基于混合智能的新型电力系统运行方式分析决策架构及其关键技术

Architecture and Key Technologies of Hybrid-Intelligence-Based Decision-Making of Operation Modes for New Type Power Systems

随着新型电力系统建设,方式编制需要考虑的运行场景数量和计算工作量大大增加,安全稳定机理愈发复杂,安全运行边界的不确定性增强,运行方式调整的难度显著增大。传统基于人工经验的运行方式决策模式难以为继,人工智能提供了新的解决思路,但单纯依靠人工智能方法仍面临样本不足、可解释性差、探索效率低等挑战。聚焦新型电力系统运行方式决策这一具体问题,提出了基于混合智能的新型电力系统运行方式决策研究框架,从运行方式样本生成、安全稳定影响因素分析与边界刻画、运行方式智能调整、模型可解释性与迁移更新4个方面展开分析和探讨,为将混合智能应用于新型电力系统提供可行的技术路径。

With the construction of new type power systems,the number of operating scenarios and computational workload that need to be considered in analyzing operation modes increases greatly,the safety and stability mechanism becomes more complex,the uncertainty of the safe operation boundary increases,and the difficulty of adjusting operating mode increases significantly.The traditional decision-making method based on human experience is facing major challenges.Artificial intelligence provides new solutions but still faces challenges such as insufficient samples,poor interpretability,low exploration efficiency,etc.Focusing on the specific problems of the operation mode decision-making of the new type power systems,this paper proposes a research framework for operation mode decision-making of the new type power systems based on hybrid intelligence.The analysis and discussion are carried out from four aspects:sample generation of operation mode,analysis of the safety boundary and stability influencing factors,intelligent adjusting of operation mode,model interpretability and update,which provide a feasible technology path for applying hybrid intelligence to new type power systems.