This work presents a theoretical and numerical study on leakage in sliding vane compressors and expanders, in collaboration with Ing. Enea Mattei. The aim of the numerical study is to develop a new thermodynamic model that can include opensystem loss due to leakages, in order to expand the the pre-existing computational design tool SVEC designed by Ing. Enea Mattei on Matlab®. The new model is able to predict lubricant and gas leakages via the minimum radial clearance in a wide set of operating conditions, taking into account the difference in clearance type and geometry, rotational speed and gas/lubricant mass and volume ratio elaborated. User can choose between six different leakage models, comprehensive of gaseous, incompressible and semi-empirical models and two two-phase bulk properties models, used to take into account the global leakage dynamics, which are not yet clear. The model has been implemented and validated using experimental results of MATTEI compressors and expanders. Poiselle-Couette leakage model, combined with Maxwell bulk property model was found to best fit the experimental data for compressors, with an average relative error of 3.93%, while in expander the best fitting was achieved with Poiselle-Couette leakage model combined with Dukler bulk property model, resulting in an average relative error of 7.63%. These results were obtained assuming that the leakage via the minimum radial clearance is 41% of the total leakage. Furthermore, the dynamic and pneumatic models implemented in SVEC have been analysed, after having observed unrealistic results during the simulations of the tensile safety coefficient to traction. The analysis highlighted the relation between pressure field acting on the concealed side of the vane and dynamic results. These results have been used to modify the aforementioned pressure and dynamic models, restoring the expected results and reducing the average error on the vane internal stress analysis from 27.3% to 9.1%. Assessment on the computational velocity on new SVEC version shows an average reduction of 51% respect to the previous version, as result of the optimization activity carried out on code structure.
Questo lavoro presenta uno studio teorico e numerico sui trafilamenti in compressori ed espansori a palette, in collaborazione con Ing. Enea Mattei. Lo scopo dello studio numerico è stato quello di sviluppare un nuovo modello termodinamico che possa includere le perdite a causa dei trafilamenti, al fine di espandere lo SVEC (Sliding Vane Efficiency Compressor), il tool numerico di progettazione preesistente e sviluppato da Ing. Enea Mattei con Matlab®. Il nuovo modello è in grado di prevedere i trafilamenti di lubrificante e gas attraverso il gioco radiale minimo in un’ampia serie di condizioni operative, tenendo conto del tipo di gioco e della sua geometria, della velocità di rotazione e del rapporto gas/lubrificante sia che massico che volumetrico elaborato. L’utente può scegliere tra sei diversi modelli di trafilamento, comprensivi di modelli gassosi, incomprimibili e semi-empirici e due modelli per il calcolo delle proprietà di miscele bifase, utilizzati per tenere conto delle dinamiche globali di trafilamento, ancora non tutto chiare. Il modello è stato implementato e convalidato utilizzando risultati sperimentali di compressori ed espansori MATTEI. Il modello di trafilamento di Poiselle-Couette, combinato con il modello per proprietà bifase di Maxwell, è risultato essere il migliore per i compressori, con un errore relativo medio di 3.93%, mentre negli espansori il miglior fitting si è ottenuto combinando il modello trafilante di Poiselle-Couette e quello per le proprietà bifase di Dukler, con conseguente errore relativo medio di 7,63%. Questi risultati sono stati ottenuti supponendo che la perdita attraverso il gioco radiale minimo sia pari al 41 % della perdita totale relativa ai trafilamenti. Inoltre, sono stati analizzati i modelli dinamici e pneumatici implementati in SVEC, dopo aver osservato risultati non realistici durante le simulazioni per il calcolo del coefficiente di sicurezza a trazione. L’analisi ha evidenziato la relazione tra il campo di pressione che agisce sul lato nascosto della paletta e i risultati dinamici. Questi risultati sono stati usati per modificare il suddetto campo di pressione e il modello dinamico, ripristinando i risultati attesi e riducendo l’errore medio sull’analisi della sollecitazione interna delle palette da 27.3% a 9.1%. La valutazione del tempo computazionale della nuova versione SVEC mostra una riduzione media del 51% rispetto alla versione precedente, frutto del lavoro di ottimizzazione svolto a livello di struttura del codice.
Developing and validating models of leakages through the minimum radial clearance via a simulation tool for sliding-vane compressors and expanders
VALLONE, ANTONIO
2017/2018
Abstract
This work presents a theoretical and numerical study on leakage in sliding vane compressors and expanders, in collaboration with Ing. Enea Mattei. The aim of the numerical study is to develop a new thermodynamic model that can include opensystem loss due to leakages, in order to expand the the pre-existing computational design tool SVEC designed by Ing. Enea Mattei on Matlab®. The new model is able to predict lubricant and gas leakages via the minimum radial clearance in a wide set of operating conditions, taking into account the difference in clearance type and geometry, rotational speed and gas/lubricant mass and volume ratio elaborated. User can choose between six different leakage models, comprehensive of gaseous, incompressible and semi-empirical models and two two-phase bulk properties models, used to take into account the global leakage dynamics, which are not yet clear. The model has been implemented and validated using experimental results of MATTEI compressors and expanders. Poiselle-Couette leakage model, combined with Maxwell bulk property model was found to best fit the experimental data for compressors, with an average relative error of 3.93%, while in expander the best fitting was achieved with Poiselle-Couette leakage model combined with Dukler bulk property model, resulting in an average relative error of 7.63%. These results were obtained assuming that the leakage via the minimum radial clearance is 41% of the total leakage. Furthermore, the dynamic and pneumatic models implemented in SVEC have been analysed, after having observed unrealistic results during the simulations of the tensile safety coefficient to traction. The analysis highlighted the relation between pressure field acting on the concealed side of the vane and dynamic results. These results have been used to modify the aforementioned pressure and dynamic models, restoring the expected results and reducing the average error on the vane internal stress analysis from 27.3% to 9.1%. Assessment on the computational velocity on new SVEC version shows an average reduction of 51% respect to the previous version, as result of the optimization activity carried out on code structure.| File | Dimensione | Formato | |
|---|---|---|---|
|
Developing and validating models of leakages through the minimum radial clearance via a simulation tool for sliding-vane compressors and expanders.pdf
non accessibile
Descrizione: Master degree thesis - Vallone Antonio
Dimensione
6.3 MB
Formato
Adobe PDF
|
6.3 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/144185