Abstract
The real time phenomenon inside an IC engine air-path is highly stochastic and non-linear in nature. In-order to understand & model the system phenomena, mathematical analogue of the phenomenon needs to be constructed and these are ODEs with higher order. Due to resource constraints in EMS and solving the MIMO equation is practically not possible in real time.With enhanced core resource optimization, the system engineer must decide a tradeoff between accuracy and computation load. In this paper the author proposes to use numerical methods to solve the ODEs. Author derives a semi-implicit Euler solver using implicit Euler method combined with Newton’s method to higher dimensional ODEs for engine air-path system to reduce the order. Author has also discussed two advantages of this method: 1) Model order reduction by decoupling coupled system at each time step, 2) Guaranteed convergence with sufficiently smaller time steps in each iteration. This enhanced reduced model with small perturbation to accommodate modeling uncertainties and non-linear behavior of flow dynamics is then checked for accuracy and convergence mathematically using Eigenvalues of Jacobian matrix. The reduced model is time independent and asymptotically stable.