Mingxin Kang

Continuation/GMRES Method based Nonlinear Model Predictive Control for IC Engines

Abstract This paper presents a model-based receding horizon optimal control algorithm for the engine speed tracking control. A mean-value model including the air intake dynamics and the rotational dynamics is exploited in the tracking controller design, which is calibrated based on the physical rules combined with curve fitting techniques. Based on this mean-value model, the dynamical model of the speed tracking error is derived for any given speed command. The design problem is reduced to the receding horizon control problem under the constraint of the tracking error dynamics. The online computational algorithm based on C/GMRES approach is adopted to solve this nonlinear receding optimal problem. Finally, simulation and experiments results demonstrate the satisfactory tracking performance by using the proposed controller.

Nonlinear model predictive torque control for IC engines

This paper presents a model predictive online optimization scheme for the engine torque control problem. The control-oriented model is based on the intake air charging dynamics and torque generation model which are derived from the mean value model. In order to reduce the tracking error induce by the insufficient accurate predictive model, an embedded integrator about the tracking error is designed. Then, the online optimization algorithm namely C/GMRES is adopted to solve the nonlinear optimal problem. Finally, experimental validations are conducted and demonstrated the robustness and transient performance for the proposed engine torque controller.

A Torque Demand Strategy of IC Engines for Fuel Consumption Improvement using Traffic Information

Abstract This paper presents a torque demand control scheme aimed at fuel efficiency improvement. Synthesizes the traffic flow information and driver acceleration intention, this control scheme can adjust targeted trajectory of the torque demand real-timely by a parametric approach in oder to reduce the fuel consumption. The precise torque estimation based on a static torque detection map determined by engine intake manifold pressure and engine speed has been adopted here. A torque feedback control gives a final throttle command to track with the reference torque demand. Finally, some drive cycle tests have been implemented on the engine-in-the-loop simulation test bench. It has been proved that adjusting the torque demand response trajectory on the basis of driver acceleration intention and traffic flow information can obtain a better fuel efficiency.

Model predictive control of gasoline engines with nonlinear feedback linearized model

Model predictive control (MPC) have received wide attention in many industrial field owing to its optimization capability for the practical control plant with constraints. However, the control performance of the closed-loop system is rarely considered in MPC design. In this paper, a relatively simple performance tuning approach for MPC-based engine speed control is investigated based on the inverse linear quadratic (ILQ) regulator design technique. Considering the nonlinear and time-varying properties of the engine system, the linear tracking dynamical system is deduced from the mean value model of gasoline engines by means of the feedback linearization method. Then the MPC optmization problem can be formulated with this linear tracking dynamics. In order to realize the convenient performance tuning, the MPC quadratic weightings are designed to be only related to a single tuning parameter according to the inverse optimality conditons of LQ problem. The experimental results validate the effectiveness of the proposed speed tracking controller and tuning method provides a trade-off between the response sensitivity and the magnitude of the control input.

Transient control of gasoline engines with C/GMRES

The oil crisis and strict emission legislation have greatly motivated the development of internal combustion engine technology. To improve fuel economy performance and reduce emissions, engine transient control has attracted wide research interests. However, engine system is a nonlinear physical plant with constraints on itself and actuators, and also it is a sophisticated control system involving many control loops to achieve single or multiple objectives. Therefore it is really a challenging issue on the transient control of the engine. In recent years, receding horizon control (RHC) was gained much attention in the field of engine control, owing to its advantages that it can explicitly tackle the constrained optimization and multi-variable control problem. However, a remarkable drawback of RHC is the heavy computation load for on-line optimization algorithm, especially for the nonlinear control system. Indeed, this bottleneck restricts its practical implementation on the industrial electronic controller of a fast control system for a long time. This tutorial paper proposes a systematic receding horizon controller design approach for the transient control applications of the gasoline engines. Two independent RHC-based tracking controllers aimed to achieve the engine torque and speed tracking control are designed, respectively. The control oriented model is derived from the mean-value model of gasoline engines, meanwhile the integrator of the tracking error is embedded to improve the tracking accuracy. All the proposed controllers are verified in real-time on a full-scale gasoline engine and the on-line optimization algorithm for RHC adopts C/GMRES method, which can provide an approximately optimal solution by solving the linear equation instead of the Riccati differential equation. The experimental results demonstrate a large potential for improving the engine transient control performance with RHC scheme.