Abstract:With the construction of UHV power grid, integration of large-scale renewable clean energy, and large-scale energy base putting into operation, the power grid dispatching faced with more and more complex challenges. On the basis of existing research results, architecture of intelligent dispatching based on situation awareness is proposed, so as to accurately achieve prevention and control of the power system. The shortcomings of traditional dispatching mode are analyzed firstly, and the concepts and characterization approaches of grid situational awareness and operation state trajectory of power grid are then introduced. The overall objective of intelligent dispatching is presented, including data processing and integrated knowledge mining, predictive perception of grid operation, risk analysis and comprehensive early warning, so as to achieve "automatic cruise under normal operating conditions, automatic navigation under abnormal operating conditions ". The functional framework of intelligent dispatching is also proposed in details, including four major aspects of the perception and forecasts, risk analysis, decision-making support, and automatic control, as well as three supporting functions such as post-assessment of dispatching, trajectory index calculation, and human-computer interaction (HCI).Technical innovations to support automatic intelligent dispatching are discussed and organised in three levels, i.e. perception, comprehension and projection. The breakthroughs are: construction of index system, trajectory recognition based on massive information and knowledge mining, trajectory projection taking into accounts the uncertainties, online risk assessment and early warning, power grid intelligent decision-making support, automatic coordination of grid operation control, online assessment, natural human-computer interaction mode, and etc ... These are the future research areas of automatic intelligent dispatching.

Power grid optimal topology dispatch and early fault warning using GAT and HDRL

A Graph-based Deep Reinforcement Learning Framework for Autonomous Power Dispatch on Power Systems with Changing Topologies

A multi-hierarchical interpretable method for DRL-based dispatching control in power systems

Fault Detection for agents on power grid topology optimization: A Comprehensive analysis

Adaptive Power System Emergency Control using Deep Reinforcement Learning

Reinforcement Learning and Stochastic Optimization with Deep Learning-Based Forecasting on Power Grid Scheduling

Optimizing Load Scheduling in Power Grids Using Reinforcement Learning and Markov Decision Processes

Day-Ahead Optimal Dispatch in Active Distribution Network Based on Deep Reinforcement Learning with Improved Feature Extraction Network

Hierarchical Task Planning for Power Line Flow Regulation

Reinforcement Learning for Resilient Power Grids

SmartGridDispatchPowered byDeepLearning: ASurvey

Optimal Operation Strategy for Distribution Grid with High Penetration Renewable Energy: a Novel Deep Reinforcement Learning Based Approach

Hierarchical deep reinforcement learning for self-adaptive economic dispatch

Deep Reinforcement Learning for Optimal Power Flow with Renewables Using Graph Information

Distributed Hierarchical Deep Reinforcement Learning for Large-Scale Grid Emergency Control

A Deep Reinforcement Learning Based Approach for Optimal Active Power Dispatch

A deep reinforcement learning algorithm for the order optimization allocation of total power in the interconnected power grids

Multi-Agent Graph-Attention Deep Reinforcement Learning for Post-Contingency Grid Emergency Voltage Control

Smart Grid Dispatch Powered by Deep Learning: a Survey

Learning and Fast Adaptation for Grid Emergency Control via Deep Meta Reinforcement Learning

Architecture and key technologies of intelligent dispatching based on situational awareness