Development of combustion indicators for control of multi-fuel engines based on new combustion concepts
- Autores: Irina Ayelén Jiménez
- Directores de la Tesis: Benjamín Pla Moreno (dir. tes.), Carlos Guardiola García (dir. tes.)
- Lectura: En la Universitat Politècnica de València ( España ) en 2022
- Idioma: español
- Tribunal Calificador de la Tesis: José Galindo Lucas (presid.), Octavio Armas Vergel (secret.), Bryan Patricio Maldonado Puente (voc.)
- Programa de doctorado: Programa de Doctorado en Sistemas Propulsivos en Medios de Transporte por la Universitat Politècnica de València
- Materias:
- Enlaces
- Tesis en acceso abierto en: RiuNet
- Resumen
The need to satisfy emissions and CO2 regulations is pushing the automotive industry to develop different innovative technologies. These technologies include alternative fuels and new modes of combustion, among others.
Therefore, the need for the development of new methods for combustion control in these mentioned conditions arises. For this reason, in this work different models and indicators have been developed aimed at the diagnosis and control of combustion in both normal and abnormal conditions.
For normal combustion cases, a combustion model has been developed, the objective of this model is to estimate the mean of evolution of the mass fraction burned and the in-cylinder pressure. An observer had been implemented, based on knock sensor signal, in order to improve the estimation in transient conditions and also to be able to make use of the model with different fuels. A cyclic variability model is also presented, where from the combustion model, a probability distribution is propagated over two of the parameters of such model. Both models had been applied for a spark ignition engine and a turbulent jet ignition combustion engine.
For the abnormal combustion cases, an analysis of the resonance within the combustion chamber had been included, where two models capable of estimating the evolution of the resonance were also developed.
These models, for both normal and abnormal conditions, were used for the diagnosis of combustion: on the one hand, for knock recognition, where three knock detection strategies were developed: two based on the in-cylinder pressure sensor and one on the knock sensor. On the other hand, an application of a resonance model was carried out in order to improve the estimation of the trapped mass from the resonance excitation.
Finally, to show the potential of such models and applications, two control strategies were developed: one for the control of knock through the actuation of the spark advance, and a second for the control of residual gases, through the actuation of the variable valve timing, while optimizing the combustion through the actuation of the spark advance.
