Optimal Planning Of Transmission And Distribution Systems With Uncertainties Adaptable

Uncertainties widely exist in modern power systems.From the layer of generation,uncertainties include traditional generation output,renewable energy output,etc.From the layer of network,uncertainties include line contingency,switch contingency,transformer contingency,etc.From the layer of load,uncertainties include time-varying load,demand side management,etc.All these uncertainties compose the risk sources of power system planning and operation,and they increase system economy and techinical risk,which make it hard for existing planning methods to handle these problems.From the perspectives of transmission,distribution and integrated transmission/distribution networks,this paper proposes the planning framework for optimizing transmission and distribution networks under uncertainties,and the main contributions are as follows.(1)An optimal planning model for transmission networks considering correlated scenario clustering is presented.According to stochastic optimization theory,this paper formulates a mixed-integer linear programming problem for transmission network expansion,and considers uncertainties associated with wind turbine generation output,load variation and component contingency.The correlated hourly wind speed and load scenarios in one year are proposed based on scenario analysis.Moreover,an adaptable fuzzy C-means clustering algorithm is adopted to reduce the number of the correlated scenarios.The solved center and number of the clustering method can maintain the characteristics of the original data,and they are introduced into the planning model as an input,which guarantee the secure and economic operation of the planning solution under stochastic scenarios.(2)An optimal planning model for transmission networks considering interruptible load management is presented.According to stochastic programming and robust optimization theory,uncertainty related to component contingency is formulated as a probabilistic model while uncertainties associated with wind turbine generation output and load variation are described as a polyhedral uncertainty set.A DC power flow based stochastic robust planning model for transmission networks is proposed considering interruptible load management measures.This model can be divided into three layers.The upper layer is the new line construction decision problem,the medium layer is the worst operation scenario selection problem,and the lower layer is the optimal operation problem under the worst scenario.The planning scheme solved has the advantages of flexibility and anti-risk ability.(3)An optimal planning model for distribution networks considering network transfer capability is presented.Under the background of the interconnection of substations in the smart distribution grid,this model considers the improvement of total supply capability in the presence of network transfer,which can postpone system upgrade and improve asset utilization.This model takes the minimum of the sum of investment and operation costs as an objective function,and integrates four active management schemes(i.e.,on-load tap-changer voltage control,distributed generation curtailment,distributed generation power factor control,and demand side management).According to decomposition coordination methodology,this model is a two-stage framework.The first stage deliver line reformation,line construction,substation expansion and distributed generation allocation to the second stage.Then,the second stage optimizes the operation and sends the results back to the first stage to guide its decision.The model is handled using improved differential evolution algorithm and primal-dual interior-point method.The planning solution optimizes the coordination of distributed generations and distribution networks,and decreases the planning cost and future operation risk.(4)An optimal planning model for integrated transmission and distribution networks considering renewable energy output regulation is presented.Based on decomposition optimization theory,the planning problem takes the dispatchable renewable energy output into account,and divides into a transmission network planning sub-problem and a series of distribution network planning sub-problems on the premise of the consistence of shared information between transmission and distribution networks.For the distribution network planning sub-problem,the primal problem can be transformed into a convex optimization problem using second-order cone relaxation and big-M method,and the mixed-integer convex programming based optimization model for integrated transmission and distribution networks can be formulated.A distributed optimization algorithm(i.e.,analysis target cascading)is adopted to solve the integrated model,which can guarantee the convergence of iteration.(5)An optimal multi-stage planning model for integrated transmission and distribution networks considering the correlation among random variables is presented.The model is to minimize the sum of the investment and operation costs of integrated transmission and distribution networks,and considers transmission network multi-stage planning constraints,distribution network multi-stage planning constraints and boundary constraints.In detail,the boundary constraints include boundary active power equation,boundary reactive power equation and boundary bus voltage equation.For the transmission network planning sub-problem,the AC power flow constrained model is relaxed using semi-definite programming.Considering the correlation among wind speed,illumination intensity and load,the wind-photovoltaic-load correlated sample matrix generated by quasi Monte Carlo simulation and singular value decomposition can improve the accuracy of the planning results.The transmission and distribution planning solutions in different stages can guarantee the penetration of centralized and distributed renewable energy sources,and satisfy the load growth.