Abstract
The present paper deals with analysing a multi-stage ring-section pump concerning head efficiency and optimisation potential and comparison with existing test bench results and systematic optimisation. The investigation was carried out within several stages. An extracted fluid volume with many details is necessary to enable the highest possible accuracy of the simulations.
The impeller side spaces, radii at impeller and diffuser blades, and narrow gaps were not taken into account to obtain a computational grid with the highest possible quality (mostly block-structured grid). The impeller and diffuser were rebuilt and meshed using reverse engineering. This way, the blade surfaces of the impeller and guide vanes could be extracted and used in a blade model tool for turbomachinery to modify the blade accordingly during the optimisation run. Within the transient calculations of the initial situation, vortex structures and dissipation were analysed. The multi-stage centrifugal pump HZ optimisation was mainly carried out in a simplified model to minimise the calculation effort. Impeller, inflow area and volute were optimised manually. Automated optimisation was carried out of the guide vane and refeed channel. Multi-objective optimisation with evolutionary algorithms was applied to the meta-model.
The shape of the inflow suction contour can significantly influence the shape of the head curve. However, a possible reduction in efficiency must also be taken into account. Even a slight alteration in the blading of the guide vanes (variation of the discharge angle by +/- 2°) can significantly change the shape of the head curve as well as the position of the efficiency optimum with an otherwise unchanged geometry.
An efficiency breakdown of the optimised two-stage pump geometry and a comparison with the original geometry are also made. As a result of the optimisation, the efficiency is mainly increased in the entire operating range; in the single-stage variant h by approx. 10%, and in the two-stage variant by approx. 7%! The characteristic stability based on the CFD simulations is improved, and the geometric specifications of the pump connection dimensions (retrofit) are also adhered to.
The impeller side spaces, radii at impeller and diffuser blades, and narrow gaps were not taken into account to obtain a computational grid with the highest possible quality (mostly block-structured grid). The impeller and diffuser were rebuilt and meshed using reverse engineering. This way, the blade surfaces of the impeller and guide vanes could be extracted and used in a blade model tool for turbomachinery to modify the blade accordingly during the optimisation run. Within the transient calculations of the initial situation, vortex structures and dissipation were analysed. The multi-stage centrifugal pump HZ optimisation was mainly carried out in a simplified model to minimise the calculation effort. Impeller, inflow area and volute were optimised manually. Automated optimisation was carried out of the guide vane and refeed channel. Multi-objective optimisation with evolutionary algorithms was applied to the meta-model.
The shape of the inflow suction contour can significantly influence the shape of the head curve. However, a possible reduction in efficiency must also be taken into account. Even a slight alteration in the blading of the guide vanes (variation of the discharge angle by +/- 2°) can significantly change the shape of the head curve as well as the position of the efficiency optimum with an otherwise unchanged geometry.
An efficiency breakdown of the optimised two-stage pump geometry and a comparison with the original geometry are also made. As a result of the optimisation, the efficiency is mainly increased in the entire operating range; in the single-stage variant h by approx. 10%, and in the two-stage variant by approx. 7%! The characteristic stability based on the CFD simulations is improved, and the geometric specifications of the pump connection dimensions (retrofit) are also adhered to.
Originalsprache | englisch |
---|---|
Titel | Proceedings of the Conference on Modelling Fluid Flow (CMFF’22) |
Redakteure/-innen | Janos Vad |
Herausgeber (Verlag) | Budapest University of Technology and Economics |
Seiten | 1-8 |
Seitenumfang | 8 |
ISBN (elektronisch) | 978-963-421-881-4 |
Publikationsstatus | Veröffentlicht - 2 Sept. 2022 |
Veranstaltung | 18th International Conference on Fluid Flow Technologies: CMFF 2022 - Budapest, Ungarn Dauer: 30 Aug. 2022 → 2 Sept. 2022 |
Konferenz
Konferenz | 18th International Conference on Fluid Flow Technologies |
---|---|
Kurztitel | CMFF '22 |
Land/Gebiet | Ungarn |
Ort | Budapest |
Zeitraum | 30/08/22 → 2/09/22 |
ASJC Scopus subject areas
- Allgemeine Energie
- Allgemeiner Maschinenbau
Fields of Expertise
- Sustainable Systems
Treatment code (Nähere Zuordnung)
- Application