A Deep Learning-Based Strategy to Initiate Diesel Particle Filter Regeneration

Deep learning (DL)-based approaches enable unprecedented control paradigms for propulsion systems, utilizing recent advances in high-performance computing infrastructure connected to modern vehicles. These approaches can be employed to optimize diesel aftertreatment control systems targeting the reduction of emissions. The optimization of the Trapped Soot Load (TSL) reduction in the Diesel Particulate Filter (DPF) is such an example. As part of the diesel aftertreatment system, the DPF stores the soot particles resulting from the combustion process in the engine. Periodically, the stored soot is oxidized during a DPF regeneration event. The efficiency of such a regeneration influences the fuel economy, and potentially the service interval of the vehicle. The quality of a regeneration depends on the operating conditions of the DPF, the engine, and the ability to complete the regeneration event. The favorable occurrence of these conditions is determined by a high number of variables including the speed profile, the state of the road, and the influence of traffic conditions. Control algorithms aim to find the drive cycle intervals with optimal conditions for executing a regeneration. It is a challenging task to optimize regenerations using rule-based control approaches. Such algorithms have a limited capability to handle a wide variety of drive cycles. This article proposes a DL-based control strategy that aims to reduce oil dilution while increasing fuel efficiency by minimizing the number of regenerations and maximizing the oxidized soot load. Based on the analysis of the driving conditions, the proposed strategy targets the most conducive regeneration opportunities. The proposed strategy is evaluated using historical drive cycle data of 10 vehicles, covering a year of vehicle operation. The effectiveness of the DL-based control approach compared to a rule-based control strategy is discussed. The results show that the DL-based control approach leads to fewer interrupted regenerations and more soot oxidation per regeneration while reducing oil dilution and increasing fuel efficiency.