Abstract
This paper describes a novel approach to model-based optimisation of dual-fuel LPG/diesel engine control maps. Optimal offline optimisation seeks to achieve low fuel consumption, low emissions, and high driveability at the same time. Experiments were performed with a single-cylinder, four-stroke, direct injection diesel engine with a rated power of 3.5 kW in dual fuel mode for a range of loads from 0 to 12 kg. First, using experimental data, engine performance and emissions were predicted in function of the engine load and the LPG/diesel fuel ratio by two architectures of neural networks: The multilayer perceptron and the radial basis function neural networks. Based on the experimental testing data, the root mean square error was found to be 2.75 × 10−5 for engine torque, 0.0024 for fuel consumption, 0.7 for CO2, and 0.8 for NOx. Then, a multi-objective optimisation environment was developed, which computes the basic control maps for the dual fuel engine settings based on the modelled emission behaviour of the engine. Results showed that engine torque increased by 0.1% and fuel consumption decreased by 20%. These results proved that the proposed method has the capability to develop optimised engine control maps for dual-fuel engines.
Acknowledgements
All authors contributed to the study’s conception and design. Material preparation, data collection, modelling, validation, and analysis were performed by TCHATO YOTCHOU Giovani Vidal. The first draft of the manuscript was written by TCHATO YOTCHOU Giovani Vidal and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Disclosure statement
No conflict of interest was reported by the author(s).