Modelling and Multi-Objective Combustion Optimization of Marine Engine with Speed Maintaining Control Target
Jian Zhang,Yangqian Meng,Dai Liu,Long Liu,Changzhao Jiang,Xiannan Li,Li Huang,Xiuzhen Ma
DOI: https://doi.org/10.1016/j.tsep.2023.101852
IF: 4.56
2023-01-01
Thermal Science and Engineering Progress
Abstract:In this paper, the modelling and optimized control problem for the marine diesel engine combustion charac-teristics has been investigated. For marine diesel engine, it is of significant importance to select proper com-bustion control parameters, by which the target engine speed is maintained stable while economy and emission can be upgraded simultaneously. Firstly, a data-driven model of the marine diesel engine combustion charac-teristics is established by combining the artificial neural network and Takagi-Sugeno fuzzy modelling methods, with which the overall operating conditions are covered. Then, a double closed-loop control scheme is proposed. The outer loop adjusts the engine speed according to time-varying load requirements, whereas the inner loop aims at optimizing the combustion performance and maintaining the target speed stabilized by the outer loop. Furthermore, the NSGA III algorithm with constraints is applied to the data-driven model for multi-objective optimization of combustion performance. The results show that the data-driven model has high generalization and prediction capabilities and that it can forecast diesel engine performance characteristics with an accuracy of up to 99.77% and a maximum error of less than 5%. Meanwhile, the proposed double-loop control loop can optimize the combustion while maintaining navigation stability. Compared with the original machine, the average fuel consumption decreased by 3.57 g/kWh, from 202.31 g/kWh to 198.74 g/kWh. Additionally, the average NOx emission is reduced to 4.82 g/kWh, which is 39.2% lower than the TIER II regulations, and a remarkable reduction of 9.69% is achieved.