
Antonín Živný ◽  
Aleš Macálka ◽  
Robert Kalista ◽  
Kamil Sedlák ◽  
...  

The paper presents the results of measurements of flow parameters behind the last stage of a 1090 MW nominal power steam turbine in a nuclear power plant. The results were obtained by traversing a pneumatic probe at a distance of about 100 mm from the trailing edges of the LSB (Last Stage Blade). Furthermore, both side walls as well as the front wall of one flow of the LP (Low Pressure) exhaust hood were fitted with a dense net of static pressure taps at the level of the flange of the turbine. A total of 26 static pressures were measured on the wall at the output from the LP exhaust hood. Another 14 pressures were measured at the output from the condenser neck. The distribution of static pressures in both cross sections for full power and 600 and 800 MW power is shown. Another experiment was measured pressure and angle distribution using a ball pneumatic probe in the condenser neck area in a total of four holes at a distance up to 5 metres from the neck wall. The turbine condenser is two-flow design. In one direction perpendicular to the axis of the turbine cold cooling water comes, it heats partially. It then reverses and it heats to the maximum temperature again. The different temperature of cooling water in the different parts of the output cross section should influence the distribution of the output static pressure. Differences in pressures may cause problems with uneven load of the tube bundles of the condenser as well as problems with defining the influential edge output condition in CFD simulations of the flow of the cold end of the steam turbine Due to these reasons an extensive 3D CFD computation, which includes one stator blade as well as all moving blades of the last stage, a complete diffuser, the exhaust hood and the condenser neck, has been carried out. Geometry includes all reinforcing elements, pipes and heaters which could influence the flow behaviour in the exhaust hood and its pressure loss. Inlet boundary conditions were assumed for the case of both computations from the measurement of the flow field behind the penultimate stage. The outlet boundary condition was defined in the first case by an uneven value of the static pressure determined by the change of the temperature of cooling water. In the second case the boundary condition in accordance with the measurement was defined by a constant value of the static pressure along all the cross section of the output from the condenser neck. Results of both CFD computations are compared with experimental measurement by the distribution of pressures and other parameters behind the last stage.

Author(s):  
Tadashi Tanuma ◽  
Hiroshi Okuda ◽  
Gaku Hashimoto ◽  
Satoru Yamamoto ◽  
Naoki Shibukawa ◽  
...  

The aim of this paper is to present some of research results of our current collaborative program to increase steam turbine efficiency with the development of high-performance blade and exhaust hood design methodology using large-scale aerodynamic and structural interaction analysis. Aerodynamic optimum designs of stator blades are already introduced in many designs of actual operating commercial steam turbine units. However, aerodynamic optimum designs of rotating blades are still difficult due to high centrifugal force and vibration stress on rotating blades. This paper focuses on rotating blades and exhaust diffusers that affect the flow field just downstream of last stage long blades. The large-scale high-accuracy CFD analysis of unsteady wet steam flows has been successfully introduced for simulations of low pressure exhaust diffuser using the Earth Simulator of Japan Agency for Marine-Earth Science and Technology. This result shows that the diffuser domain analysis can provide static pressure recovery coefficients and its circumferential deviations with enough accuracy for design use except correct location predictions of separations. The unsteady flow analyses of the typical designed last stage with the measured and calculated downstream static pressure distribution as the outlet boundary condition were conducted. The unsteady flow analyses of the typical designed low pressure exhaust diffuser with the measured and calculated upstream flow conditions as the inlet boundary condition were also conducted. Some of the calculated results were compared with measured data. The large-scale parallel computing Finite Element Analysis of turbine blades with inter-connection parts has been also successfully introduced on the Earth Simulator. The calculation result shows that the eigen frequencies of the present group of loosely-connected rotating blades correspond well to the existing measured data. For the next step, the unsteady structural analysis is being conducted with the calculated unsteady forces on the rotating blades as the FEA boundary conditions. Some of the FEA results are also presented in this paper.


Author(s):  
Liu Meng ◽  
Chen Yang ◽  
Zhong Zhuhai ◽  
Zhang Xiaodan ◽  
Deng Guoliang ◽  
...  

Kinetic energy recovery is a key objective for low pressure exhaust hood design and optimization. Numerical simulation of the exhaust hood helps the engineers to explore and confirm the causes of the loss in the hood. Many studies have suggested that it is necessary for the simulation to include the last stage blade to get a realistic assessment. For the sole exhaust hood study, the inlet boundary condition is hard to set precisely like the downstream flow of the last stage blade. And the studies have also shown that the performances generated from the simulations may vary evidently between the sole exhaust hood and exhaust hood with last stage blade. It is obvious that the blade influences the exhaust hood, but the exact effect factors of the blade and the way they work are not thoroughly discussed. This paper has conducted many numerical tests to audit the influence of the common effect factors of the last stage blade. The internal flow field of the exhaust hood was numerically investigated using three-dimensional Reynolds-Averaged Navier-Stokes (RANS) solutions based on the ANSYS-CFX. In the first part of the paper, the tests are conducted by changing each effect factor of the inlet boundary condition for sole exhaust hood studies. These factors include the mass flow flux, the angle of the exit flow of the last stage, both the circumferential and the radial ones, and the speed and position of the jet-flow downstream of the seal over the shroud of the bucket. The tests show that each factor has its own distinctive style and extent for influence. Some of them may maximize the performance at some certain point, and some may deteriorate the performance rapidly beyond a threshold. And some factors may change the performance insignificantly within a wide range. However, these influences are not good enough to be consistent with the difference between the sole exhaust hood and the hood with blade simulations. In the second part of this paper, the focus locates on the direction of the jet-flow of the bucket seal. The tests prove that this direction is the prominent factor to influence the exhaust hood performance. Some extra tests for the seal have also been conducted to analyze this factor. The static pressure recovery for the simulation with labyrinth seal is about only half of the sole exhaust hood simulation. The discussion of these tests show that the seal jet is the main cause for this performance dive, and explain how the seal jet direction changes the flow field of the exhaust hood. It also suggests that the procedure to optimize the seal design is not mature yet, for some nature of the jet-flow remains unclear. It may need more detailed study in the future.


Author(s):  
Michal Hoznedl ◽  
Kamil Sedlák ◽  
Lukáš Mrózek ◽  
Tereza Dadáková ◽  
Zdeněk Kubín ◽  
...  

Abstract The paper deals with experimental research of the flow and dynamics of the blades in the last stage of a steam turbine with nominal output of 34 MW and a connected axial exhaust hood. The experiments were carried out on a turbine with relatively low inlet steam parameters “- 64 bars and 445 °C. It was possible to change the operating modes of the turbine during the course of measurement so that significant ventilation would be achieved in the last stage up to the point when aerodynamic throttling occurred in the last stage. In other words the turbine output varied from about 2 to 35 MW. The output of 2 MW was for the case of the island mode turbine operation. The experiments were carried out using static pressure taps and measurements of temperatures at the root and tip limiting wall. In addition to static pressure taps and temperature measurement, it was also possible to carry out probing by pneumatic probe with a diameter of 30 mm. Blade vibration monitoring sensors, so called last stage blade tip-timing, were also installed. The blade tip-timing acquisition hardware was used to monitor rotor blades tip amplitude. Due to the obtained experimental data, it was possible to verify the behaviour of the last stage and the connected exhaust hood for four measured variants. The courses of pressures and steam angles along the length of the LSB were determined. Furthermore, basic parameters of the last stage were determined, i.e. reactions of the stage, Mach and Reynolds numbers and values of pressure recovery coefficients. Based on experimental data the boundary conditions for CFD calculations were determined. Comparison of CFD calculations done for ventilation modes and for a nominal mode was also included. Another phenomenon which occurred during the probing of the flow parameters, particularly in ventilation modes, was the inability to determine parameters of steam due to low values of measured dynamic pressure in the vortex area at the root of the blade. The probe was able to detect dynamic pressure at the level of 50 Pa and more. In other words the transition point between backward and forward flows was identified. This limit point was used for further analysis of ventilation character of the steam flow depending on the ventilation coefficient c2x/u. where c2x is the average axial velocity at the LSB outlet, calculated from volumetric mass flow and u is LSB circumferential velocity calculated at LSB middle diameter. Due to the fact that it was also possible to measure vibration amplitudes of blades using the tip-timing method for a variety of modes, the relationships between pressure ratio over the tip and root of the last moving blade and vibration amplitude were also determined. This verified that the highest amplitude of blade tips occurred just when the compression of the medium on the blade tip was maximum, i.e. c2x/u = 0.05.


2017 ◽  
Author(s):  
Robert Kalista ◽  
Lukáš Mrózek ◽  
Michal Hoznedl

As is well known, the performance of the last stage of the low pressure part of a steam turbine is strongly influenced by the effectivity of the downstream exhaust casing. The efficiency of the exhaust hood depends on many structural factors such as the design of the diffuser parts, dimensions of the outer casing or arrangement of internal supports. The aim of this paper is the experimental study of the influence of the internal supports of the axial-radial exhaust hood on its pressure recovery factor. For one geometry of its diffuser parts a few different variations of internal supports such as T-rib, tube grid or BV were tested. The effect of reducing the width of exhaust hood in the horizontal joint and the changing of axial length of the diffuser were observed. The width of exhaust hood in horizontal joint and the axial length of the diffuser define the area in the horizontal joint of the exhaust hood. How the diffuser behaves when reducing this area is very important in retrofitting of old machines, where there are so many geometric constrains. The effect of wall jet blowing into the diffuser wall was also evaluated. In this paper we concentrate to examine the sensitivity of these certain geometrical parameters of exhaust hood on the pressure recovery of the whole exhaust system of the low pressure part of the steam turbine. The main purpose of our analysis and experimental measuring was optimising the axial-radial exhaust hood of the steam turbine. For this reason, wind tunnel facilities with relevant measuring and traversing systems were designed and built. The measurements have been performed on 1/5th scale test rig which enabled rapid and efficient evaluation of multiple geometrical variants. The observed exhaust hood was designed for an extra long 54inch last stage blade. For measurements of flow parameters was used multi-hole pneumatic pressure probes and wall pressure taps in conjunction with CFD tools to explore physics based alterations to the exhaust configuration.


Author(s):  
Zoe Burton ◽  
Simon Hogg ◽  
Grant L. Ingram

It has been widely recognized for some decades that it is essential to accurately represent the strong coupling between the last stage blades (LSB) and the diffuser inlet, in order to correctly capture the flow through the exhaust hoods of steam turbine low pressure cylinders. This applies to any form of simulation of the flow, i.e., numerical or experimental. The exhaust hood flow structure is highly three-dimensional and appropriate coupling will enable the important influence of this asymmetry to be transferred to the rotor. This, however, presents challenges as the calculation size grows rapidly when the full annulus is calculated. The size of the simulation means researchers are constantly searching for methods to reduce the computational effort without compromising solution accuracy. However, this can result in excessive computational demands in numerical simulations. Unsteady full-annulus CFD calculation will remain infeasible for routine design calculations for the foreseeable future. More computationally efficient methods for coupling the unsteady rotor flow to the hood flow are required that bring computational expense within realizable limits while still maintaining sufficient accuracy for meaningful design calculations. Research activity in this area is focused on developing new methods and techniques to improve accuracy and reduce computational expense. A novel approach for coupling the turbine last stage to the exhaust hood employing the nonlinear harmonic (NLH) method is presented in this paper. The generic, IP free, exhaust hood and last stage blade geometries from Burton et al. (2012. “A Generic Low Pressure Exhaust Diffuser for Steam Turbine Research,”Proceedings of the ASME Turbo Expo, Copenhagen, Denmark, Paper No. GT2012-68485) that are representative of modern designs, are used to demonstrate the effectiveness of the method. This is achieved by comparing results obtained with the NLH to those obtained with a more conventional mixing-plane approach. The results show that the circumferential asymmetry can be successfully transferred in both directions between the exhaust hood flow and that through the LSB, by using the NLH. This paper also suggests that for exhaust hoods of generous axial length, little change in Cp is observed when the circumferential asymmetry is captured. However, the predicted flow structure is significantly different, which will influence the design and placement of the exhaust hood internal “furniture.”


2015 ◽  
Vol 137 (8) ◽  
Author(s):  
Zoe Burton ◽  
Grant Ingram ◽  
Simon Hogg

The exhaust hood of a steam turbine is an important area of turbomachinery research as its performance strongly influences the power output of the last stage blades (LSB). This paper compares results from 3D simulations using a novel application of the nonlinear harmonic (NLH) method with more computationally demanding predictions obtained using frozen rotor techniques. Accurate simulation of exhausts is only achieved when simulations of LSB are coupled to the exhaust hood to capture the strong interaction. One such method is the NLH method. In this paper, the NLH approach is compared against the current standard for capturing the inlet circumferential asymmetry, the frozen rotor approach. The NLH method is shown to predict a similar exhaust hood static pressure recovery and flow asymmetry compared with the frozen rotor approach using less than half the memory requirement of a full annulus calculation. A second option for reducing the computational demand of the full annulus frozen rotor method is explored where a single stator passage is modeled coupled to the full annulus rotor by a mixing plane. Provided the stage is choked, this was shown to produce very similar results to the full annulus frozen rotor approach but with a computational demand similar to that of the NLH method. In terms of industrial practice, the results show that for a typical well designed exhaust hood at nominal load conditions, the pressure recovery predicted by all methods (including those which do not account for circumferential uniformities) is similar. However, this is not the case at off-design conditions where more complex interfacing methods are required to capture circumferential asymmetry.


2021 ◽  
Vol 1096 (1) ◽  
pp. 012097
Author(s):  
A M Kongkong ◽  
H Setiawan ◽  
J Miftahul ◽  
A R Laksana ◽  
I Djunaedi ◽  
...  

Author(s):  
Fabian F. Müller ◽  
Markus Schatz ◽  
Damian M. Vogt ◽  
Jens Aschenbruck

The influence of a cylindrical strut shortly downstream of the bladerow on the vibration behavior of the last stage rotor blades of a single stage LP model steam turbine was investigated in the present study. Steam turbine retrofits often result in an increase of turbine size, aiming for more power and higher efficiency. As the existing LP steam turbine exhaust hoods are generally not modified, the last stage rotor blades frequently move closer to installations within the exhaust hood. To capture the influence of such an installation on the flow field characteristics, extensive flow field measurements using pneumatic probes were conducted at the turbine outlet plane. In addition, time-resolved pressure measurements along the casing contour of the diffuser and on the surface of the cylinder were made, aiming for the identification of pressure fluctuations induced by the flow around the installation. Blade vibration behavior was measured at three different operating conditions by means of a tip timing system. Despite the considerable changes in the flow field and its frequency content, no significant impact on blade vibration amplitudes were observed for the investigated case and considered operating conditions. Nevertheless, time-resolved pressure measurements suggest that notable pressure oscillations induced by the vortex shedding can reach the upstream bladerow.


Author(s):  
Dickson Munyoki ◽  
Markus Schatz ◽  
Damian M. Vogt

The performance of the axial-radial diffuser downstream of the last low-pressure steam turbine stages and the losses occurring subsequently within the exhaust hood directly influences the overall efficiency of a steam power plant. It is estimated that an improvement of the pressure recovery in the diffuser and exhaust hood by 10% translates into 1% of last stage efficiency [11]. While the design of axial-radial diffusers has been the object of quite many studies, the flow phenomena occurring within the exhaust hood have not received much attention in recent years. However, major losses occur due to dissipation within vortices and inability of the hood to properly diffuse the flow. Flow turning from radial to downward flow towards the condenser, especially at the upper part of the hood is essentially the main cause for this. This paper presents a detailed analysis of the losses within the exhaust hood flow for two operating conditions based on numerical results. In order to identify the underlying mechanisms and the locations where dissipation mainly occurs, an approach was followed, whereby the diffuser inflow is divided into different sectors and pressure recovery, dissipation and finally residual kinetic energy of the flow originating from these sectors is calculated at different locations within the hood. Based on this method, the flow from the topmost sectors at the diffuser inlet is found to cause the highest dissipation for both investigated cases. Upon hitting the exhaust hood walls, the flow on the upper part of the diffuser is deflected, forming complex vortices which are stretching into the condenser and interacting with flow originating from other sectors, thereby causing further swirling and generating additional losses. The detailed study of the flow behavior in the exhaust hood and the associated dissipation presents an opportunity for future investigations of efficient geometrical features to be introduced within the hood to improve the flow and hence the overall pressure recovery coefficient.


Sign in / Sign up

Export Citation Format

Share Document