Efficient symmetric Hessian propagation for direct optimal control

doi:10.1016/j.jprocont.2016.11.011

	Efficient symmetric Hessian propagation for direct optimal control
	Quirynen, Rien 1,3; Houska, Boris2 ; Diehl, Moritz 3
	2017-02
发表期刊	JOURNAL OF PROCESS CONTROL (IF:3.3[JCR-2023],3.6[5-Year])
ISSN	0959-1524
卷号	50 页码:19-28
发表状态	已发表
DOI	10.1016/j.jprocont.2016.11.011
摘要	Direct optimal control algorithms first discretize the continuous-time optimal control problem and then solve the resulting finite dimensional optimization problem. If Newton type optimization algorithms are used for solving the discretized problem, accurate first as well as second order sensitivity information needs to be computed. This article develops a novel approach for computing Hessian matrices which is tailored for optimal control. Algorithmic differentiation based schemes are proposed for both discrete and continuous-time sensitivity propagation, including explicit as well as implicit systems of equations. The presented method exploits the symmetry of Hessian matrices, which typically results in a computational speedup of about factor 2 over standard differentiation techniques. These symmetric sensitivity equations additionally allow for a three-sweep propagation technique that can significantly reduce the memory requirements, by avoiding the need to store a trajectory of forward sensitivities. The performance of this symmetric sensitivity propagation is demonstrated for the benchmark case study of the economic optimal control of a nonlinear biochemical reactor, based on the open-source software implementation in the ACADO Toolkit. (C) 2016 Elsevier Ltd. All rights reserved.
关键词	Optimal control Sensitivity analysis Algorithms and software Nonlinear predictive control
收录类别	SCI ; EI
语种	英语
资助项目	ShanghaiTech University[F-0203-14-012]
WOS研究方向	Automation & Control Systems ; Engineering
WOS类目	Automation & Control Systems ; Engineering, Chemical
WOS记录号	WOS:000392682300003
出版者	ELSEVIER SCI LTD
EI入藏号	20170103217223
EI主题词	Benchmarking ; Bioreactors ; Open source software ; Open systems ; Optimal control systems ; Optimization ; Problem solving ; Sensitivity analysis ; Software engineering
EI分类号	Biotechnology:461.8 ; Computer Software, Data Handling and Applications:723 ; Computer Programming:723.1 ; Control Systems:731.1 ; Mathematics:921 ; Optimization Techniques:921.5 ; Systems Science:961
WOS关键词	DIFFERENTIAL-ALGEBRAIC EQUATIONS ; ADJOINT SENSITIVITY-ANALYSIS ; DYNAMIC OPTIMIZATION ; PERIODIC OPERATIONS ; ALGORITHMS ; SOFTWARE ; MPC
原始文献类型	Article
引用统计	正在获取...
文献类型	期刊论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/1522
专题	信息科学与技术学院_PI研究组_Boris Houska组
通讯作者	Quirynen, Rien
作者单位	1.KU Leuven Univ, Dept ESAT STADIUS, Kasteelpk Arenberg 10, B-3001 Leuven, Belgium 2.ShanghaiTech Univ, Sch Informat Sci & Technol, 319 Yueyang Rd, Shanghai 200031, Peoples R China 3.Univ Freiburg, Dept IMTEK, Georges Koehler Allee 102, D-79110 Freiburg, Germany
推荐引用方式 GB/T 7714	Quirynen, Rien,Houska, Boris,Diehl, Moritz. Efficient symmetric Hessian propagation for direct optimal control[J]. JOURNAL OF PROCESS CONTROL,2017,50:19-28.
APA	Quirynen, Rien,Houska, Boris,&Diehl, Moritz.(2017).Efficient symmetric Hessian propagation for direct optimal control.JOURNAL OF PROCESS CONTROL,50,19-28.
MLA	Quirynen, Rien,et al."Efficient symmetric Hessian propagation for direct optimal control".JOURNAL OF PROCESS CONTROL 50(2017):19-28.