摘要：AbstractThe NOχStorage Catalyst is currently envisaged to be implemented in light-duty passenger cars for nitrogen oxides reduction, in order to comply with strict emission legislation targets. Since robustness and durability of the engine and emission control system is the first priority in automotive application, to satisfy the need of robust on-board real time monitoring, diagnosis and control, computing efficient methods are needed. In this framework, a control oriented model that describes the dynamics of the main physical-chemical processes within the NSC catalyst, while still maintaining affordable computational burden, has been developed and validated. Model calibration has been performed, for light-duty application, along the NEDC test cycle, by using a statistical-based sub-optimal procedure, based on a parametric analysis which allows identifying the more suitable section of NeDC cycle for model identification, without the need to perform cost- and time- expensive experiments on the engine test bench. The procedure also accounts for missing information and sensors inaccuracies. The great potential of this methodology is the possibility to adopt not optimal designed tests for model parameters identification. The proposed methodology is proven to be effective for real time control strategy, directly embedded in ECU, and provide a sub-optimal but effective strategy for complex models calibration.
关键词：KeywordsAutomotive system identificationmodelingNSCLNTNOχemissionsexhaust gas after-treatment systems
