As the use of turbocharged engines has expanded in recent years, turbocharging systems,
such as two-stage turbochargers, have become more complex, and as a result, the issue of
controlling such systems has become more challenging. This paper presents a high-speed optimal
control method developed to address this issue. This model predictive control contains an internal
mathematical model of the engine and turbocharger, allowing it to predict system behavior. By
adopting this model predictive control and with some further algorithm improvements, this control
system can now be executed in real-time. Simulation results show that, compared with conventional
systems, an automotive supercharged engine with the proposed method could attain boost pressure
response improvement of approximately 50%. In addition, the basic operation of this control
method was verified using an in-house engine. With our progress in this development, Mitsubishi
Heavy Industries, Ltd. (MHI) is aiming to actively participate in the development of
next-generation products such as two-stage electric turbocharging systems in the future.
