Surge Detection in Turbocharger Compressors for Automotive Application and Development of a Control-Oriented Model

The increasingly strict regulations on emissions and fuel consumption of internal combustion engines for automotive applications has pushed researchers and manufacturers to develop new technologies and control strategies. One of the solutions widely adopted for gasoline engines is downsizing coupled with turbocharging. While this improves an engine’s emissions and fuel consumption, turbocharging has its own challenges. The limitation to high performance boosting and boost pressure control lies in the difficulty of predicting and controlling the instability phenomenon that occurs in centrifugal compressors for small mass flow rates. Many different approaches have been followed to study and analyze the behavior of the compressor during this unstable operation and many different models have been developed as a result: ranging from high-fidelity multi-dimensional models to simplified lumped parameter models. The development of control-oriented models is still limited due to the difficulty of accurately representing the complex fluid dynamic phenomena occurring in the compressor while being able to execute on embedded hardware in real time. This work will identify the main causes that lead to a transition from stable to unstable behavior which will provide the knowledge required to develop a control-oriented model. This work describes the design of two experimental investigations performed at the University of Genoa test bench and the development of a one-dimensional distributed-parameter model of the centrifugal compressor and piping system. The aim of the first experimental investigation was to analyze the behavior of the compressor in stable conditions far from and close to the stability limit and to investigate the effect of the compressor outlet circuit on the transition to unstable behavior. The data and the knowledge collected were used to develop a compressor and piping system model that could represent the physical phenomena observed. Following the development of the model, the second experimental investigation was performed to collect data to calibrate and validate the model. The second experimental investigation also offered the opportunity to study the behavior of the compressor during deep surge in more detail. Specifically, the increased number of measurements allowed for study of the shape and position of the cycles occurring during deep surge conditions. The model was then used to perform a preliminary analysis of the parameters and loss mechanisms that cause a transition from stable to unstable behavior.