Zongxuan Sun

Challenges and opportunities in automotive transmission control

Automotive transmission is a key element in the powertrain that connects the power source to the wheels of a vehicle. To improve fuel economy, reduce emission and enhance driving performance, many new technologies have been introduced in the transmission area in recent years. This paper first reviews different types of automotive transmissions and explains their unique control characteristics. We then address the challenges facing automotive transmission control from three aspects: calibration, shift scheduling, and sensing, actuation and electronics. Along the way, research opportunities to further improve system performance are discussed.

Pressure-Based Clutch Control for Automotive Transmissions Using a Sliding-Mode Controller

Clutch shift control is critical for efficient and high-performance transmission designs, including automatic, dual clutch, and hybrid transmissions. To ensure a smooth clutch to clutch shift, appropriate controls for two consecutive processes are critical. One is the precise coordination between the on-coming and off-going clutches, which requires the on-coming clutch to be filled and ready for engagement at the predetermined time (clutch fill). The other is the proper torque control during clutch engagement. In this paper, we will investigate the closed-loop “wet” clutch control enabled by a pressure sensor in the clutch chamber. The main challenges of the pressure-based “wet” clutch control lie in the complex nonlinear dynamics due to the interactions between the fluid and the mechanical systems, the ON/OFF behavior of the clutch assembly, the time-varying clutch loading condition, the required short time duration for a precise and robust clutch shift, and the lack of the displacement information. To enable precise and robust pressure-based control, this paper focuses on the following three aspects. First, a clutch dynamic model is constructed and validated, which precisely captures the system dynamics in a wide pressure range. Second, a sliding-mode controller is designed to achieve robust pressure control while avoiding the chattering effect. Finally, an observer is constructed to estimate the clutch piston motion, which is not only a necessary term in the nonlinear controller design but also a diagnosis tool for the clutch fill process. To validate the proposed methods, a transmission clutch fixture has been designed and built in the laboratory. The experimental results demonstrate the effectiveness and robustness of the proposed controller and observer.

Transient Control of Electro-Hydraulic Fully Flexible Engine Valve Actuation System

Fully flexible valve actuation (FFVA) system, often referred to as camless valvetrain, employs electronically controlled actuators in place of the camshaft to drive the intake and/or exhaust valves for internal combustion engines. This system enables the engine controller to tailor the valve event according to the engine operating condition in real-time to improve fuel economy, emissions, and torque output performance. This paper presents the transient control of a laboratory electro-hydraulic fully flexible valve actuation system. Transient control of the FFVA system includes lift transient, duration transient, phase transient, speed transient, and mode transient. With constant engine speed, the valve profile is periodic in time domain and the lift, phase, and duration transients can be realized using robust repetitive control. When the engine speed varies, the period of the valve profile changes in real-time. This phenomenon poses a fundamental challenge to the transient control problem and repetitive control cannot be applied anymore. To overcome this challenge, we propose a new valve profile consisting of a periodic portion and a dwell portion with time-varying duration. Robust repetitive control is then applied to the periodic portion and proportional plus integral and derivative (PID) control is applied to the dwell portion. These two controls are switched in real-time to achieve asymptotic valve profile tracking performance. To demonstrate the effectiveness of the proposed control method, we show real-time valve-lift profiles used to explore homogeneous charge compression ignition (HCCI) combustion at different engine operating conditions.

Dynamics and control of an electro-hydraulic fully flexible valve actuation system

As the internal combustion engine moves into the 21/sup st/ century, fully flexible valve actuation (FFVA) systems are being proposed as an enabling technology for advanced internal combustion engine concepts. This paper focuses on exploring the dynamics and control of an electro-hydraulic fully flexible valve actuation system. Challenges and approaches for developing variable valve actuation systems are first outlined and discussed. Dynamic model of the electro-hydraulic actuator was obtained based on experimental data. A robust repetitive controller is then designed and implemented using a dSpace system. We are able to achieve valve profile tracking errors within 140 microns for a 9.5 mm lift at engine speed up to 3000 rpm. To demonstrate the flexibility and robustness of the system, we show real-time valve-lift profiles used to explore advanced combustion technologies.

Trajectory tracking and disturbance rejection for linear time-varying systems: Input/output representation

This paper considers the problem of asymptotic trajectory tracking and/or disturbance rejection for linear time-varying systems via input/output representation. Inspired by the structure of repetitive control, a novel feedback mechanism is proposed based on which the necessary and sufficient condition for the problem is derived. This condition is a generalization to the time-varying case of the result in time-invariant setting. Based on the complete knowledge of the plant model, a design is provided and it is applicable to repetitive control for time-varying plants. An important feature of the proposed approach is that a systematic way of constructing time-varying internal model unit is provided.

Modelling, control, and hardware-in-the-loop simulation of an automated manual transmission

Abstract To enable a realistic automated manual transmission (AMT) system performance evaluation and rapid prototyping, this paper focuses on constructing a practical Simulink model and a hardware-in-the-loop (HIL) real-time simulation. The working principle and dynamic characteristics of the AMT system are first explored. The driveline, the engine control, and the dry clutch are modelled and analysed. In particular, a dynamic model for the hydraulic clutch actuator that is currently used in the actual transmissions is developed. To evaluate the performance and operation characteristics of the AMT system, the control methodology to ensure desirable performance is studied. The gearshift control logic and the proportional—integral—derivative-based clutch control are analysed and implemented. Based on the dynamic model and the controller, a Simulink model is developed and implemented into an HIL real-time simulator to enable rapid prototyping. In addition, to realize an energy-efficient and smooth clutch engagement, the possibility of using the dynamic programming method to generate the optimal clutch and engine torque control inputs is investigated as well. In particular, the controllability of the AMT system is studied to determine the number of control inputs necessary for optimal control. To this end, the simulation and experiment results in the HIL environment are presented.

Active Motion Control of a Hydraulic Free Piston Engine

The free piston engine (FPE), as an alternative of the conventional internal combustion engine, could significantly impact the energy consumption and emissions of both on-highway and off-highway vehicles. In an FPE, the piston motion is dependent on the combustion chamber gas dynamics and the loading dynamics in real time. One of the technical barriers that prevents the wide spread of this technology is the lack of precise piston motion control. In this paper, we present the modeling and control of a hydraulic FPE with an opposed-piston opposed-cylinder design. Specifically, a comprehensive system model is first built and validated to study the dynamics of the engine operation. The simulation studies lead to the design of an active piston motion controller, which acts as a virtual crankshaft by utilizing energy in the storage element to regulate the piston to follow a predefined trajectory. With a designated optimal trajectory, the virtual crankshaft ensures stable and robust operation in engine motoring. Preliminary combustion testing results and analysis are also included in this paper.

Adaptive control with asymptotic tracking performance and its application to an electro-hydraulic servo system

This paper presents a discrete-time adaptive controller, which incorporates internal model principle for asymptotic tracking performance of systems with parametric uncertainties, unmodeled dynamics and disturbances. Global stability and tracking performance of the adaptive system are derived under conditions on the systems stabilizability and bounds of noise and unmodeled dynamics. It is shown that asymptotic tracking can be achieved while the unmodeled dynamics and disturbances exist. The adaptive algorithm is applied to an electrohydraulic servo system for periodic trajectory tracking and disturbance rejection. Experimental results based on an eighth-order adaptive system updated at 2560 Hz demonstrate the adaptive systems ability in high bandwidth tracking performance under effects of system variations and finite word length real-time computation.

A novel internal model-based tracking control for a class of linear time-varying systems

This paper provides a novel method of constructing an internal model-based design of reference tracking and/or disturbance rejection for a class of linear time-varying plants with a known linear time invariant (LTI) exosystem. It is shown how the realization of an appropriate time-varying internal model can be constructed by means of a novel feedback mechanism. The design of the internal model consists of two ingredients: (1) a time-varying system immersion of the exosystem, and (2) an automatic generation of the desired control input to render the error-zeroing subspace invariant, based on the complete knowledge of the plant model. The important features of the proposed method lie in that the tracking problem setup and the proposed feedback mechanism allow us to avoid explicitly calculating the desired input, which keeps the regulated error identically at zero. Moreover the time-varying immersion is guaranteed to hold for the class of plant models under consideration. These features significantly broaden the range of applications of the proposed method, and simplify the control implementation process.

Design and Implementation of Clutch Control for Automotive Transmissions Using Terminal-Sliding-Mode Control and Uncertainty Observer

This paper investigates the clutch control problem for automotive transmission systems. This paper focuses on the pressure tracking control for a wet clutch, which has a much shorter stroke and asks for a more precise position and pressure regulation. The system dynamics are highly nonlinear, and the developed model includes unmodeled dynamics and parameter uncertainties. All these factors will influence the closed-loop system performance if not considered carefully during the control design, which makes the control synthesis for clutch systems challenging. Compared with the existing sliding-mode control (SMC) method, an improved method based on nonlinear sliding mode is presented in this paper to further improve the convergence rate and tracking precision of the closed-loop clutch system. This controller ensures that the pressure tracking error not only reaches the sliding manifold in finite time but converges to the equilibrium point in finite time as well. Moreover, considering the chattering phenomenon caused by high switching gains, a composite terminal SMC (TSMC) method based on uncertainty observer is proposed to reduce chattering. Using an extended state observer (ESO) technique, an uncertainty observer is designed for estimating the model uncertainties of pressure dynamics, pressure-reducing-valve flow dynamics, and estimation error of piston displacement. Then, a feedforward compensation term is combined with the terminal-sliding-mode feedback control. Thus, the composite terminal-sliding-mode controller may take smaller control gains for the whole pressure tracking procedure without sacrificing uncertainty compensation performance. Experimental results based on an xPC Target platform of the clutch system are provided to show the superiority of the proposed methods.