Automated Calibration of Heavy-Duty Low NOx Aftertreatment System Controls using Physics-Informed Machine Learning and Global Optimization Methods

Heavy duty diesel engines provide a robust power plant for transportation applications both on highway and off road. Control of criteria pollutants such as particulate matter and NOx at tailpipe for these applications based on standards set by regulatory bodies such CARB and EPA is critical. SwRI has demonstrated the capability to achieve 0.02 g/bhp-hr. Tailpipe NOx standard through application of a model-based controls in EPA and CARB funded projects. This control mechanism enables precise urea dosing for both steady and transient conditions by leveraging the estimated ammonia storage state in a dual dosing system using chemical kinetics-based SCR observer models. This controller is highly nonlinear, with a significant amount of controller tuning and up to 55 calibratable parameters. To improve the accuracy and reduce the time required for calibration of this controller, this work proposes the deployment of a PINN-based SCR plant model in conjunction with a Genetic Algorithm-based optimization script in a closed loop with the low NOx controller that can enable identification of near-optimal controller calibration in a simulation environment. This work describes the optimization framework and its validation with real-world experimental data. The calibration determined in this framework achieved 0.02 g/bhp-hr. for regulatory cycles consisting of CFTP, HFTP, RMC, and LLC. A quantitative and qualitative analysis of this proposed framework's results is performed and compared against the existing expert-driven manual calibration process.