Contents lists available at ScienceDirect Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman Developing a multi-zone model for a HCCI engine to obtain optimal conditions using genetic algorithm Mohsen Nazoktabar a , Seyed Ali Jazayeri b , Kamran Arshtabar c , Davood Domiri Ganji d, ⁎ a Islamic Azad University, Roudehen Branch, Tehran, Iran b K.N. Toosi University of Technology, Tehran, Iran c Mazandaran Engineering Organization, Babol, Iran d Noshirvani University of Technology, Babol, Iran ARTICLE INFO Keywords: HCCI engine Multi-zone model Optimal condition Grey-box modeling ABSTRACT The potential benefits of Homogenous Charge Compression Ignition (HCCI) engines are higher thermal effi- ciency, lower Nitrogen Oxides (NOx) and Particulate Matter (PM) emissions in contrast with conventional en- gines. This mode of combustion faces its own set of challenges, such as lack of a direct means to control timing initiation of combustion together with higher unburned hydrocarbon and Carbon Monoxide emissions. In the present study, a multi-zone thermo-kinetic model developed to investigate the HCCI engine performance and to predict its emissions. In addition, to determine the optimal HCCI combustion phasing, a grey-box control model used to predict major variables. In order to optimize conditions for higher thermal efficiency and lower emis- sions, the thermo-kinetic model coupled with a Genetic Algorithm. The octane number and equivalence ratio are two qualitative parameters, which determine the optimal conditions. The grey-box model consists of a combi- nation of physical models and Artificial Neural Networks (ANN) models. The steady state and transient vali- dations show that the grey-box model is capable of predicting HCCI engine outputs including combustion phasing, load, THC, CO emissions. By means of coupling the GA and thermos-kinetic model, the optimal engine operating conditions can be derived without the use of test results. The validation results show that the grey-box model is able to predict CA50, IMEP, CO and THC with the average errors of 1.2 CAD, 0.4 bar, 10 PPM, 0.8%, and 394 PPM, respectively. 1. Introduction There has been a global trend to enforce more stringent emission regulations since the early years of the 21st century. Homogenized lean mixture and low temperature combustion mode leads to higher thermal efficiency lower NOx and negligible particulate matter in HCCI engines. On the other hand, difficulty in controlling the combustion phasing, a restricted range of engine operation, high hydrocarbon and CO emis- sions are among the major challenges in these engines [1,2]. In recent years, more efforts has been carried out to get better understanding of HCCI engine performance not only by experimental investigation but also by means of thermo-kinetic models. In order to simulate the HCCI combustion and study its performance, a single-zone or multi-zone HCCI engine combustion chamber coupled to a chemical kinetic me- chanism is used. The single-zone thermo-kinetic model is capable of predicting the combustion initiation timing in a HCCI engine [3]. Single-zone models downfalls are over prediction of peak cylinder pressure, over prediction of NOx, under prediction of burn duration, and inability to predict hydrocarbon and CO [4]. In this study, the in-cylinder interior is di- vided into 10 zones. A detailed fluid mechanics code and a detailed chemical kinetics codes used to predict HCCI combustion parameters. The result indicates accurate prediction of maximum pressure, burn duration and emissions .A seven zones model, using a reduced model to investigate combustion parameters such as auto-ignition and in-cy- linder pressure carried out earlier [5]. A modified method to determine the initial condition using a multi-zone model also presented [6]. Komninos proposed a modified multi-zone model to predict HCCI en- gine emissions fueled with iso-octane [7]. In this model, heat transfer to the wall and heat and mass transfer between different zones considered. Nobakht et al. [8] investigated the effect of engine variables on HCCI combustion and its performance fueled with natural gas by a six-zone model. They showed that equivalence ratio and intake pressure are the main parameters that affect HCCI combustion and performance. Neshat and Saray [9] proposed an 11-zone model together with a new heat transfer model and a semi detailed chemical kinetic https://doi.org/10.1016/j.enconman.2017.12.001 Received 20 August 2017; Received in revised form 21 November 2017; Accepted 1 December 2017 ⁎ Corresponding author. E-mail address: ddg_davood@yahoo.com (D.D. Ganji). Energy Conversion and Management 157 (2018) 49–58 0196-8904/ © 2017 Elsevier Ltd. All rights reserved. T