MAiNGO – McCormick-based Algorithm for mixed-integer Nonlinear Global Optimization

MAiNGO – 基于 McCormick 的混合整数非线性全局优化算法

• 批准号: 442664501
• 负责人: Professor Alexander Mitsos, Ph.D.
• 金额: --
• 依托单位: Lehrstuhl Systemverfahrenstechnik (SVT)
• 依托单位国家: 德国
• 项目类别: Research Grants
• 财政年份: 2021
• 资助国家: 德国
• 起止时间: 2020-12-31 至 2021-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/442664501?language=en
• 关键词: MAiNGO; McCormick; based; Algorithm; mixed

The objective of this project is to develop and provide a reliable and novel open-source software for deterministic global optimization of mixed-integer nonlinear optimization problems. A first version of MAiNGO is already being developed at the chair for Process Systems Engineering (AVT.SVT) RWTH Aachen University with the goal of an open-source publication. In contrast to other state-of-the-art deterministic global optimization solvers, MAiNGO does not increase the dimensionality of the user-defined optimization problem. This is achieved through the reduced-space formulation and the application of the so-called McCormick relaxations. The McCormick relaxations are constructed via the open-source package MC++. MAiNGO implements a specialized heuristic for the improvement of the resulting McCormick relaxations and also specialized relaxations for specific intrinsic functions (including functions commonly found in the field of chemical engineering). We have already shown, that MAiNGO exhibits computational advantages for problems formulated in a reduced-space manner. Problems profiting from a reduced-space formulation can be found in flowsheet optimization or data-driven optimization with artificial neural networks.MAiNGO can be used, adapted and extended by the research community. Currently, MAiNGO is a prototype software and misses an adequate documentation and unit testing. Moreover, the software has to be tested on published benchmarks from the global optimization community. The implementation of interfaces to common programming and modelling languages also represents one of goals of this project.

本项目的目标是为混合整数非线性优化问题的确定性全局优化开发并提供一个可靠的、新颖的开源软件。过程系统工程（AVT）主席已经在开发MAiNGO的第一个版本。（SVT）亚琛工业大学，目标是开源出版物。与其他最先进的确定性全局优化求解器相比，MAiNGO不会增加用户定义优化问题的维度。这是通过简化空间公式和所谓麦考密克弛豫的应用来实现的。McCormick松弛是通过开源包mc++构建的。MAiNGO实现了一种专门的启发式方法，用于改进所得到的McCormick松弛和特定内在函数（包括化学工程领域中常见的函数）的专门松弛。我们已经表明，对于以简化空间方式表述的问题，MAiNGO具有计算优势。从简化空间公式中获益的问题可以在流程图优化或人工神经网络的数据驱动优化中找到。MAiNGO可以被研究界使用、调整和扩展。目前，MAiNGO是一个原型软件，缺少足够的文档和单元测试。此外，软件必须在全球优化社区发布的基准测试上进行测试。实现通用编程和建模语言的接口也代表了这个项目的目标之一。