Application of machine learning tools for constrained multi-objective optimization of an HCCI engine

Ayat Gharehghani, Hamid Reza Abbasi and Pouria Alizadeh

Energy, 2021, vol. 233, issue C

Abstract: Because of strict emission regulations, significant efforts are made by researchers to reduce the pollutants from internal combustion engines. Homogeneous mixture of fuel and air in Homogeneous Charge Compression Ignition (HCCI) engine results in lower amounts of thermal NOx while CO and HC emissions would be higher in this process. Besides, performance instability due to knock and misfire is another problem that HCCI engines are facing. In this paper, machine learning tools have been implemented to learn the behavior of a single-cylinder engine using three different fuels with the HCCI strategy. Changes in six objective functions with engine Inlet Valve Closing (IVC) temperature and air-fuel equivalence ratio are studied with the help of an Artificial Neural Network (ANN). Those objective functions are thermal efficiency, combustion efficiency, IMEP, together with NOx, CO, and HC emissions. A multi-objective optimization algorithm based on a decomposition method and genetic algorithm is used to identify the best operating condition of studied fuels. Results show that HCCI engine fueled with Ethanol and Methanol could produce significantly lower HC and CO emissions at their optimum point. However, the produced NOx would be higher in these cases comparing to the engine fueled with CNG. The genetic algorithm indicates that the optimum point of operation for CNG-fueled engine is TIVC ≈ 445 K while λ ≈ 2.25. In contrast, for the Ethanol-fueled case, this optimum point is located in the IVC temperature of ≈450 K and λ ≈ 2.75. The TOPSIS point is located in minimum air-fuel equivalence ratio and lower IVC temperatures (TIVC ≈ 420 K) when the engine is being operated with Methanol as fuel.

Keywords: HCCI engine; Stable operating range; Artificial neural network (ANN); Multi-objective optimization; Decomposition method; Genetic algorithm (GA) (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (6)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544221013542
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:233:y:2021:i:c:s0360544221013542

DOI: 10.1016/j.energy.2021.121106

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().