【摘要】 为解决柴油机预测模型参数过多导致的标定困难问题，采用遗传基因算法对搭建的船用中速机DIjet预测燃烧模型进行智能标定。借助最大缸内压力以及对应的曲轴转角快速完成缸压曲线标定，相较于传统手动标定，标定时间可缩短到几十分之一。在进行柴油机整体功率验证后，对NOx排放进行标定，形成一套简便可行的智能标定流程。在完成标定的模型上进行NOx排放及有效燃油消耗率（brake specific fuel consumption,BSFC）优化，在考虑两者不同权重系数的情况下得到更优的帕累托前沿面。更多还原

【Abstract】 In order to solve the calibration difficulties caused by excessive parameters in the diesel engine prediction model,a genetic algorithm was used to intelligently calibrate the combustion prediction model DIjet for a marine medium speed engine. By utilizing the maximum cylinder pressure and corresponding crankshaft angle, the cylinder pressure curve calibration was quickly completed, which reduced the calibration time to the several tenth of its original time compared to traditional manual calibration. After verifying the overall power of the diesel engine,the NOx emissions was calibrated to form a simple and feasible intelligent calibration process. The NOx and brake specific fuel consumption(BSFC) was optimized on the calibrated model, and a better Pareto front surface considering the different weight coefficients of the two was obtained.更多还原