Multi-time Scale Optimal Dispatch In AC/DC Hybrid Distribution Networks Based On SOCP

In order to deal with the depletion of fossil fuels and the pollution of environment,countries around the world are focusing on technologies like renewable energy resources and electric vehicles.However,most of the renewable energy resources generate DC power and are connected to the network as distributed generations(DG)meaning that they cannot be connected to the AC network without a converter.Besides,an increasing number of DC loads,such as electric vehicles also need converters to plug in the AC distribution network.Hence constructing DC distribution networks,which saves the cost of converter construction,has gained much more attention,resulting in the appearance of a hybrid AC/DC topology configuration.Considering uncertainty analysis and optimal scheduling strategy,this paper does a research on AC/DC hybrid distribution system based on second-order corn programming(SOCP),opportunity constraints and voltage sensitivity.The main work of this paper can be summarized as follows(1)Based on the SOCP theory,a convex optimization model of AC/DC hybrid distribution system is established.Firstly,Distflow model is used to establish the AC/DC hybrid power flow equation.Then SOCP is used to convert power flow constraints into the second-order cone type.The nonlinear equality constraint in VSC converter is transformed into a linear inequality constraint by formula deriving.Thus the nonlinear non-convex program is convexed into a linear program,which can guarantee that the optimized result is the global optimal solution.Besides,the load model has been carefully analyzed to implement conservation voltage reduction(CVR).Tests on IEEE 33-bus test system demonstrate the effectiveness of the proposed method by using Cplex toolbox through MATLAB(2)Considering the uncertainty and correlation of the DGs' outputs,a day-ahead scheduling model is established based on opportunity constraint programming and Nataf transformation.The uncertainty of wind power is described with opportunity constraint.Confidence coefficient is optimized to get the economy optimum.The opportunity constraint is dealt with big M method and Monte Carlo simulation.Then the correlation between DG outputs is considered by Nataf transformation.After that the day-ahead scheduling model is proposed to minimize both the active power loss and electricity purchasing cost by coordinating active and reactive power outputs of voltage source converters(VSCs),energy storage systems(ESSs)and static var compensations(SVCs),as well as the voltage regulation capability of on-load tap-changer(OLTC).The IEEE 33-bus test system is employed to demonstrate the validity of the proposed method.(3)A day rolling correction model of AC/DC hybrid distribution system is established based on the voltage sensitivity.Considering the load characteristics,an analytical derivation of the sensitivities of node voltages to the nodal power injections and transformers tap-changers positions is provided.After that the deduced voltage sensitivity is used as the prediction model of model predictive control(MPC)for the day rolling optimization.The dispatch schedule of OLTC and capacitor banks are determined in the day-ahead scheduling model,while the outputs of SVCs,CBs,ESSs and VSCs are determined in the day rolling optimization.Thus the multi-time scale optimization scheduling is established.Results from the IEEE 33-bus test system demonstrate the effectiveness of the model.