Research On The Collaborative Optimization Method Of Sectionlization And Path Restoration Strategy Using Mixed Integer Linear Programming

Under the background of energy transition,the high proportion of renewable energy and the high proportion of power electronic equipment in the power system make the operation environment of power grid increasingly complex,and the risk of large-scale power system blackout is increasing.For a large power outage network with multiple black start power sources,it can be divided into multiple subsystems to restore the power system in parallel,so as to speed up the restoration of the power system and reduce the social and economic losses caused by load outage.Existing studies have proposed many sectionalizing methods for power outage systems,including GN algorithm,ordered binary decision graph,theoretical spectrum clustering algorithm and other methods.However,existing methods mainly focus on the internal topology characteristics of power grid,and seldom consider the influence of partitioning results on subsequent restoration.Some scholars have considered collaborative optimization of sectionalizing and restoration path of outage system,and established a nonlinear two-stage optimization model to make the results of sectionalizing more conducive to subsequent generation and path restoration,so as to shorten the restoration time of the whole system and improve recovery efficiency.However,the nonlinear model is difficult to solve,the solving efficiency is low,and the optimality of the solution cannot be guaranteed.Therefore,this paper proposes a co-optimization method of sectionlization and path restoration strategy using mixed integer linear programming,which mainly completes the following work:Firstly,a nonlinear sectionalizing optimization method considering the restoration path optimization is established to construct the power undirected graph of the power failure system network considering the topological structure of the system after power failure,and the preliminary sectionalizing results are obtained based on the spectral clustering algorithm.The artificial bee colony algorithm is used to solve the path recovery scheme of each sectionalizing.Finally,based on the path restoration method within the sectionalizing,rough set theory was used to modify the sectionalizing situation,and the optimal sectionalizing and path restoration scheme was obtained repeatedly.An example of IEEE 39-node system was taken to verify the validity of the model.Second,the linear optimization model of power system parallel restoration based on network flow theory sectionalizing is established.Then,the linear modeling method is used to model the generator restoration sequence problem and the sectionalizing problem as a whole linear model,considering the path restoration time limit,adding the starting safety constraint to ensure that the generator start time is longer than the node start time.Finally,the model is solved by commercial CPLEX and simulated in IEEE 10-generator 39-node system.Thirdly,a collaborative optimization method of sectionlization and path restoration strategy using mixed integer linear programming is proposed.First,a linear sectionalizing optimization model considering the restoration path is established based on virtual power-on agent method.Then,the coupling relationship between the linear sectionalizing model and the restoration sequence optimization of the generator is found considering the recovery time of the generator.Finally,the nonlinear variables in the cooperative optimization model are dealt with,and the cooperative optimization model of sectionalizing and restoration path of power outage system based on mixed integer linear programming is established.The model is solved by commercial CPLEX,and simulated in IEEE 10-generator 39-node system.Fourthly,the method proposed in this paper is applied to the actual system,and the actual system in a certain area of Jiangsu Province is taken as an example to solve the optimal sectionalizing and optimal restoration path,and a comparative analysis is made with the previous two methods to verify the effectiveness of the model and method proposed in this paper in the actual system.