4 Challenge X Control Strategy Development

As part of the Challenge X competition, the University of Tennessee has chosen a hybrid electric powertrain architecture for conversion of a 2005 Chevrolet Equinox SUV to a more fuel efficient and environmentally responsible vehicle. A brief overview of the powertrain configuration is presented, as well as the corresponding supervisory control system architecture. The methodology for developing the control strategies and algorithms are described. OVERALL VEHICLE CONTROL STRATEGY GOALS UTK VEHICLE POWERTRAIN AND CONTROLS ARCHITECTURE PRIMER The University of Tennessee has applied the systems engineering approach to determine the appropriate advanced powertrain architecture for their entry into the Challenge X competition. Based on their analysis, the team has elected to design a charge sustaining, through-the-road, parallel hybrid electric vehicle with small displacement diesel engine and low storage requirement (LSR) high voltage system. A PSAT block diagram representation of the selected powertrain architecture is shown in Figure 1. Perhaps one of the greatest benefits of this powertrain configuration versus others is the high voltage system (traction motor and battery pack) is completely isolated from the heat engine. This provides a redundant system with the ability for reduced-power operation under possible subsystem failures. The basic components of this powertrain that must be controlled by the supervisory vehicle control system are the Cobasys Nickel Metal Hydride high voltage battery, EV Ranger traction motor/transaxle, and Fiat 1.3L diesel engine. The basic vehicle system control architecture is illustrated in Figure 2. Note that the foundation for the supervisory controller is represented as the Vehicle System Control Module (VSCM). This architecture houses four (4) fundamental control processes necessary for the operation and control of the hybrid electric drivetrain. These processes include the Vehicle Mode Control Process (VMCP), the Battery Mode Control Process (BMCP), the Regenerative Braking Control Process (RBCP), and the Energy Management Control Process (EMCP). The functionality of each of these processes is described in greater detail in subsequent sections of this document. Figure 1 PSAT representation of a through-the-road parallel configuration Figure 2 Basic block diagram of UTK vehicle system controller VEHICLE SYSTEM CONTROL OBJECTIVES The overall function of the University of Tennessee Challenge X control system is to coordinate the interaction of the heat engine, the traction motor, and the energy storage system. The manner in which the control system carries out this function relies on several factors. Translate driver intent The most fundamental objective of the vehicle system control is to translate the intent of the driver. The control system must interpret what the driver is trying to do, and to deliver what is expected up to the limitations of the entire system. The primary interfaces for the driver to the vehicle are the accelerator and brake pedals. These inputs are transformed into control signals for the traction motor and heat engine. These two primary motive forces work together to provide the necessary torque to satisfy the demands of the driver. Maintain state-of-charge (SOC) of HV battery Since the design philosophy of this control system employs a charge-sustaining approach, the hybrid control system must maintain the state of charge of the high voltage battery pack. This function must be integral to the control algorithm and, more importantly, be transparent to the driver. Fluctuations to the delivered torque to the drive wheels are not desired from a consumer acceptability and drivability point-of-view. There are two (2) basic methods, or sources of energy in the powertrain, to charge the battery pack. The first method, dubbed regenerative braking, is to make use of the otherwise wasted kinetic energy from a braking event. Regenerative braking can lead to a more efficient drivetrain. Regenerative braking can be applied in two (2) basic versions. The most efficient means of regenerative braking is referred to as series regenerative braking. In this approach, the traction motor absorbs all of the energy from the wheels to slow the vehicle up to a charge limitation on the battery. At this point, the foundation brakes are then applied. While this is the best system to use, it is inherently more difficult to implement. The second approach to regenerative braking is referred to as parallel regenerative braking. The basic difference with this version versus the series approach is the traction motor and foundation brakes work in parallel to slow the vehicle down. For this reason, it is less efficient since less energy is returned to the high voltage battery pack. Parallel regenerative braking is much easier to implement. However, due the overall system efficiency gains that can be attained and the powertrain configuration chosen by the team, the series regenerative braking approach will be incorporated in the UTK controls design. The second method for charging the battery pack is to use the traction motor as a generator that utilizes energy from the heat engine. Due to the architecture selection of the UTK team, this can only be accomplished when the vehicle is moving. Idle charging of the high voltage battery pack is not possible in this configuration. Protect high voltage (HV) battery One of the key items for HEV durability is the life span of the high voltage battery pack. The vehicle control system should provide a means of limiting available battery power based on the limitations of the pack itself. In order for the battery to survive for a predetermined warranty period, strict adherence to battery pack manufacturer limitations should be maintained. Such items as charge and discharge limitations, maximum module temp, and state of charge limitations must be taken into account when coordinating the interactions of the traction motor and the heat engine. CONTROL STRATEGY DESIGN CONTENT The high level control strategy objectives have been established. The control system features developed to deliver these objectives are presented in the following sections. BASIC DESIGN STRATEGY FEATURES AND ALGORITHMS Each control process of the UTK VSCM plays a major role in defining and executing the control strategy for the overall vehicle. Engine ON/OFF conditions are determined by the VMCP (Vehicle Mode Control Process). The operation of the engine for this vehicle is very basic in the sense that it is allowed to shut off only during idle periods. This comes as a result of the powertrain architecture being developed as a through the road parallel. Idle charging of the high voltage battery pack is simply not possible in this configuration. The vehicle must be moving in order to charge the battery. The engine can be shut down during periods of “prolonged” braking events at low speeds where the vehicle is assumed to be coming to rest. The Battery Mode Control Process (BMCP) has the responsibility for reporting the appropriate and corrected HV battery power limits to the Energy Management Control Process for further manipulation. The strategy makes use of a calibrateable table to determine the additional power required of the engine to maintain the state of charge (SOC) of the HV battery pack. This power, termed PSOC, is a one-dimensional function of battery SOC. An example of this is illustrated in Figure 3. Here, a negative value for PSOC indicates that the battery needs to be charged towards the target SOC. Conversely, a positive value means that the HV battery should be discharged to utilize stored energy in the pack. It should be noted that the power necessary to maintain the SOC can be calculated in a variety of methods. The method presented in Figure 3 is merely a baseline. An optimization involving the efficiencies of the battery pack and traction motor should be employed in such as way as to determine the optimal value for PSOC for a given set of pertinent conditions. The BMCP is also responsible for modifying the battery power limits that are calculated inside the BCM. The BMCP corrects these limitations for such conditions as high SOC, low SOC, high module temperature, and low module temperature. Figure 4 represents a typical discharge power curve for the Cobasys NiMH battery pack that will be used in the UTK Challenge X Equinox. This is the type of data that will be output from the internal battery control module (BCM). -25 -20 -15 -10 -5 0 5 10 15 20 25 0 10 20 30 40 50 60 70 80 90 100 State of Charge, SOC (%) P S O C (k W ) CHARGE REGION DISCHARGE REGION Figure 3 Determination of battery SOC maintenance power for BMCP Figure 4 Physical discharge power limitations for Cobasys NiMHax 288 battery The Regenerative Braking Control Process (RBCP) is responsible for coordinating the necessary traction motor braking torque values during braking events. The RBCP must monitor such vehicle parameters as wheel speed, SOC, and brake pedal demands to determine the appropriate amount of negative torque to request from the motor. This brake torque command becomes an input to the EMCP for further processing. In addition, the RBCP must constantly monitor the existing anti-lock braking system (ABS) on the vehicle. The RBCP must cancel any traction motor brake torque request during an ABS event so that the positive effects of ABS are not cancelled and no wheel lock-up occurs. The Energy Management Control Process (EMCP) is the most critical process within the VSCM. The EMCP is responsible for coordinating the interaction of the heat engine and the traction motor. The EMCP must ensure that the driver demanded power is satisfied while at the same time maintaining the state-of-charge of the HV battery pack. The EMCP must deliver these items while also administering overall system limitations for subsystem component protection. The EMCP joins the outputs from the BMCP, the VMCP, and the RBCP to determine what is required of the heat engine and the traction motor. The root output of the VMCP is the driver demanded power, dubbed Pdrv. The prime output of the BMCP is the power necessary to maintain the SOC of the HV battery pack, referred to as PSOC. These variables together form the total power required of the engine in HEV mode. It is worth noting that PSOC is ignored in a ZEV mode. The total engine power desired, designated Ptot, thus becomes