Study On Adaptive Robust Control Of Automotive Electronic Throttle System With Fuzzy Optimal Design

With the development of automobile industry,the increasing shortage of oil resources and the wors-ening air pollution,people put forward higher requirements of automotive economy,safety and emission when pursuing good power performance and stability.As an important part of the electronic control sys-tem in automotive engine,the electronic throttle(ET)control system aims to reduce gas emission,provide a better fuel economy and further enhance the vehicle drivability.Hence the precise control of the ET system has attracted a significant amount of research interest from automotive engineers and researchers.This dissertation focuses on studying deterministic Lyapunov-based robust and adaptive robust con-trol for automotive electronic throttle(ET)system with uncertainty and fuzzy optimal design.The main research contents and achievements of this dissertation are as follows:(1)This dissertation presents a systematic approach to study Udwadia-Kalaba theory which provides a general three-step procedure to establish the fundamental equation of motion for constrained system.When only passive constraints(i.e.,modeling constraints)are considered,the environment or structure of the mechanical system should provide the required constraint forces.When the mechanical system is expected to follow a set of servo constraints(i.e.,control objectives),the task then is to find the generalized control forces applied to the system.Udwadia-Kalaba theory serves as a basis for the robust control and adaptive robust control design for the mechanical system with uncertainty.(2)This dissertation establishes the dynamical model of the ET system,considering the three typical nonlinearities of the ET system:nonlinear friction,nonlinear spring and gear backlash.In the case that the uncertainties are known but there exists the problem of initial condition deviation from the constraint(i.e.,the initial.condition of the opening angle of the ET system can not satisfy the given constraint),this dissertation proposes a modified control on the basis of Udwadia-Kalaba theory.In comparison with the direct control of Udwadia-Kalaba theory,numerical simulation results show that this modified control can effectively address the problem of initial condition deviation from the constraint.(3)This dissertation creatively uses the fuzzy set theory to describe the system uncertainty of the ET system and establishes the fuzzy ET dynamic system.On this basis,from the perspective of trajectory tracking,this dissertation proposes a robust control scheme which is deterministic and is not the usual if-then rules-based control to guarantee the controlled system to achieve the deterministic performance:uniform boundedness and uniform ultimate boundedness.Furthermore,a fuzzy-based system performance index including the size of the uniform ultimate boundedness and control cost is proposed based on the fuzzy information.Then an optimal design problem is formulated and finally solved.(4)This dissertation proposes a robust control scheme which is designed on the basis of Udwadia-Kalaba theory and second order servo constraint from the perspective of constraint following to deal with the control problem of the fuzzy ET dynamic system.It is important to stress that various control prob-lems,including stabilization,trajectory following,and optimality,can be cast into the second order servo constraint.That is,the trajectory tracking problem can be considered as a special case of the constrained following control problem.The proposed robust constraint following control scheme is also deterministic and not the usual if-then rules-based,and is able to guarantee the deterministic performance:uniform boundedness and uniform ultimate boundedness.A fuzzy-based system performance index which consists of three parts:the average of the overall transient performance,the average of the steady state performance and the control cost is proposed.The optimal design problem is finally solved by minimizing the perfor-mance index.(5)This dissertation proposes an adaptive robust constraint following control method to deal with the fuzzy ET dynamic system considering that robust control is usually conservative.The adap-tive law which is constructed to estimate the possible bound of the uncertainty,has three typical features:(i)it is given in a general form;(ii)it is a leakage type;(iii)the leakage gain is performance-based.For optimal design,a novel concept:confidence index is proposed to measure the expectation value of a fuzzy number.With this optimal adaptive robust constraint following control,the performance of the fuzzy ET system is both deterministically guaranteed and fuzzily optimized.In comparison with the robust con-straint following control,this control method has higher tracking accuracy and better robustness.