Abstract:Closed-loop load transfer is an effective measure to improve the power supply reliability of radial distribution networks. However, the traditional closed-loop load transfer mode is passive for operators: the loop closing operation relies on complex network models and calculation methods, which is feasible for the distribution network with many sensors, but it is not practical for the low-voltage distribution network with a complex structure. The field operation is still based on the experience of operators, resulting in uncertainty of the loop closing and large loop-closing current. Besides, the coordination of multiple power sources for guaranteeing power supply can not guarantee the power quality and the economic cost is high, in which the adopted closed-loop device is not movable and has poor reusability. This article presents an active loop-closing mode: a seamless closed-loop load transfer system (SLTS), which can not only realize uninterrupted load transfer but also eliminate loop closing surge current and expand the application range of closed-loop load transfer. In the proposed system configuration, a series compensation voltage source is connected to two substations. Therefore, closed-loop current can be controlled flexibly by regulating this voltage source. The proposed system can be applied to adjacent substations to carry out closed-loop load transfer operations during equipment maintenance. Besides, it can be installed between two radial feeders to achieve seamless fault seal-healing. The control strategy of seamless closed-loop load transfer and fault seamless self-healing are proposed, respectively. The effectiveness of the proposed SLTS and the designed control strategy are validated via MATLABSimulink.

A Novel Load Transfer Method for Distribution Systems Based on Reinforcement Learning

Multi-agent Deep Reinforcement Learning Algorithm for Distributed Economic Dispatch in Smart Grid.

Multi-Agent Reinforcement Learning-based Distributed Economic Dispatch Considering Network attacks and Uncertain Costs

Multi-agent Deep Reinforcement Learning for Distributed Load Restoration

Reinforcement Learning-Based Net Load Volatility Control of Active Distribution Power Networks

A Novel Method for Seamless Closed-Loop Load Transfer in Radial Distribution Networks

Deep Reinforcement Learning Based Charging Scheduling for Household Electric Vehicles in Active Distribution Network

DDPG-based Reactive Power Optimization Strategy for Active Distribution Network

Curriculum-based Reinforcement Learning for Distribution System Critical Load Restoration

Load Shedding Scheme with Deep Reinforcement Learning to Improve Short-term Voltage Stability

Deep Reinforcement Learning for Load Shedding Against Short-Term Voltage Instability in Large Power Systems

Deep Reinforcement Leaming for Short-term Voltage Control by Dynamic Load Shedding in China Southem Power Grid

Deep Q-Learning-based Distribution Network Reconfiguration for Reliability Improvement

Deep-Learning-Aided Voltage-Stability-Enhancing Stochastic Distribution Network Reconfiguration

Direct Thermal Load Control in Active Distribution Networks Based on Deep Reinforcement Learning

Electric Vehicles Management in Distribution Network: A Data-Efficient Bi-level Safe Deep Reinforcement Learning Method

Voltage Control of Distribution Grid with District Cooling Systems Based on Scenario-Classified Reinforcement Learning

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

Development and Application of Load Transfer Auxiliary Decision-making System in Distribution Grid

Many-Objective Distribution Network Reconfiguration via Deep Reinforcement Learning Assisted Optimization Algorithm

Safe Reinforcement Learning for Emergency LoadShedding of Power Systems