Research On Two-stage Cloud-side Cooperative Regulation Method For Active Distribution Network

Driven by the "double carbon" goal,the power grid needs to speed up to improve the efficiency of energy use,increase the proportion of non-fossil energy consumption,clean and renewable solar,wind and other distributed energy in the grid has been rapidly developed.However,the uncertainty of new energy output,weak controllability,weak support,poor support for securing power supply during peak load hours,insufficient proportion of consumption during centralized grid connection,and its large-scale access brings complex challenges to the real-time balance of the power system.Large-scale controllable load with demand response capability actively participates in the daily control of the power grid,and the active distribution network can coordinate the mobilization of demand-side controllable resources and regulate the consumption behavior of controllable load to minimize the impact of peak power consumption tension and low valley power consumption shortage on the safety of the power grid,and at the same time help promote the energy consumption of wind and light after they are connected to the grid.With the development of electric power IOT technology,the research on the application of cloud-side cooperative computing in active distribution network is emerging.In this paper,we take advantage of cloud-side collaborative computing to conduct research on the optimal regulation methods for active distribution grids,as follows:First,considering the differences in advance notice time of interruptible load,transferable load and levelable load,they are further classified into day-ahead type and intra-day type,and a load regulation model is established.The structure composition and functions of active distribution network cloud and edge are studied,and the overall architecture of active distribution network cloud-edge cooperation including central cloud,feeder edge and station area edge is proposed.Based on the fit between the alternating direction multiplier method and cloud-side cooperation,a cloud-side cooperative control calculation system based on synchronous alternating direction multiplier method is established.Secondly,we propose a two-stage cloud-side cooperative regulation method for active distribution networks,and establish a day-ahead regulation model and an intra-day correction model that consider the intra-day controllable load.By making the intraday controllable load forecast value participate in the optimization together with the intraday controllable load and energy storage resources in the day-ahead regulation stage,and correcting the day-ahead regulation result according to the actual declaration of intraday controllable load in the intraday regulation stage,the peak-shaving and valley-filling performance of the optimized resources can be improved.It can not only realize the effective connection between day-ahead and intra-day regulation strategies,but also ensure the global optimal regulation.Simulation analysis of the four-feeder distribution system using synchronous alternating direction multiplier method verifies the effectiveness and economy of the proposed method.Finally,a two-stage cloud-side cooperative active-reactive joint scheduling strategy is proposed for the active distribution network,and a day-ahead joint scheduling model and an intra-day joint scheduling model are established.The joint active-reactive power dispatching strategy takes into account the effects of static reactive power compensation devices,distributed photovoltaic and energy storage on reactive power and voltage,enables all resources to participate in the day-ahead stage,and corrects the day-ahead active-reactive power dispatching results in the intra-day stage to fully mobilize demand-side controllable resources to achieve peak and valley reduction and high proportional consumption of new energy.Simulation analysis is performed using the improved IEEE33 node system,and the results verify the effectiveness of the proposed strategy.