Simulation-based optimal calibration of spark ignition engines with multiple objectives: fuel economy and combustion variability

作     者：Lee, T-K Kramer, D. Ohl, G. Filipi, Z. S. 

作者机构：Univ Michigan Dept Mech Engn Ann Arbor MI 48109 USA USA RDECOM TARDEC Warren MI USA Adapt Mat Inc Ann Arbor MI USA 

出 版 物：《PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART D-JOURNAL OF AUTOMOBILE ENGINEERING》 (机械工程师学会会报；D辑：机动车辆工程杂志)

年 卷 期：2011年第225卷第D10期

页      面：1427-1442页

核心收录：

学科分类：08[工学] 0802[工学-机械工程] 0823[工学-交通运输工程] 

基　　金：Chrysler LLC 

主　　题：internal combustion engine optimal calibration combustion variability fuel economy multiple-objective optimization artificial neural network virtual sensing 

摘      要：Optimal engine calibration methodology at part-load operating conditions using a multi-scale simulation approach is proposed and demonstrated on a spark ignition engine with dual-independent variable valve timing and a charge motion control valve. Fuel economy is typically selected as a calibration objective at part-load operating conditions. However, to secure vehicle driveability and smooth engine operation with low vibration and harshness, the combustion variability is considered by introducing the coefficient of variation in the indicated mean effective pressure characterized through a statistical analysis of experimental data. Other engine responses are fully predicted through a co-simulation approach by implementing the quasi-dimensional combustion simulation into the one-dimensional gas exchange simulation, and subsequently captured by an artificial neural network for fast computation. The best actuator set points are determined by solving the constrained multiple-objective optimization problem with fuel economy and combustion variability objectives under part-load conditions. The calibration eliminates a high combustion variability at low loads and low speeds, maintaining most of the fuel economy benefit achievable with variable valve timing, and it can be used for an implementable feedforward control strategy.