RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication
Artículo científico (JournalArticle)
Assessing the optimum combustion under constrained conditions
[EN] This work studies the optimum heat release law of a direct injection diesel engine under constrained conditions. For this purpose, a zero-dimensional predictive model of a diesel engine is coupled to an optimization tool used to shape the heat release law in order to optimize some outputs (maximize gross indicated efficiency and minimize NOx emissions) while keeping several restrictions (mechanical limits such as maximum peak pressure and maximum pressure rise rate). In a first step, this methodology is applied under different heat transfer scenarios without restrictions to evaluate the possible gain obtained through the thermal isolation of the combustion chamber. Results derived from this study show that heat transfer has a negative effect on gross indicated efficiency ranging from -4% of the fuel energy (m(f)H(v)), at high engine speed and load, up to -8% m(f)H(v), at low engine speed and load. In a second step, different mechanical limits are applied resulting in a gross indicated efficiency worsening from -1.4% m(f)H(v) up to -2.8% m(f)H(v) compared to the previous step when nominal constraints are applied. In these conditions, a temperature swing coating that covers the piston top and cylinder head is considered obtaining a maximum gross indicated efficiency improvement of +0.5% m(f)H(v) at low load and engine speed. Finally, NOx emissions are also included in the optimization obtaining the expected tradeoff between gross indicated efficiency and NOx. Under this optimization, cutting down the experimental emissions by 50% supposes a gross indicated efficiency penalty up to -8% m(f)H(v) when compared to the optimum combustion under nominal limits, while maintaining the experimental gross indicated efficiency allows to reduce the experimental emissions 30% at high load and 65% at low load and engine speed., This work was partially funded by GM Global R&D and the Government of Spain through Project TRA2017-89894-R. In addition, the authors acknowledge that some equipment used in this work has been partially supported by FEDER project funds (FEDER-ICTS-2012-06), framed in the operational programme of unique scientific and technical infrastructure of the Ministry of Science and Innovation of Spain. Diego Blanco-Cavero is partially supported through contract FPI-S2-2016-1356 of the Programa de Apoyo para la Investigacion y Desarrollo (PAID) of Universitat Politcenica de Valencia.