大规模非线性优化软件包UniOptima的设计和开发

Design and Implementation of Large-scale Nonlinear Optimization Software UniOptima

根据将SQP算法应用于大规模问题求解的两种改造思路,并结合目前这方面的研究成果,在M atlab环境下开发了大规模非线性优化工具箱Un iOptim a。Un iOptim a包括两个非线性求解器RSQP和THSQP,RSQP专门应用于自由度相对较小的等式约束优化问题,THSQP则应用于一般的非线性优化问题。本优化工具箱采用M atlab语言开发,并与M atlab优化工具箱保持了良好的兼容性,大大方便了用户使用。求解器中设有很多优化选项,用户可根据不同情况或不同需要选择不同的优化选项,增强了用户和程序的交互性。

SQP algorithms are viewed as the best method for solving nonlinear problems, and when applied to solve large- scale problems, they can be classified as full-space methods and reduced space methods. Based on the two kinds of SQP algorithms, large-scale nonlinear optimization software named UniOptima is developed. In UniOptima there are two nonlinear solvers named RSQP and THSQP. RSQP is designed to solve large-scale problems with large number of equality constraints and relatively few degrees of freedom, and THSQP is designed to solve general nonlinear problems. The optimization software has many options for users to select, is developed under Matlab language and is compatible with the optimization tools in Matlab, which will largely increase the convenience for users.