This study utilizes Design of Experiments methodology to optimize the performance of a multi-cylinder aviation diesel engine. Through a rigorous approach, design variables and response parameters were selected, and their ranges defined. Based on the specified variable ranges, a software tool was used to generate a test matrix of the minimum-size required to define the relationships between variables. Experimental work was conducted to acquire real engine data for all points in the test matrix, and used to build a Stochastic Process Model of engine performance. The empirical model was used to predict the optimal values for the design variables that would produce the most power from the engine. These conditions were then experimentally validated. Through this approach it was found that engine power at sea level could be increased by as much as 12% compared to the reference calibration, while still maintaining safe operating conditions in terms of exhaust gas temperature and in-cylinder peak pressure. This approach using Design of Experiments was found to dramatically reduce the amount of experimental work required to quantify the full capability of the engine.
展开▼