Intelligent optimization of diesel engine Selective catalytic reduction urea injection based on multi-model state estimation to reduce NH3 slip and NOx emission

To address the challenge of enhancing NOx conversion efficiency and reducing NH3 slip in diesel vehicle Selective Catalytic Reduction (SCR) system, a multi-model optimization-based intelligent optimization method for SCR urea injection was proposed. An SCR state estimator employing multi-model fusion accurately predicts NH3 coverage, SCR downstream NH3 concentration, and NOx concentration. The intelligent optimization strategy of urea injection used Non-dominated Sorting Genetic Algorithm (NSGA-II) to optimize MAP in intelligent calibration mode. In the real-time optimization mode, the optimal control parameters and singular perturbation adaptive correction method were used to optimize the urea injection amount in real-time. Validation through Worldwide Harmonized Transient Cycle (WHTC) testing demonstrates significant results. The results show that the state estimation closely aligns with SCR catalyst simulation, with an average error of less than 5 %. Urea injection strategy under calibration mode, particularly at lower exhaust gas temperature, NOx conversion rate increases from 45 % to 75 %·NH3 slip decreases at higher exhaust temperature, especially with increased exhaust flow rates and SCR carrier temperature, consistently staying below 10 ppm. Under real-time optimization mode, a significant reduction in NOx emission from 8.9 g/kW·h to 0.46 g/kW·h, with NOx conversion efficiency exceeding 95 %, representing a 13 % improvement·NH3 slip is optimized, decreasing from an average of 12.89 ppm to 7.99 ppm, with an average NH3 slip consistently below 10 ppm. This method intelligently adjusts urea injection quantity under actual operating conditions, achieving maximum NOx conversion rates while meeting regulatory requirement. This research offers new theoretical and methodological support for optimizing SCR system performance in diesel vehicles.