Control of over-actuated systems with application to advanced turbocharged diesel engines.

摘要

The automotive industry is currently striving to improve vehicle fuel economy, while complying to stringent emission standards and keeping affordable costs. Engine downsizing with advanced turbochargers is one of the main technical solutions nowadays implemented by the industry in response to the above challenges. In particular, the adoption of advanced actuation techniques brings more degrees of freedom to improve the efficiency of breathing and combustion in internal combustion engines. However, it is understood that improvements in engine design can be effective only if matched by the ability to closely control engine breathing and combustion performance.;This dissertation aims at exploiting systematic methodologies for control and optimization of over-actuated systems. Following the path of the state of the art in control theory and practice, a novel control design methodology is proposed that hinges on the characterization of redundancy for a class of over-actuated systems in geometric terms to determine the reconfigurable structure of the controller. The method relies upon the concept of inverse model allocation, where the integration of allocation module is effectively separated from stabilization of the regulated outputs, allowing one to shape the transient response by optimizing on-line a given cost function without affecting the output tracking performance. The proposed approach overcomes fundamental limitations found in current techniques and specifically addresses the need for simultaneous control and optimization of the air-path systems for advanced downsized engines.;The proposed control methodology is applied to the case study of a Diesel engine air-path system equipped with Variable Geometry Turbine (VGT), Variable Geometry Compressor (VGC) and Exhaust Gas Recirculation (EGR) systems. A comparative study against different conventional control methods shows the effectiveness of the proposed methodology to improve the engine performance variables, as well as the overall reduction in design and calibration costs.