计及压缩机运行域的动态最优电-气能流凸化算法
Convexification Algorithm for Dynamic Optimal Power-gas Flow Considering Compressor Operation Domain
陈飞雄 1郑翔昊 1邵振国 1林勇棋 1颜熙颖1
作者信息
- 1. 福建省电器智能化工程技术研究中心(福州大学电气工程与自动化学院),福建省 福州市 350108
- 折叠
摘要
压缩机是电-气互联系统的关键元件,其运行工作点是否位于运行域内关系到系统的安全稳定运行,同时其模型非凸性大幅增加了最优电-气能流的求解难度.为此,该文在压缩机准确建模的基础上,针对电力和天然气系统迥异的传输特性,提出一种计及压缩机运行域的动态最优电-气能流凸化算法.首先,根据压缩机运行参数与状态变量间的映射关系,构建压缩机三维运行域模型,并利用超平面实现运行域的凸包络松弛.其次,考虑差异化传输特性建立动态最优电-气能流模型,实现模型的混合整数二阶锥凸重构,并提出一种基于动态线性化与紧缩凸包络的凸化算法.最后,算例验证所提模型的有效性,表明所提算法在计算精度和效率上均优于连续锥规划算法.
Abstract
The compressor is a key component of the integrated electricity and gas systems.The position of its operational point within the operational domain significantly impacts the safe and stable functioning of the entire system.Meanwhile,its model non-convexity significantly increases the difficulty of solving the optimal power-gas flow.A dynamic optimal power-gas flow convexification algorithm considering the operation domain is proposed based on the accurate compressor modeling and different transmission characteristics of the power and natural gas systems.First,a 3D compressor operation domain model is constructed based on the mapping relationship between the operating parameters and the state variables,and the convex envelope relaxation of the operation domain is realized through the hyperplanes.Then,a dynamic optimal power-gas flow model considering the different transmission characteristics is developed,a mixed integer second-order cone convex reformulation of the model is realized,and a convexification algorithm based on dynamic linearization and tightening convex envelope is proposed.Finally,results demonstrate the effectiveness of the proposed model,indicating that the proposed algorithm outperforms the sequential cone programming algorithm in both computational accuracy and efficiency.
关键词
电-气互联系统/最优电-气能流/压缩机运行域/凸包络/二阶锥松弛Key words
integrated electricity and gas system/optimal power-gas flow/compressor operation domain/convex envelope/second-order cone relaxation引用本文复制引用
基金项目
国家自然科学基金项目(52107080)
福建省自然科学基金项目(2021J05135)
出版年
2024
国家科技期刊平台
NSTL
万方数据