Flexible Reactive Power Resources Coordination And Optimization In Transmission And Distribution Networks

There is an increasing number of countries identifying the availability and potential of renewable and eco-friendly resources.With high penetration of distributed generators(DG)in distribution networks,as well as the construction of the UHV projects between cross-regional power grids,a part of large power plants in transmission networks will decommission in order to avoid reversed power flow at their transmission-distribution network interfaces.However,the reduction of generator-based reactive power resources may lead to considerable investment costs by transmission system operators(TSO)due to the necessity of installing additional reactive power compensation devices.Therefore,TSOs are looking for another way to obtain controllable reactive power without extra investment expenses and one of that is the reactive power potential from distribution networks.Some resources,such as photovoltaic(PV),wind turbines(WTs),battery storage systems(BSSs)and electric vehicles,have the capability to generate/absorb reactive power with their inverters.Therefore,the benefit of reactive power supporting by distribution networks has been recognized by researchers and many European countries have already legislated documents that admit distribution system operators(DSOs)to participate in reactive power markets by providing reactive power support.The purpose of this work is to fully activate reactive power potential from distribution networks and formulate reactive power resources coordination scheme in transmission networks.In this dissertation,the main research work can be summarized as the following three aspects:Firstly,serving as the most abundant and widely used generation resource in distribution networks,their total installed capacity determines the overall power supply level of distribution networks.However,the active power output of PV is greatly affected by transient weather conditions(e.g.cloud movement).The volatility and uncertainty of PVs pose challenges to the system operators while making dispatchable plans.Therefore,before estimating the reactive power potential of distribution networks,it is of great importance to predict the volatility level of PV's active output.In this dissertation,a hidden Markov model is established to predict the degree of active power output fluctuation of PVs.After that,we could obtain the predictable reactive power capacity of PV.Meanwhile,there are diversity of reactive power resources in distribution networks and their control modes are different from each other.In order to provide a theoretical reference to the power grid managers while dispatching these resources,this dissertation establish the mathematical models of other types of reactive power resources.Secondly,based on the existing active operating points of generators and power flow constraints,this dissertation proposes a method to quickly.and accurately assess the reactive power potential of distribution networks.The optimization model uses the upper and lower limits of reactive power on TSO-DSO surface as objective functions.In order to quickly calculate the reactive capacity on the point of TSO-DSO common coupling with the adjustable range of control variables,the optimization model is transformed into a interval power flow model,and the improved Krawczyk interval iterative algorithm is used to solve this problem.The simulation result is compared with the direct optimization algorithm to verify the calculation accurancy.The analysis of factors in simulation such as network load level,reactive capacity,distribution network location,voltage constraint range,etc.shows an ideal condition to maximize the reactive power flexibility of distribution network.In addition,the active-reactive regulation curve on the point of TSO-DSO common coupling is put forward to coordinate continuously and discretely regulated reactive resources,which enables distribution network to exert the reactive power potential to a greater extent.Finally,the existing study about TSO-DSO reactive power coordinated optimization mainly focuses on the rationality and feasibility of reactive power exchange between TSO-DSO surfaces,but pays less attention to multi reactive resources(incl.reactive power potential of distribution network)coordination in transmission network.Therefore,this dissertation begins with the definition of reactive power regulation performance indexes in transmission network.By sorting their importance degrees,we could clarify the dispatching sequence of resources during reactive power management.A multi-reactive resource coordinated optimization model in transmission network is established,with the goal of reducing network loss,limiting voltage deviation and optimizing reactive power outputs.Generally,in the study of reactive power optimization problems,the active and reactive operating points cannot be easily decoupled,hence,dealing with a reactive power optimization model becomes complex.In this dissertation,we use Taylor's second-order expansion to simplify the nonlinear constraints in the model,lowering the computational complexity while maintaining the accuracy of the results.For the linearized multi-objective optimization model,the Pareto solution frontier is found using the normal boundary intersection method and a compromise solution is given.