A Global Convergent Line Search Filter SQP Method
Yuqing Wang
Abstract
Yuqing Wang
Abstract
The paper presents a line search filter sequential quadratic programming(SQP) method for inequality constrained optimization.Compared to traditional SQP methods,the advantages are that the quadratic programming(QP) subproblem is always consistent and the penalty function is not required by filter strategy.Under some mild conditions the global convergence can be induced.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The paper presents a line search filter sequential quadratic programming(SQP) method for inequality constrained optimization.Compared to traditional SQP methods,the advantages are that the quadratic programming(QP) subproblem is always consistent and the penalty function is not required by filter strategy.Under some mild conditions the global convergence can be induced.
Key concepts: Sequential quadratic programming, Line search, Mathematical optimization, Convergence (economics), Quadratic programming, Filter (signal processing), Penalty method, Line (geometry)