Khalil, Waël (2006) Développement d'un appareil automatisé de mesure simultanée d'équilibres de phases et de propriétés volumétriques. Exploitation des données volumétriques pour le calcul prédictif de grandeurs thermodynamiques dérivées. PhD thesis Génie des Procédés, ENSMP - CEP Centre Energétique et Procédés, ENSMP.

Full text available as: - These_Wael_KHALIL_19-12-2006pdf.pdf ( 2012 Kb )

Licence: Copyright

Abstract

In process engineering science, volumetric properties and phase equilibria are used together to design the various implied unit operations. This work mainly aims at automating an equipment based on the vibrating tube densimetry, equipment being used for simultaneous measurements of phase equilibria and volumetric properties. This densimeter works in a very broad range of pressures (from vacuum to 70MPa) anda t tempertures between 253.15 and 423.15 K. The purpose of automation is to ensure better security of workers, to allow a higher degree of accuracy of the obtained results and to generate numerous data without particular human care. Automatisation concerns first controlled introduction of th e fluid from loading reservoir via two alternative-turn valves, then the monitoring of a special pressurizing device connected to a motor reducer to compress and decompress very finely the liquid phase. The speed of compression and the relaxation is controllable by means of the intervalle of time between valves actuations and choice of the motor reducer rotation speed. The rate of pressure variations is adjustable with values lower than 0.001 MPa/s. This is completely invaluable in the case of fluid close to critical conditions where the heating effects are marked, ac it is proven inside the manuscript. Automatic valves where also added to protect the pressure transducers. A program was written to control the whole setup and to acquire and treat the values of the measuref deta. Equipment was used successfully for measurements on the system : carbon dioxide + isopropanol. A complete study of the volumetric properties of the mixture CO2 + isopropanol was carried out. The mixture was studied in temperatures ranging from 308 to 348 K and with pressures up to 15 MPa. Modeling of the experimental results is presented and comparaisons with literature data are performed. Neural networks are used to represent the volumetric properties of the mixture, as well as carbon dioxide derived properties (enthalpy, entropy and heat capacity). The results display the very good capacity of a neural network to represent volumetric properties and derived properties via learning and validation staps. In parallel, study of the detection of hydrate formation has been carried out in order to estimate the ability of vibrating tube densimetry to detect appearance of solids in gaseous mediums.

Statistiques de consultation

Repository Staff Only: edit this item