Affiliation:
1. Naval University of Engineering
Abstract
Mechanism method and neural network identification method are widely used for system modelling. Due to the complexity of actual investigated objects, models built with individual method are usually not good enough for the problems to be solved. Therefore, a hybrid modelling method based on mechanism and identification is presented in the paper. In the hybrid model, the mechanism model is applied as the fundamental model in order to ensure the overall coincidence performance and the identification model is adopted as the compensated model in order to ensure the accuracy performance. The developing process of the hybrid model of a marine turbo-charged diesel is demonstrated in detail in the paper. Testing result shows that the hybrid modelling method is very suitable for modelling complex investigated objects.
Publisher
Trans Tech Publications, Ltd.
Reference4 articles.
1. Paolo F, Stefania S, Peter L, Torque Generation Model for Diesel Engine, C. Proceedings of the 42nd IEEE Conference on Decision and Control, Hawaii, USA, 2003, pp.1771-1776.
2. RakoPoulos C, Giakoumis G, Sensitivity Analysis of Transient Diesel Engine Simulation, J. Proceedings of the Institution of Mechanical Engineers, 2006, pp.89-101.
3. Theotokatos G, On the Cycle Mean Value Modelling of a Large Two-Stroke Marine Diesel Engine, J. Proceedings of the Institution of Mechanical Engineers, 2010, pp.193-205.
4. Galindo J, Luja M, Serrano R, Combustion Simulation of Turbocharger HSDI Diesel Engines During Transient Operation Using Neural Networks, J. Thermal Engineering, 2005, pp.877-898.
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献