Inproceedings,

NUMERICAL MODELLING OF THE TRANSITION FROM A CLOSED WALL FILM TO DISCRETE LIQUID RIVULETS

, and .
(2019)

Abstract

Fogging systems in industrial stationary gas turbines are more and more used as they offer a cheap, easy and very adjustable method to increase a turbine's power output. With the use of high-fogging, water droplets enter the compressor cascade, where they evaporate and continuously cool the air throughout the compression. Depending on the spray characteristics and droplet interactions various droplet sizes can occur. Due to inertia, only small droplets with diameter dμm are able to follow the surrounding air flow. Larger droplets are likely to hit the compressor blades or casing, where they interact with the component. A distinct part of the impacting droplet deposits on the surface and a wall film develops. Experiments have shown, that the wall film on the suction side of a compressor blade splits up into liquid rivulets. Since now, only the wall film has been investigated numerically. Within the current study an adaption to the numerical wall film formulation is developed to enable the transition from a closed wall film to discrete liquid rivulets. The adaption also allows to simulate the propagation of the ligaments due to surrounding conditions, may it be an air flow, or potential forces. The core of the adaption is a new treatment of the surface tension, as it is distinguishes between cells with and without a contact line. This results in one main advantage, that liquid streaks can emerge from a closed film, but also coalescence of different streaks is possible. Besides single liquid rivulets under varying conditions of the surrounding flow, mainly the applicability to liquid flows over a tilted plate has been investigated. Here cases with and without external air flow have been simulated. Finally, the application case of a compressor blade of a staggered grid under high-fogging conditions has been simulated. The transition point can be evaluated and compared with experimental data.

Tags

Users

  • @adrianseck

Comments and Reviews