Optimal Problem with Discontinuity

This thesis aims at developing efficient mathematical methods for solving hierarchical dynamic optimization problems. The main motivation is to model processes in nature, for which there is evidence to assume that they run optimally. We describe models of such processes by optimal control problems (called optimal control models (OCMs)). However, an OCM typically includes unknown parameters that cannot be derived entirely on a theoretical basis, which is in particular the case for the cost function. Therefore, we develop parameter estimation techniques to estimate the unknowns in an OCM from observation data of the process. Mathematically, this leads to a hierarchical dynamic optimization problem with a parameter estimation problem on the upper level and an optimal control problem on the lower level. We focus on multi-stage equality and inequality constrained optimal control problems based on nonlinear ordinary differential equations. The main goal of this thesis is to derive numerically efficient mathematical methods for solving hierarchical dynamic optimization problems, and to use these methods to estimate parameters in high-dimensional OCMs from real-world measurement data. We develop parameter-dependent OCMs for the gait of cerebral palsy patients and able-bodied subjects. The unknown parameters in the OCMs are then estimated from real-world motion capture data provided by the Heidelberg MotionLab of the Orthopedic University Clinic Heidelberg by using the mathematical methods developed within this work. The main novelties and contributions of this thesis to the field of hierarchical dynamic optimization are summarized herein. - We establish a novel mathematical method, a so-called direct all-at-once approach, for solving hierarchical dynamic optimization problems based on the direct multiple shooting method and first-order optimality conditions. - Furthermore, we propose an efficient numerical algorithm for large-scale hierarchical dynamic optimization problems, which fully exploits the structures inherited from both the hierarchical setting and the discretization. - Pontryagin's maximum principle is used to analyze solution properties of hierarchical dynamic optimization problems like second-order optimality conditions of the lower-level problem. - In addition, we propose and discuss alternative methods for hierarchical dynamic optimization that are based on derivative-free optimization and a bundle approach. These methods keep the hierarchical problem setting and do not reformulate the lower-level problem using first-order optimality conditions. - We establish a novel lifting method for regularizing mathematical programs with complementarity constraints, which is discussed and numerically investigated by means of a well-known collection of benchmark problems. - Proofs of regularity and convergence results for sequential quadratic programming methods applied to lifted mathematical programs with complementarity constraints are provided. - Efficient state-of-the-art implementations of all mathematical methods derived in this thesis, as well as a benchmark collection of hierarchical dynamic optimization problems are presented. - High-dimensional optimal control gait models for cerebral palsy patients and able-bodied subjects are developed. The mathematical methods derived in this thesis are used to estimate the unknown model parameters from real-world motion capture data provided by the Heidelberg MotionLab of the Orthopedic University Clinic Heidelberg. The theoretical and practical results presented in this thesis can be considered an initial motivating step towards answering open questions in current medical research in fields like treatment planning, classification of gaits, or the evaluation of surgeries by means of hierarchical dynamic optimization.

The work presented in this thesis concerns the optimisation of the hour-to-hour generation schedule (over a scheduling interval of 24 hours) of the generating stations in an all-hydro electric - system. The day-to-day system storage policy is assumed known from the long-term schedule. System load demand is first treated as deterministic. Later, attempts are made to treat it as stochastic. Throughout the text, system transmission losses are represented by means of the B-coefficient loss formula. Chapter 1 presents brief discussions on various topics which may play - a part in the optimization process. In particular, the concept of frequency control and the distinction between long-term and daily scheduling are outlined. The optimization objective is then formulated and some of the known methods of obtaining the solution are briefly discussed. In chapter 2, the optimization problem is first formulated . as that of minimizing the instantaneous total power drawn from the system subject to satisfying a . predetermined time-varying load demand.This is analogous to the minimization of the instantaneous total cost rate in thermal systems. Station characteristic curves are approximated by straight line segments and head variations are assumed negligible. The optimising - conditions are derived using the Lagrange multiplier technique. A small system consisting of 4 stations is studied, where optimisation is achieved by means of an analogue computer. By introducing weighting factors in the objective function, the storage policy can be satisfied. (Analysis carried out in chapter 3 later reveals that this is the correct measure which must be taken to meet the storage policy). Problems do arise, however, when - there are multi-unit stations present in the system. Due to the non-monotonic nature . of the incremental Output curves' of such stations, a. number of solutions would satisfy the optimising conditions. The optimum solution can only be established after testing each possible solution, a prohibitive task for large systems. Approximating the incremental output curves of such stations may create undesirable errors. Another weakness of the method is that it cannot readily take into account the effects of head variations. The application of Pontryagin's Maximum Principle is attempted in chapter 3. It is shown that, for systems with fixed‚ÄövÑvÆ head stations, the optimising conditions are exactly identical to that derived using the Lagrange multiplier technique with weightings being introduced in the objective function. Unfortunately, those problems associated with the presence of multi‚ÄövÑvÆunit stations are still unsolved. Under ideal system conditions, it is shown that optimum generation schedule would result if all stations; except the frequency control station, were step loaded. Step loading mode is defined as an operating mode where the instantaneous station discharge rate is only allowed to take one of two discrete values with stepwise transitions. The analysis indicates that the switching instants do not play any part, except they must be so chosen that the output of the frequency control station is kept within its limits. When step loading technique is applied to non‚ÄövÑvÆideal systems, the optimisation. becomes that of searching for the optimum switching instants of the step loaded stations. This is discussed in chapter 4. For practical reasons, only two step loading modes are assumed admissible. A system consisting of 18 stations is studied. Experimental results indicate that the optimum step loading modes can be established within - a reasonably short computation time. One striking feature of the step loading technique is that it can handle the optimal scheduling problem where the load demand is treated as being stochastic. The effects of head variations can be estimated fairly accurately. The application of the principle of dynamic programming is discussed in chapter 5. Due to the multi-dimensional nature of the problem, this principle must be combined with the relaxation principle to minimise the computation time and computer storage requirements. A good initial guess for the optimum schedule may be obtained using step loading technique. The step loading modes are then modified by applying the combination of dynamic programming and relaxaxtion principles. Finally, chapter 6 outlines the general conclusions drawn from the work presented in the preceeding chapters. The possibility. of applying Pontryagin's Maximum Principle to obtain the long-term storage policy is also studied. Related future research areas are suggested. To the author's best knowledge, the materials presented in chapters 3, 4 and 5 are original. Chapter 2 is based on the work of Chandler Jr. and Gabrielle, where the mathematical error found in their text has been corrected and the weighting procedure, among other materials has been included.

SUMMARYA mathematical model for leukemia therapy based on the Gompertzian law of cell growth is investigated. The effect of the therapy is described in terms of therapy functions. Two types of therapy functions are considered: monotone and non‐monotone. In the first case, the effect of the chemotherapy on the normal cells directly depends on the quantity of the chemotherapeutic agent. In the second case, we consider a therapy function with a threshold effect, which influences the leukemia cells. This effect can describe resistance to the therapy. The optimal control problem has the purpose of minimizing the number of the leukemia cells subject to the constraints of the normal cells, whose number must not fall below a prescribed critical limit and of the amount of the chemotherapeutic agents in the body of the patient (i.e. constraint on the toxicity). Furthermore, a third phase constraint on the cumulative amount of the medicine over the whole therapy process is considered, which can be interpreted as an economic factor of the therapy. This optimization problem is solved analytically with the help of the Pontryagin's Maximum Principle, extension of penalty functions and additional analysis. Copyright © 2013 John Wiley & Sons, Ltd.

This paper treats an optimization problem of time-varying network topology for continuous-time linear network systems. Although the topology from controllers to nodes has been widely investigated in the literature, this paper alternatively focuses on the topology among nodes. For the cost function, we employ the trace of the controllability Gramian to reduce the required control energy. For the constraint, we introduce an upper bound of the L <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sup> norm of the data transmission among nodes to find an effective communication schedule. Since the formulated optimization problem is combinatorial, we introduce a convex relaxation for its computational tractability. After reformulating the problem into an optimal control problem where Pontryagin's maximum principle can be applied, we give a sufficient condition under which the relaxation problem gives a solution of the original problem. We finally provide a numerical algorithm based on a derived optimality condition and illustrate the proposed method with an example.

&lt;p style='text-indent:20px;'&gt;In this work, we have proposed a new approach for solving the linear-quadratic optimal control problem, where the quality criterion is a quadratic function, which can be convex or non-convex. In this approach, we transform the continuous optimal control problem into a quadratic optimization problem using the Cauchy discretization technique, then we solve it with the active-set method. In order to study the efficiency and the accuracy of the proposed approach, we developed an implementation with MATLAB, and we performed numerical experiments on several convex and non-convex linear-quadratic optimal control problems. The obtained simulation results show that our method is more accurate and more efficient than the method using the classical Euler discretization technique. Furthermore, it was shown that our method fastly converges to the optimal control of the continuous problem found analytically using the Pontryagin's maximum principle.&lt;/p&gt;

Recent work linking deep neural networks and dynamical systems opened up new avenues to analyze deep learning. In particular, it is observed that new insights can be obtained by recasting deep learning as an optimal control problem on difference or differential equations. However, the mathematical aspects of such a formulation have not been systematically explored. This paper introduces the mathematical formulation of the population risk minimization problem in deep learning as a mean-field optimal control problem. Mirroring the development of classical optimal control, we state and prove optimality conditions of both the Hamilton-Jacobi-Bellman type and the Pontryagin type. These mean-field results reflect the probabilistic nature of the learning problem. In addition, by appealing to the mean-field Pontryagin's maximum principle, we establish some quantitative relationships between population and empirical learning problems. This serves to establish a mathematical foundation for investigating the algorithmic and theoretical connections between optimal control and deep learning.

1 A Survey on Computational Optimal Control.- Issues in the Direct Transcription of Optimal Control Problems to Sparse Nonlinear Programs.- Optimization in Control of Robots.- Large-scale SQP Methods and their Application in Trajectory Optimization.- Solving Optimal Control and Pursuit-Evasion Game Problems of High Complexity.- 2 Theoretical Aspects of Optimal Control and Nonlinear Programming.- Continuation Methods In Boundary Value Problems.- Second Order Optimality Conditions for Singular Extremals.- Synthesis of Adaptive Optimal Controls for Linear Dynamic Systems.- Control Applications of Reduced SQP Methods.- Time Optimal Control of Mechanical Systems.- 3 Algorithms for Optimal Control Calculations.- Second Order Algorithm for Time Optimal Control of a Linear System.- An SQP-type Solution Method for Constrained Discrete-Time Optimal Control Problems.- Numerical Methods for Solving Differential Games, Prospective Applications to Technical Problems.- Construction of the Optimal Feedback Controller for Constrained Optimal Control Problems with Unknown Disturbances.- Repetitive Optimization for Predictive Control of Dynamic Systems under Uncertainty.- Optimal Control of Multistage Systems Described by High-Index Differential-Algebraic Equations.- A New Class of a High Order Interior Point Method for the Solution of Convex Semiinfinite Optimization Problems.- A Structured Interior Point SQP Method for Nonlinear Optimal Control Problems.- 4 Software for Optimal Control Calculations.- Automated Approach for Optimizing Dynamic Systems.- ANDECS: A Computation Environment for Control Applications of Optimization.- Application of Automatic Differentiation to Optimal Control Problems.- OCCAL: A mixed symbolic-numeric Optimal Control CALculator.- 5 Applications of Optimal Control.- A Robotic Satellite with Simplified Design.- Nonlinear Control under Constraints of a Biological System.- An Object-Oriented Approach to Optimally Describe and Specify a SCADA System Applied to a Power Network.- Near-Optimal Flight Trajectories Generated by Neural Networks.- Performance of a Feedback Method with Respect to Changes in the Air-Density during the Ascent of a Two-Stage-To-Orbit Vehicle.- Linear Optimal Control for Reentry Flight.- Steady-State Modelling of Turbine Engine with Controllers.- Shortest Paths for Satellite Mounted Robot Manipulators.- Optimal Control of the Industrial Robot Manutec r3.

This paper addresses the problem of time-optional motions for a mobile platform in a planar environment. The platform has two non-steerable independently driven wheels. The overall mission of the robot is expressed in terms of a sequence of via points at which the platform must be at rest in a given configuration (position and orientation). The objective is to plan time-optimal trajectories between these configurations assuming an unobstructed environment. Using Pontryagin's maximum principle (PMP), we formally demonstrate that all time optimal motions of the platform for this problem occur for bang-bang controls on the wheels (at each instant, the acceleration on each wheel is either at its upper or lower limit). The PMP, however, only provides necessary conditions for time optimality. To find the time optimal robot trajectories, we first parameterize the bang-bang trajectories using the switch times on the wheels (the times at which the wheel accelerations change sign). With this parameterization, we can fully search the robot trajectory space and find the switch times that will produce particular paths to a desired final configuration of the platform. We show numerically that robot trajectories with three switch times (two on one wheel, one on the other) can reach any position, while trajectories with four switch times can reach any configuration. By numerical comparison with other trajectories involving similar or greater numbers of switch times, we then identify the sets of time-optimal trajectories. These are uniquely defined using ranges of the parameters, and consist of subsets of trajectories with three switch times for the problem when the final orientation of the robot is not specified, and four switch times when a full final configuration is specified. We conclude with a description of the use of the method for trajectory planning for one of our robots.

In this article, the optimal boundary control for a generalised Korteweg–de Vries equation is considered. The Dubovitskii and Milyutin functional analytical approach is adopted in the investigation of Pontryagin's maximum principles of the system in both fixed and free final horizon cases. The necessary conditions are, respectively, presented for the optimal boundary control problems in these two cases.

A geometric approach to time-dependent optimal control problems is proposed. This formulation is based on the Skinner and Rusk formalism for Lagrangian and Hamiltonian systems. The corresponding unified formalism developed for optimal control systems allows us to formulate geometrically the necessary conditions given by a weak form of Pontryagin's maximum principle, provided that the differentiability with respect to controls is assumed and the space of controls is open. Furthermore, our method is also valid for implicit optimal control systems and, in particular, for the so-called descriptor systems (optimal control problems including both differential and algebraic equations).

This article presents a unified approach to simultaneous solution of optimization and optimal control problems in batch distillation, operating under different modes of operation: variable, constant, or optimal reflux. The simplified, computationally efficient short‐cut method and a novel algorithm to solve the optimal control problems in batch distillation is the basis of this unified approach. The short‐cut method identifies the feasible region of operation essential for optimization and optimal control problems, and provides analytical partial derivatives of the model parameters crucial to the solution.The new algorithm for the solution of optimal combination of the maximum principle and NLP optimization techniques. It circumvents the problems associated with the maximum principle approach (iterative solution of a two‐point boundary value problem, undounded control variables, and inability to handle the simultaneous optimization and optimal control problem), and the coupled ODE discretization‐NLP optimization scheme for nonlinear models (higher system nonlinearities, multiplicity of solutions, sensitivity of convergence to initial guesses). This algorithm reduces the dimensionality of the problem, and the nature of the algorithm allows a common platform to optimal solutions of different operating conditions. This article also shows that different categories of the optimal control problems in batch distillation essentially involve the solution of the maximum distillate problem.

Some optimization problems concerning a substrate in a fluid are considered. The concentration of the substrate is affected by diffusion, convection, and elimination by enzymes, and the problem is to find the optimal distribution of enzymes. In this paper, the rate of elimination and the transmission coefficient are optimized. Mathematically, these problems are optimal control problems, and they are analyzed by means of Pontryagin's maximum principle.

In this paper we define a class of optimal control problems which we denote “embedded optimal control problems”. These are not true optimal control problems since the control system is not locally controllable on the manifold on which it is defined. Despite this, they allow for a well defined associated optimal control problem which does not admit abnormal extremals. We apply Pontryagin's maximum principle to the embedded optimal control problem to derive the generating differential equations for the normal and abnormal extremals. We show that the normal extremal generating equations in a sense contain the extremal generating equations for the associated optimal control problem. We show that this is not the case for the abnormal extremal generating equations. This has applications to the study of the optimal control of systems constrained to a given submanifold of a configuration space, for example the sphere or hypersphere. We apply the theory to three examples in order to illustrate its applicability and to show how it relates to well known results.

A co-infection model between HIV and COVID-19 that takes into account COVID-19 vaccination and public awareness is discussed in this article. Rigorous analysis of the model is conducted to establish the existence and local stability conditions of the single-infection models. We discover that when the corresponding reproduction number for COVID-19 and HIV exceeds one, the disease continues to exist in both single-infection models. Furthermore, HIV will always be eradicated if its reproduction number is less than one. Nevertheless, this does not apply to the single-infection COVID-19 model. Even when the fundamental reproduction number is less than one, an endemic equilibrium point may exist due to the potential for a backward bifurcation phenomenon. Consequently, in the single-infection COVID-19 model, bistability between the endemic and disease-free equilibrium may arise when the basic reproduction number is less than one. From the co-infection model, we find that the reproduction number of the co-infection model is the maximum value between the reproduction number of HIV and COVID-19. Our numerical continuation experiments on the co-infection model reveal a threshold indicating that both HIV and COVID-19 may coexist within the population. The disease-free equilibrium for both HIV and COVID-19 is stable only if the reproduction numbers are less than one. Additionally, our two-parameter continuation analysis of the bifurcation diagram shows that the condition where both reproduction numbers equal one serves as an organizing center for the dynamic behavior of the co-infection model. An extended version of our model incorporates four different interventions: face mask usage, vaccination, and public awareness for COVID-19, as well as condom use for HIV, formulated as an optimal control problem. The Pontryagin's Maximum Principle is employed to characterize the optimal control problem, which is solved using a forward-backward iterative method. Numerical investigations of the optimal control model highlight the critical role of a well-designed combination of interventions to achieve optimal reductions in the spread of both HIV and COVID-19.

This paper aims to apply an optimal control theory for the autonomous model of the leptospirosis epidemic to examine the effect of four time-dependent control measures on the model dynamics with cost-effectiveness. Pontryagin's Maximum Principle was used to derive the optimality system associated with the optimal control problem. Numerical simulations of the optimality system were performed for different control strategies and the results were presented graphically with and without controls. The optimality system was simulated using the Forward–Backward Sweep method in the Matlab programme. The numerical results revealed that the combination of all optimal control measures is the most effective strategy for minimizing the spread and impact of disease in the community. Furthermore, a cost-effectiveness analysis was performed to determine the most cost-effective strategy using the incremental cost-effectiveness ratio approach and we observed that the rodenticide control-only strategy is most effective to combat the spread of disease when available resources are limited.