ScienceDirect IFAC-PapersOnLine 48-14 (2015) 235–240 ScienceDirect Available online at www.sciencedirect.com 2405-8963 © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2015.09.463 © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Abstract: In this paper, a robust nonlinear control of an engine cooling system for vehicles is presented, where an electrically driven radiator fan serves as the control input. A simplified control-oriented model of the engine cooling system is derived using the first law of thermody- namics. The control design is based on an integral sliding mode approach and aims at tracking of desired trajectories for the engine outlet temperature. A gain-scheduled modified Utkin sliding mode observer, which uses both a switching term and an output error feedback, is employed to estimate unknown heat flows acting as system disturbances. The estimated heat flows are used in the control structure for a disturbance compensation. An experimental analysis highlights the effectiveness of the integral sliding mode control strategy in combination with the gain-scheduled sliding mode observer. Saif S. Butt * , Robert Prabel * , Harald Aschemann * * Chair of Mechatronics, University of Rostock, D-18059 Rostock, Germany (e-mail: {Saif.Butt, Robert.Prabel, Harald.Aschemann}@uni-rostock.de) Robust Nonlinear Control of an Innovative Engine Cooling System Keywords: Engine control, sliding mode control, robust control, sliding mode observer 1. INTRODUCTION Recent trends concerning the growing environmental im- pact of greenhouse gases, strict regulations towards a cleaner environment as well as the depletion of fossil fuels seek for possible solutions to improve the fuel effi- ciency of vehicles. This results in a paradigm shift in the vehicle design concepts regarding their architecture and propulsion technology accompanied by the introduction of integrated intelligent systems. Incorporating the latest developments and trends, the majority of automotive ve- hicles still uses internal combustion engines with either gasoline or diesel fuel. Moreover, the use of vehicles has in- creased significantly, especially in urban traffic, where the engine is mainly operated in part-load conditions, cf. Arici (1999). These part-load conditions are related to higher fuel consumption and pollutant emissions, which adversely affect the efficiency of the combustion engine, see Heywood (1988). One research direction to maximize the efficiency of the engine is given by the development of sophisticated active engine cooling systems. Such a system can be con- sidered as the integration of electro-mechanical and ther- mal components providing thermodynamic control action and an appropriate cooling heat flow. On the one hand, a properly designed active cooling system contributes to a reduced heating-up time and an accurate temperature tracking control. On the other hand, it improves the re- liability and durability of the engine, see Page (2005). A conventional engine cooling system as depicted in Fig. 1 consists of an internal combustion engine, a radiator, a thermostat and a mechanically operated bypass valve, cf. Zou (1999). Recent research on energy savings and an electrification of the automotive cooling system have been published in several contributions, cf. Stetlur (2005); Salah (2008, 2010) and Butt (2014a). In Salah (2008), detailed Radiator Engine Thermostat bypass Fan Fig. 1. Structure of an engine cooling system. dynamic models of all the components of an engine cooling system are described. Therein, a nonlinear control strategy based on backstepping control for the engine outlet tem- perature is presented. In Butt (2014b), a dedicated control design aiming at a short heating-up phase of the engine has been described. In Wang (2015), it was pointed out that a considerable amount of energy is needed to operate the radiator fan(s) in comparison to the electric bypass valve and the electrically actuated pump. Aschemann (2011) in- vestigated a flatness-based control scheme to track desired trajectories of the engine outlet temperature based on a simplified model of a cooling system. Therein, the angular velocity of the radiator fan is employed as the control input. Regarding the influence of parameter uncertainties, robust control schemes are required. Therefore, an integral sliding mode control (integral SMC or I-SMC) is proposed in this contribution to track desired trajectories for the engine outlet temperature. Moreover, a gain-scheduled version of the Utkin sliding mode observer (SMO) is cho- sen to estimate unknown heat flows. The convergence of