To address the challenge of pollutant control during transient operations of heavy-duty diesel engines under the China VI Emission Standard, this study develops a model-based multi-parameter cooperative control algorithm using time iteration. Traditional steady-state MAP-based control methods exhibit significant limitations under dynamic conditions due to fixed parameter mapping and single-variable regulation, resulting in strong subsystem coupling. Focusing on a representative transient process—rapid load increase at constant speed (1200 rpm, IMEP ramping from 3 bar to 20 bar within 1 s)—a dynamic coupling model is established for fuel injection, variable geometry turbine (VGT), and retarded intake valve closing actuator (RIVCA). A novel quadratic control trajectory for fuel quantity (“fast-then-slow" pattern, R \(^2\) = 0.98) and a downward-convex VGT control curve (R \(^2\) = 0.99) are designed. Coupled with real-time intake/exhaust pressure estimation, closed-loop fuel-oxygen equivalence ratio control is achieved. Experimental results demonstrate a reduction in the peak equivalence ratio from 1.02 to 0.78, a 60% decrease in critical duration, a 35% increase in turbine acceleration, and a 40% reduction in intake response time. Under the WHTC, soot emissions are reduced by 64.9% (87% peak reduction), fuel consumption drops by 6.48%, and while NOx increases by 16.6%, the specific emission rises only by 10.9%. The proposed approach overcomes the limitations of traditional steady-state strategies and provides an effective solution for transient emission control in modern diesel engines.