Investigation on fuel injection quantity of low-speed diesel engine fuel system based on response surface prediction model. (15th November 2020)

• Record Type: Journal Article
• Authors: Lan, Qi; Bai, Yun; Fan, Liyun; Gu, Yuanqi; Wen, Liming; Yang, Li
• Is Part Of: Energy. Volume 211(2020)
• Publication Date: 2020-11-15
• Subjects: Low-speed diesel engine -- Fuel system -- Fuel injection quantity -- Response surface prediction model -- Significance analysis; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2020.118946
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: To analyze the fuel injection quantity of low-speed diesel engine fuel system, an AMESim model was established. Its accuracy was validated by fuel injection quantity and system pressures. Based on D-optimal design of experiment and partial least squares regression, the response surface prediction model of fuel injection quantity was obtained. The R 2 and R a d j 2 of the prediction model are 0.995 and 0.984 respectively and the standardized residuals of the predicted values are distributed between −2 and 2. Meanwhile, the maximum relative error between the calculated and predicted fuel injection quantity is only 3.48%. The predictive capacity of the model is proved. Then the significance analysis upon the structural parameters was performed. The results show that the diameters of fuel plunger and hydraulic piston and the interactions between the diameter of fuel plunger and the diameter of high-pressure fuel tube, between the diameter of hydraulic piston and the pre-tightening force of needle spring are the key factors influencing the fuel injection quantity for their p -values are less than 0.05. Since the p -values of the diameters of high-pressure fuel tube and nozzle hole and the interaction between the two are less than 0.001, they have significant effects on fuel injection quantity. Highlights: The fuel injection quantity of a double-valve controlled fuel system was analyzed. An accurate response surface prediction model of injection quantity was obtained. Significance analysis was performed upon the prediction model. The effects of key factors on fuel injection quantity were analyzed in detail.