The Large Hadron Collider (LHC) is the next large particle accelerator developed at CERN, constructed to enable studies of particles. The acceleration of the particles is carried out using magnets operating at about 1.9 K, a temperature achieved by regulating flow of superfluid helium. For economical reasons, control of the helium flow is based on feedback of virtual flow meter (VFT) estimates instead of real instrumentation.
The main purpose of this work is to develop a virtual flow meter with the possibility to estimate the flow by means of two different flow estimation methods; the Samson method that has previously been tested for the LHC, and the Sereg-Schlumberger method that has never before been implemented in this environment.
The virtual flow meters are implemented on PLCs using temperature and pressure measurements as input data, and a tool for generating the virtual flow meters and connect them to the appropriate physical instrumentation has also been developed.
The flow through a valve depends, among others, on some pressure and temperature dependent physical properties that are to be estimated with high accuracy. In this project, this is done by bi linear interpolation in two dimensional tables containing physical data, an approach that turned out to be more accurate than the previously used method with polynomial interpolation.
The flow measurement methods have been compared. Since they both derive from empirical studies rather than physical relations it is quite futile to find theoretical correspondencies, but the simulations of the mass flows can be compared. For low pressures, the results are fairly equal but they differ more for higher pressures. The methods have not been validated against true flow rates since there were no real measurements available before the end of this project.
Source: Linköping University
Author: Ödlund, Erika