Search Results

This research proposes a control system for Spark Ignition (SI) engines with external Exhaust Gas Recirculation (EGR) based on model predictive control and a disturbance observer. The proposed Economic Nonlinear Model Predictive Controller (E-NMPC) tries to minimize fuel consumption for a number of engine cycles into the future given an Indicated Mean Effective Pressure (IMEP) tracking reference and abnormal combustion constraints like knock and combustion variability. A nonlinear optimization problem is formulated and solved in real time using Sequential Quadratic Programming (SQP) to obtain the desired control actuator set-points. An Extended Kalman Filter (EKF) based observer is applied to estimate engine states, combining both air path and cylinder dynamics. The EKF engine state(s) observer is augmented with disturbance estimation to account for modeling errors and/or sensor/actuator offset.

To improve torque management algorithms for drivability, the powertrain controller must be able to compensate for the nonlinear dynamics of the driveline. In particular, the presence of backlash in the transmission and drive shafts excites sharp torque fluctuations during tip-in or tip-out transients, leading to a deterioration of the vehicle drivability and NVH. This paper proposes a model-based estimator that predicts the wheel torque in an automotive drivetrain, accounting for the effects of backlash and drive shaft flexibility. The starting point of this work is a control-oriented model of the transmission and vehicle drivetrain dynamics that predicts the wheel torque during tip-in and tip-out transients at fixed gear. The estimator is based upon a switching structure that combines a Kalman Filter and an open-loop prediction based on the developed model.

Combustion phasing of Spark Ignition (SI) engines is traditionally regulated with map-based spark timing (SPKT) control. The calibration time and effort of this feed forward SPKT control strategy becomes less favorable as the number of engine control actuators increases. This paper proposes a model based combustion phasing control frame work. The feed forward control law is obtained by real time numerical optimization utilizing a high-fidelity combustion model that is based on flame entrainment theory. An optimization routine identifies the SPKT which phases the combustion close to the target without violating combustion constraints of knock and excessive cycle-by-cycle covariance of indicated mean effective pressure (COV of IMEP). Cylinder pressure sensors are utilized to enable feedback control of combustion phasing. An Extended Kalman Filter (EKF) is applied to reject sensor noise and combustion variation from the cylinder pressure signal.

As engines are equipped with an increased number of control actuators to meet fuel economy targets they become more difficult to control and calibrate. The large number of control actuators encourages the investigation of physics-based control strategies to reduce calibration time and complexity. Of particular interest is spark timing control and calibration since it has a significant influence on engine efficiency, emissions, vibration and durability. Spark timing determination to achieve a desired combustion phasing is currently an empirical process that occurs during the calibration phase of engine development. This process utilizes a large number of stored surfaces and corrections to account for the wide range of operating environments and conditions that a given engine will experience. An obstacle to realizing feedforward physics-based combustion phasing control is the requirement for an accurate and fast combustion model.

The highly transient operational nature of passenger car engines makes cylinder pressure based feedback control of combustion phasing difficult. The problem is further complicated by cycle-to-cycle combustion variation. A method for fast and accurate differentiation of normal combustion variations and true changes in combustion phasing is addressed in this research. The proposed method combines the results of a feed forward combustion phasing prediction model and “noisy” measurements from cylinder pressure using an iterative estimation technique. A modified version of an Extended Kalman Filter (EKF) is applied to calculate optimal estimation gain according to the stochastic properties of the combustion phasing measurement at the corresponding engine operating condition. Methods to improve steady state CA50 estimation performance and adaptation to errors are further discussed in this research.

As engines are equipped with an increased number of control actuators to meet fuel economy targets, they become more difficult to control and calibrate. The additional complexity created by a larger number of control actuators motivates the use of physics-based control strategies to reduce calibration time and complexity. Combustion phasing, as one of the most important engine combustion metrics, has a significant influence on engine efficiency, emissions, vibration and durability. To realize physics-based engine combustion phasing control, an accurate prediction model is required. This research introduces physics-based control-oriented laminar flame speed and turbulence intensity models that can be used in a quasi-dimensional turbulent entrainment combustion model. The influence of laminar flame speed and turbulence intensity on predicted mass fraction burned (MFB) profile during combustion is analyzed.

Two on-line algorithms have been developed to acquire driveline component loads in terms of revolutions at torque and rainflow cycle counting matrix. These algorithms have been implemented in real-time on a standard engine controller unit and have been optimized for fast run-time and low memory requirements. The revolutions at torque algorithm is intended to count the number of driveshaft revolutions in each torque level for each gear and store the number of counts in the engine controller memory. The rainflow cycle counting algorithm is intended to count driveshaft torque cycles and to store the number of counts in a two dimensional “from-to” matrix format in the engine controller memory. The revolutions at torque histogram data and the rainflow cycle counting matrix are then downloaded from the vehicle using the data collection device. Download occurs when the vehicle is serviced at a dealership.

Due to the multitude of external design constraints, such as increasing fuel economy standards, and the increasing number of global vehicle programs, developers of automotive transmission controls have to cope with increasing levels of powertrain system complexity. Achieving these requirements while improving system quality, reducing development cost and improving time to market is a very challenging task. To achieve this goal, a rapid prototype controller was used to develop a new transmission clutch temperature model. This model is used to detect clutch surface overheating, improve design and enhance shift quality.

Physical modeling has been used by the industry to improve development time and produce a quality product. In this paper, we will describe two methods used in system control to take advantage of the physical model. One method describes a complete transmission physical model with a full system control utilizing co-simulation techniques. Data will be presented, and comparison to vehicle data will be conducted and verified. The second method will illustrate how to utilize the physical model to improve system design and modification. In this method, vehicle data will be used as inputs to the model, the model output will be verified against vehicle output data. The two methods are excellent tools for the Design For Six Sigma process (DFSS design).

The Chrysler Ultradrive four-speed transaxle 41TE was the first production transmission to pioneer fully adaptive direct clutch-to-clutch electronic controls without overrunning clutches. "Single stage" high flow PWM solenoids have been used for transmission control since 1989 and still being used in the later-developed 545RFE and 62TE transmissions. The proposed target volume-based shift control method allows the usage of flow-based control device such as PWM solenoids to implement torque-based control strategy. Vehicle test results with this new method have shown excellent shift quality and improved system consistency.

This paper presents methods for analyzing and visualizing the relationship between input torque, clutch torque, output torque and input acceleration during the inertia phase of a shift. The methods presented are an expansion of the lever analogy [1]. The methods are useful for understanding how geartrain inertia affects control, both its magnitude and distribution. Clutch energy and shift speeds are also easy to calculate and understand using the tools presented. Lastly the methods show why the optimum control strategies for various transmission configurations (such as DCT's, planetary transmissions, etc.) are different in the inertia phase.

The development of a robust feedback slip controller for a torque converter clutch (TCC) is presented in this paper. The dynamic behavior of the TCC is modeled utilizing the principles of input-output system identification. An H∞ loop shaping controller design technique is applied in order to ensure robust stability against unmodeled system dynamics and large variations in system parameters. Road driving tests indicate that the control system achieves high levels of reliability and stability.

The new Chrysler six-speed transaxle makes use of an underdrive assembly to extend a four-speed automatic transmission to six-speed. It is achieved by introducing double-swap shifts. During double-swap shift, learning the initial clutch torque capacity of the underdrive assembly's subsystem has a direct impact on the shift quality. A new method is proposed to compute and learn the initial clutch torque capacity of the releasing element. In this paper, we will outline a new mathematical method to compute and learn the accurate starting point of the clutch torque capacity for double swap shift control. The performance of the shift is demonstrated and the importance of the adaptation to shift quality is highlighted. An nth order lookup table is presented; this table contains n rows and m columns. Every row defines a relationship between the dependent variable such as actuator duty cycle and one independent variable such as transmission oil temperature, input torque or battery voltage.

A new six-speed transaxle has been introduced by the Chrysler Group of DaimlerChrysler AG. Along with the six forward ratios in the normal upshift sequence, this transaxle features a seventh forward ratio used primarily in a specific downshift sequence. A significant technical challenge in this design was the control of so-called double-swap shifts, the exchange of two shift elements for two other shift elements. In the case at hand, one of the elements is a freewheel. A unique solution is discussed for successful control of double-swap shifts. The new design replaces a four-speed transaxle but makes use of a large percentage of parts and processes from the four-speed design. This approach enabled the new transaxle to reach production in three years from concept. The new transaxle, referred to as the 62TE, has substantially improved performance and passing maneuvers coupled with a new 4.0L high output engine for which the 62TE was developed.

In this paper, the equations of motion for output torque are developed for the 62TE transmission. The 62TE is an innovative six-speed transmission comprised of an existing four-speed transmission, the 41TE with an underdrive assembly added on the transfer shaft centerline. Understanding the behavior influencing the output torque profile is critical to achieving good shift quality[1]. The application of lever analogy [2] resulted in a control strategy that was based on the physics of the system.

The 45RFE is a new generation electronically controlled rear wheel drive transmission. It employs real-time feedback, closed-loop modulation of shift functions to achieve excellence in shift quality and to meet severe durability goals. The 45RFE uses no shift valves; all friction element applications are effected with high-flow electro-hydraulic solenoid valves. A unique gear train arrangement of three planetary carriers allows all sun and annulus gears to have identical numbers of teeth and to use common pinion gears in all carriers. This results in substantial manufacturing simplification. The three-planetary system is designed for four forward ratios of 3.00, 1.67, 1.00 and 0.75 and one reverse gear ratio equal to the low gear ratio. A fifth ratio of 1.50 is used mainly in certain kick-down shift sequences for highway passing. A sixth forward ratio, an additional overdrive ratio of 0.67, is available in the hardware.

The 45RFE is a new generation electronically controlled rear wheel drive automatic transmission. It employs real-time feedback, closed-loop modulation of shift functions to achieve outstanding shift quality and to meet demanding durability goals. It uses no shift valves; all friction element applications are effected with high-flow electro-hydraulic solenoid valves. A unique gear train arrangement of three planetary carriers allows all sun gears and annulus gears to have the same number of teeth respectively and use a common pinion gear in all carriers, resulting in significant manufacturing simplification. The three-planetary system is designed for four forward ratios of 3.00, 1.67, 1.00 and 0.75 and one reverse gear ratio equal to the low gear ratio. A fifth ratio of 1.50 is used only in certain kick-down shift sequences for highway passing. A sixth forward ratio, an additional overdrive ratio of 0.67, is available in the hardware.