The Global Optimization Algorithm Of Filled Function And Its Application In Optimal Control Problems

As the filled function method develops in the field of global optimization theory and optimal control,the scholars in the world pay more attention to how to use the filled function method solves the optimal control problem.In this dissertation,the optimal control problem is transformed into a nonlinear programming problem based on the con-trol parameterization method,then we discuss the optimality conditions of the discrete time optimal control problem,the global optimal algorithm of filled function and its ap-plication for agricultural products quality traceability problem.Specific content is as follows:In chapter one,we introduce the global optimization filled function method principle and the development of the status quo.Then,we give a definition of the optimal control problem and the preliminary knowledge associated with this paper,including the con-trol parameterization method,time-scaling conversion and the adjoint equation method.Finally,we get a brief introduction to optimal enforcement problems of agricultural prod-ucts quality traceability system and the applications of the related theories to agricultural products quality traceability problems.In chapter two,we give a class of Lipschitz condition for the discrete-time control problems that the system has unique solution.Further,we present the control variable transformation function by using the control parameterization method.As a result,the optimal control problem we are considering is converted to a nonlinear differentiable programming problem.Then we put forward a class of first-order optimality conditions for this optimal control problem.Finally,two examples are provided to demonstrate the effectiveness of the proposed first-order optimality conditions.In chapter three,for unconstrained optimal control problem,we give a class of filled functions with parameter free and prove the properties of these filled functions.Then,we give the corresponding algorithm and specific examples.In chapter four,comparing unconstrained optimal control problem with continuous time optimal control problem,we creatively design time-scaling transformation method.We construct a class of filled functions with one parameter and prove the properties of these filled functions.At last,we give the corresponding algorithm and specific examples.In chapter five,we discuss the optimal quality supervision and execution strategy problem of OTO agricultural product quality traceability system.We firstly make the quality of supervision and enforcement as a control variable,and apply level analysis method and logistic regression analysis to establish the discrete-time optimal control model.Secondly,used the control parameterization method,the optimal control problem is transformed into a nonlinear differentiable optimization problem.Finally,we give the globally optimal quality supervision and execution strategy applying related algorithms.