摘要
知识驱动方法与数据驱动方法是指导工程人员研究电力系统的两大方法论。然而随着电网规模日趋扩大、时变因素日益增多和非线性逐渐增强,基于知识驱动的机理模型方法或基于数据驱动的经验模型方法在电力系统相关应用中将面临更多的挑战。充分利用数据驱动方法与知识驱动方法的互补特性,将二者联合,有望实现应用中综合性能的提升。该文对各研究领域中的数据与知识联合驱动方法进行了整理归纳,进而结合电力系统的特点和需求,梳理了数据与知识联合驱动的典型应用方式,并针对潜在的应用场景进行了详细讨论。最后,在电力系统应用场景中测试验证了数据与知识联合驱动方法的应用效果。
Data-driven and knowledge-driven are two main methodologies guiding researches implemented in power systems. However, these two methodologies encounter effectiveness and flexibility limitations in power system applications, due to large network scale, vast time-varying factors, and strong nonlinearity in power systems. Combining these two methodologies is then believed as an effective way to improve application performances considering the natural complementary features of these two methodologies. In this paper, research advances in combined data-driven and knowledge-driven methodology of various research areas were summarized and archived. On this basis, four typical combined data-driven and knowledge-driven patterns were put forward and discussed on potential applications in power systems. Finally, specific application cases were presented to validate effectiveness of proposed four typical combined patterns.
作者
李峰
王琦
胡健雄
汤奕
LI Feng;WANG Qi;HU Jianxiong;TANG Yi(School of Electrical Engineering,Southeast University,Nanjing 210096,Jiangsu Province,China)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2021年第13期4377-4389,共13页
Proceedings of the CSEE
基金
国家重点研发计划项目(2018YFB0904500)。
关键词
数据驱动
知识驱动
联合驱动
电力系统
data-driven
knowledge-driven
combined drive
power system
