A Sequential Algorithm for Solving Nonlinear Optimization Problems with Chance Constraints

Frank E. Curtis, Andreas Waechter, Victor M. Zavala

Research output: Book/ReportOther report

Abstract

An algorithm is presented for solving nonlinear optimization problems with chance constraints, i.e., those in which a constraint involving an uncertain parameter must be satisfied with at least a minimum probability. In particular, the algorithm is designed to solve cardinality-constrained nonlinear optimization problems that arise in sample average approximations of chance-constrained problems, as well as in other applications in which it is only desired to enforce a minimum number of constraints. The algorithm employs a novel penalty function, which is minimized sequentially by solving quadratic optimization subproblems with linear cardinality constraints. Properties of minimizers of the penalty function in relation to minimizers of the corresponding nonlinear optimization problem are presented, and convergence of the proposed algorithm to a stationary point of the penalty function is proved. The effectiveness of the algorithm is demonstrated through numerical experiments with a nonlinear cash flow problem.
Original languageEnglish (US)
PublisherUnknown Publisher
StatePublished - 2016

Fingerprint

Chance Constraints
Sequential Algorithm
Nonlinear Optimization
Nonlinear Problem
Optimization Problem
Penalty Function
Minimizer
Sample Average Approximation
Cardinality Constraints
Quadratic Optimization
Uncertain Parameters
Stationary point
Constrained Optimization
Linear Constraints
Cardinality
Numerical Experiment

Cite this

@book{5b2a131c9a544ae8be72778ada8307ce,
title = "A Sequential Algorithm for Solving Nonlinear Optimization Problems with Chance Constraints",
abstract = "An algorithm is presented for solving nonlinear optimization problems with chance constraints, i.e., those in which a constraint involving an uncertain parameter must be satisfied with at least a minimum probability. In particular, the algorithm is designed to solve cardinality-constrained nonlinear optimization problems that arise in sample average approximations of chance-constrained problems, as well as in other applications in which it is only desired to enforce a minimum number of constraints. The algorithm employs a novel penalty function, which is minimized sequentially by solving quadratic optimization subproblems with linear cardinality constraints. Properties of minimizers of the penalty function in relation to minimizers of the corresponding nonlinear optimization problem are presented, and convergence of the proposed algorithm to a stationary point of the penalty function is proved. The effectiveness of the algorithm is demonstrated through numerical experiments with a nonlinear cash flow problem.",
author = "Curtis, {Frank E.} and Andreas Waechter and Zavala, {Victor M.}",
year = "2016",
language = "English (US)",
publisher = "Unknown Publisher",

}

A Sequential Algorithm for Solving Nonlinear Optimization Problems with Chance Constraints. / Curtis, Frank E.; Waechter, Andreas; Zavala, Victor M.

Unknown Publisher, 2016.

Research output: Book/ReportOther report

TY - BOOK

T1 - A Sequential Algorithm for Solving Nonlinear Optimization Problems with Chance Constraints

AU - Curtis, Frank E.

AU - Waechter, Andreas

AU - Zavala, Victor M.

PY - 2016

Y1 - 2016

N2 - An algorithm is presented for solving nonlinear optimization problems with chance constraints, i.e., those in which a constraint involving an uncertain parameter must be satisfied with at least a minimum probability. In particular, the algorithm is designed to solve cardinality-constrained nonlinear optimization problems that arise in sample average approximations of chance-constrained problems, as well as in other applications in which it is only desired to enforce a minimum number of constraints. The algorithm employs a novel penalty function, which is minimized sequentially by solving quadratic optimization subproblems with linear cardinality constraints. Properties of minimizers of the penalty function in relation to minimizers of the corresponding nonlinear optimization problem are presented, and convergence of the proposed algorithm to a stationary point of the penalty function is proved. The effectiveness of the algorithm is demonstrated through numerical experiments with a nonlinear cash flow problem.

AB - An algorithm is presented for solving nonlinear optimization problems with chance constraints, i.e., those in which a constraint involving an uncertain parameter must be satisfied with at least a minimum probability. In particular, the algorithm is designed to solve cardinality-constrained nonlinear optimization problems that arise in sample average approximations of chance-constrained problems, as well as in other applications in which it is only desired to enforce a minimum number of constraints. The algorithm employs a novel penalty function, which is minimized sequentially by solving quadratic optimization subproblems with linear cardinality constraints. Properties of minimizers of the penalty function in relation to minimizers of the corresponding nonlinear optimization problem are presented, and convergence of the proposed algorithm to a stationary point of the penalty function is proved. The effectiveness of the algorithm is demonstrated through numerical experiments with a nonlinear cash flow problem.

M3 - Other report

BT - A Sequential Algorithm for Solving Nonlinear Optimization Problems with Chance Constraints

PB - Unknown Publisher

ER -