Study On Nonlinear Estimation And Control Of Gasline Engine

To improve fuel economy, reduce emission and enhance driving performance, manyscholars and car manufacturers get together to accelerate the research and developmentof the automobile engine control systems and related technology products. The applica-tion of new technologies relay on many new electronic control actuators, which make thegasoline electrical control system more and more complex. In order to shorten the devel-opment cycle of the control system design, reduce the calibration of the development ofcontrol software and improve the control performance, a large number of automotive re-searchers focus on model-based control system design method. This thesis uses the newlydeveloped robust nonlinear control theory, including feedback linearization, Backsteppingtechnology and input to state stability (ISS) theory, to address some typical estimationand control issues of gasoline engine. The following issues are investigated: estimation ofengine torque, airpath control system design of gasoline engine, electronic throttle controland idle speed control.In this paper, we consider the traditional four-cylinder PFI gasoline engine as theresearch object, using the mean value modeling theory, the engine volumetric efciencyand which are difcult to mathematical description are considered as map experimen-tal data commonly utilized in automobile engineering, a physics/map mixed descriptionengine model is established for controller design. The engine dynamic model enDYNAwhich has high precision is used as a virtual engine to calibrate the parameters of enginemodel and data of the maps. In this case, torque-based engine control architecture iswidely adopted to manage of the engine torque output. In this paper, the airpath controlsystem is designed to achieve the engine torque demand. Firstly, the control system basedon inverse-dynamic is designed, including the intake manifold pressure planning, throttleopening planning and tracking. This method can be applied in the case of simple struc-tured controller, easy realization in hardware, and low precision of systems, but steadyerror is much larger when the engine speed is low. In order to improve the performanceof the engine torque output, a dual closed-loop airpath control system is designed, thecontrol system consists of a intake manifold pressure tracking controller which is designedusing feedback linearization method and a electronic throttle tracking controller. Fol-lowed by ofine simulation results show the dual closed-loop airpath control system hasbetter tracking performance, the hardware in-the-loop platform based on xPC-Target and dSPACE real-time system is used to verify that dual-closed-loop air path control systemnot only ensures a good quality of the performance for reliability in real-time but also theefectiveness in practical application. The experiment results show that the control sys-tem mentioned above gives a good tracking performance even in a more actual condition.And it is emphasized that this study is of great worth in practical application.Electronic throttle as the eventually implementing agencies, its control performanceplays a crucial role in airpath control. Therefore, we choose a electronic throttle body incommercial cars as control objective, a control-oriented nonlinear model is established.In order to verify the reliability of the model, the model validation is carried out usingthe electronic throttle body. For the nonlinear characteristics of the electronic throttlesystem, the Backstepping technology is applied to derive the nonlinear controller, and areduced-order observer is designed to estimate the unmeasurable state. Combining thenonlinear controller and the observer, an observer-based output feedback control schemeis presented. Robustness of the tracking error system is discussed in the framework of ISStheory, where model uncertainties are considered as additive disturbance inputs. Basedon this, a guideline for selecting the controller parameters is given. In order to reduce theofset, an integral action of the tracking error is introduced. On the basis of good of-linesimulation results, the rapid prototyping experiment and HiL experiments are carried outto demonstrate the efectiveness of the proposed control scheme.The idle speed control is one of the most important automotive control problemsto reduce emission. In this paper, based on the physics/map mixed description enginemodel, a nonlinear idle speed controller is derived using Backstepping and the robustnessof the tracking error system is discussed in the framework of ISS theory, and a guidelinefor selecting the controller parameters is given. For the known external load torque,ignition advance angle compensation algorithm is introduced to reduce the influence ofload torque. The simulation results verify the efectiveness of the controller.The engine torque is a key variable in engine torque management system. In a lab-oratory, the indicated torque is usually calculated by in-cylinder pressure obtained frompressure transducers. The efective torque can be also obtained by subtracting torquelosses. However, such in-cylinder pressure sensors are too expensive and not technicallyavailable for commercial production engines. Thus, indirect methods to measure enginetorque are required. In this paper, based on the physics/map mixed description enginemodel, the feedback linearization method is adopted to design the nonlinear engine speedtracking controller. The actual engine efective torque is obtained by the efective torquegenerated by the torque production model and the torque compensation from the track-ing controller. The simulation results show that the nonlinear tracking controller has good tracking performance. Therefore, this method can be estimated the engine torqueaccurately. Furthermore, the enDYNA-based simulation results verify the validity of theestimation method.In this thesis, the nonlinear control theory is applied for the nonlinear control and es-timation problem of automotive engine. Airpath control algorithm and torque estimationalgorithm are designed using feedback linearization; Nonlinear method of "Backstepping+ISS" is proposed to designed the electronic throttle algorithm and idle speed algorithm.Introduction of ISS into Backstepping derivation guarantees the controller's robustnessagainst the modeling uncertainties, the above algorithms all achieve good control efect.However, further research needs to be done since some problems are still remain to besolved gradually. For example:(1)the nonlinearity of the gear backlash is ignored in theelectronic throttle controller, which needs to be considered to improve the control efect;(2)the torque estimation algorithm and idle speed control algorithm are only tested by of-line simulation and rapid prototyping experiment, therefore, the HiL experiment shouldbe carried out for further test;(3)the airpath control algorithms and electronic throttlecontrol algorithm does not co-simulate, therefore, it is waiting for being solved in thefuture study;(4)the in-vehicle test has not been carried out for all algorithms, the majorfuture work should be the implementations of bench and in-vehicle experiments.