Application of modelling approaches of twin-screw compressors: thermodynamic investigation and reduced-order model identification
Refrigeration is an essential part of the food chain. It is used in all stages of the chain, from industrial food processing to final consumption at home. In these processes, mechanical refrigeration technologies are employed, where compressors increase gas pressure from evaporation to condensation....
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN FR |
| Publicado: |
EDP Sciences
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/076d336819694250824888ea53c7fb21 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Refrigeration is an essential part of the food chain. It is used in all stages of the chain, from industrial food processing to final consumption at home. In these processes, mechanical refrigeration technologies are employed, where compressors increase gas pressure from evaporation to condensation. In industrial refrigeration systems, twin-screw compressors represent the most widely used technology. A detailed mathematical model of a twin-screw compressor has been developed in Simulink® using differential equations for energy and mass balances to simulate the compression cycle that includes suction, compression and discharge phases. Gas pressure and enthalpy can be calculated as time functions during the cycle. However, the computational times obtained limit the possibility to extend the use of the model in the development of control strategies for the whole refrigeration plant in its real operating conditions. Therefore, the detailed model has been used to train a simplified model developed in Matlab®: the simulated mass flow rate, shaft power and the fluid discharge temperature have been employed to identify several geometrical and thermodynamic parameters of the simplified model. The latter relies on non-linear algebraic equations and, thus, requires a very short computational time. A limited performance dataset has been used to train the model, and a different dataset to test it: the results of the models have been compared, and small errors in mass flow rate, shaft power and fluid discharge temperature have been observed. |
|---|