ABSTRACT In a parallel hybrid electric vehicle, higher fuel economy gains are typically achieved if significant electric drive (or engine-off) operation is possible, shifting the engine operating schedule so that it only runs at medium to high load for best efficiency. To enable efficient engine-off driving, a typical configuration will have a disconnect clutch between the engine and the rest of the driveline. In some configurations, when engine-on operation is requested the disconnect clutch is applied in conjunction with the traction motor/generator to crank the engine (i.e., a flying engine start). In this paper we describe the development of a control system for a flying engine start using an engine disconnect clutch. The clutch is located between the engine and electric motor, which is connected to the input of a multispeed transmission. We first describe an initial control algorithm evaluation using a driveline model. After this initial evaluation, the control algorithm was incorporated into an existing hybrid vehicle control system architecture and tested on a HIL bench to validate code implementation. Finally, we describe experiments in which we tested the control algorithm in a test cell using a mule transmission. We discuss in qualitative terms the refinements in the control algorithm resulting from the experimental results. INTRODUCTION Hybrid electric vehicles (HEVs) are widely seen as an important technology in addressing future regulatory requirements on fuel economy and greenhouse gas emissions [1]. HEVs typically have one or two motor/generators that carry out a variety of functions depending on the hybrid powertrain architecture. In an input-split (also known as series-parallel) architecture, one motor may propel the vehicle while the other can be used to start the engine througha planetary gearset, as in General Motors' 2-Mode hybrid architectures [2]. In a single motor parallel HEV capable of significant engine-off electric driving ( Fig. 1), the motor/ generator must simultaneously propel the vehicle and, via a disconnect clutch #1, spin up the engine to firing speed during an engine start event. This presents a challenge for vehicle drivability, as the disturbance of the engine start can be transmitted through the driveline or engine mounts during the engagement of the disconnect clutch and some motor torque must be diverted to overcome the engine cranking torque. In this paper, we describe the controls development for an experimental engine-start setup using a mule transmission. HARDWARE MULE ARCHITECTURE Figure 1 shows schematically the mule configuration used for this work. This configuration has a single motor/generator and a wet engine disconnect clutch (EDC) located between the engine and the transmission. When the EDC is disengaged, the engine is disconnected from the transmission and can be shut off in order to save fuel when engine power is not required or when the engine would operate at a relatively inefficient load point. The motor/generator, with the power supplied from the battery, can provide traction torque to the drive wheels through the transmission. When the engine is needed to provide additional power or charge the battery, the EDC is engaged so that the motor can start the engine. Controls Development for Clutch-Assisted Engine Starts in a Parallel Hybrid Electric Vehicle2011-01-0870 Published 04/12/2011 Anthony Smith, Norman Bucknor, Hong Yang and Yongsheng He General Motors Company Copyright © 2011 SAE International doi:10.4271/2011-01-0870Downloaded from SAE International by Birmingham City Univ, Monday, August 20, 2018Figure 1. Single Motor Parallel HEV Schematic ENGINE-START CONTROL ARCHITECTURE At the highest conceptual level, the engine-start control architecture must manage two functions: 1) control the EDC to spin up the engine; and 2) manipulate the motor torque to cancel disturbances due to the torque load from the EDC. A schematic of the control architecture is shown