Characteristics of Diesel Fuel Injection Rate under Multi-Injection Condition and the Control Stategy based on ANN

Paper #:
  • 2017-01-2212

  • 2017-10-08
High-pressure common rail(HPCR)fuel injection system is the most widely used fuel system for diesel engines due to the fact that it can provide constant injection pressure and precise injection strategy. However, when multiple injection strategy is used, the pressure wave caused by the opening and closing of the needle valve will affect the subsequent injection and can not be neglected. In this paper, the influence of pressure wave on the second injection pressure, injection rate and fuel-injection quantity is carried out on a common rail fuel injection test rig under two-stage injection conditions. The results show that the pressure wave varies in terms of various rail pressure, environmental back pressure and injection intervals, resulting in a 10% fluctuation. As a consequence, the injection quantity will be changed. In detail, increasing injection intervals leads to an enhancement of injection pressure, injection pressure fluctuations and the decrease of injection quantity. The reason is the higher pressure fluctuation and the reduction of injection closing delay. With narrow injection pulse width, the injection intervals have a great influence on the change of maximum injection rate. In addition, different injection intervals affect combustion and so that the NOx and soot formation. In order to control the injection intervals and injection pulse width precisely, BP neural network model of artificial intelligence (AI) is adopted. Three trained neural network are validated by experiments and the error is under 5%. It concludes that with ANN an optimized injection control parameters (injection pressure, injection rate and injection quantity) can be achieved.
SAE MOBILUS Subscriber? You may already have access.
Attention: This item is not yet published. Pre-Order to be notified, via email, when it becomes available.
Members save up to 40% off list price.
HTML for Linking to Page
Page URL

Related Items

Training / Education
Training / Education