Industrial steam turbines are designed for application in power-, process- and chemical engineering. Particular modules ensure the optimum integration into power plants and other engineering processes. Extraction modules allow the controlled extraction of large steam quantities on certain and constant enthalpy levels. Valves regulate the amount of steam extracted from the turbine expansion path. Depending on the valve lift, different flow separation phenomena can occur peripherally inside the valves, causing undesired large unsteady fluid forces on the valve head and seat. Due to the compact design of the industrial steam turbines, these unsteady jets can influence the rotor dynamics as well as the blade loading of the adjacent stages. These fluctuations should be understood and avoided in order to enhance the reliability of steam turbines. In the present study the unsteady flow phenomena due to separation occurring circumferentially inside the valve of extraction modules are investigated numerically. First, the commercial 3D RANS CFD-solver (ANSYS CFX 14) is validated in the application to experimental results. Subsequently, the various flow patterns of the examined valve design are analyzed on a standalone numerical valve model in an extensive study. In order to assess the impact of these unsteady flow separations on other components, the complete extraction module is simulated in combination with the adjacent stages. The transient simulation results show pressure fluctuations downstream of the valves resulting in an unsteady load of the control valves, the shaft and the blading.

This paper investigates the validity of the current industrial procedure of measuring optimized blade profiles in a wind tunnel under air condition although they are applied in a steam turbine. Therefore, it is important to analyze the possibility of using air-measured profile data for optimizing steam turbine blades. To this end, experimental data is collected using the cylindrical datum blade of a steam turbine in a three-stage high pressure steam turbine and in an annular air cascade wind tunnel. Three-dimensional CFD simulations are separately performed for both setups and show a good agreement with the experimental data. The numerical simulations can therefore be assumed to represent the real flow conditions. Firstly, for analyzing aerodynamic transferability, two optimized profiles are measured in the annular air cascade wind tunnel at Reynolds number of 6 × 105. These profile sections are designed for high and intermediate pressure applications by employing an optimizer. The optimization is performed with the focus on reducing the profile loss for steam conditions. The experimental data verifies that the losses of the optimized profiles are reduced significantly compared to the datum blade profile measured in the same air rig. Secondly, the air-measured optimized blade profiles are used to design a 3D-optimized blade. In a numerical investigation, this optimized blade is analyzed in the steam turbine by applying steam conditions. The outlet Reynolds number of the 2nd stage is 8 × 105. This configuration is compared with the numerical results of the datum blade profile simulations. The relative isentropic total-to-total efficiency is increased by 0.6% due to the use of the optimized rotor blades. The benefit persists also for a maximum outlet Reynolds number of 9 × 106.

Industrial steam turbines are applied for power generation as well as drive for turbo-compressors. They combine a high level of operational flexibility with highest reliability. Especially in the field of process technology they provide process heat on a certain enthalpy level for other industrial applications. Modular design concepts are used to meet these various requirements like admission or extraction of large steam quantities. Extraction modules use valves to control the amount of steam extracted from the turbine expansion path at constant steam parameters. While extraction steam is taken from the turbine through an outlet flange, the remaining steam passes valves and downstream diffusers, flows into an annular inner casing and finally escapes through the subsequent stages. Depending on the valve lift, different flow separations can occur around the valves, resulting in unsteady transonic jets. Due to the compact and asymmetric design of the inner casing the flow into the subsequent stages is strongly disturbed. Hence, strong unsteady mechanical blade loading can occur in addition to efficiency loss. The current work focusses on the improvement of the flow conditions in the subsequent stage. Experimental results are applied to quantify the viability of the used 3D RANS CFD-solver (ANSYS CFX 14) for these numerical investigations. Compared with the experiment, the distribution of pressure, velocity and incidence angle are well predicted by the numerical code. It is evident that the unsteady transonic jets emerging around the valves have a major influence on the distribution of the parameters considered. Thus, to quantify the impact of a modified inlet chamber design, it is sufficient to simulate the domain starting from the valves. The influence of different design modifications on the flow parameters in comparison with the base design is discussed in detail in an extensive study. The results clarify that horizontal and vertical valve positions, as well as thorough contouring of the radial-axial deflection have a strong influence on the distribution of pressure, mass flux and incidence angle. Hence, in this contribution combinations of the most beneficial modifications are investigated numerically and compared with the base design.

Operation of industrial steam turbines under partial admission is usually implemented in order to cope with fluctuations in demand while maintaining a high efficiency. Subject of the article presented is the experimental investigation of the performance and efficiency of a control stage operated with air, under partial admission. Furthermore, the steady state blade loading distribution on the stator in a free passage and at the entry into a not admitted passage is observed. For the experiment, the single-staged axial-flow turbine is operated with varying shaft speed, pressure ratio and admission rate, realized by flow blockaging. Two rotationally symmetric stator endwall designs are tested: a plane and a contoured endwall. Reducing the admission rate at part load decreases the efficiency, especially at high circumferential velocities as precedent investigations showed. Pumping and mixing loss behaviour shows a different scale across the performance map. To prove CFD Simulations, results of the blade loading distribution on the stator are acquired.

No abstract