Research On Self-healing Strategy For Smart Distribution Network With DG

Compared with traditional distribution network,smart distribution network has the characteristics of reliability and high-quality,which is the development direction of distribution network in the future.Self-healing is the essential feature of smart distribution network,which is significant to the operational stability of distribution network.Fault location and fault recovery are essence means to realize self-healing,but traditional solutions are no longer fully applicable with the large scale access of distributed generator.Therefore,researching fault location method and fault recovery method with the access of distributed generator is very significant to enhance self-healing ability of distribution network.With the access of distributed generator,distribution network turns into a two-end/multi-end power network,as a result,the traditional fault location scheme based on one-way trend no longer applies.For the characteristics of center symmetry of the line parameters due to non-displacement of the distribution network,traditional analytical method such as symmetrical component method cannot realize the decoupling of sequence components.This paper presents a new fault location method for active distribution network based on improved Karrenbauer transformation.Firstly,starting from standard Karrenbauer transformation matrix and through matrix transformation,the improved Karrenbauer matrix which can realize three-phase high-precision decoupling of non-displacement distribution network is obtained.On the basis of high-precision decoupling of three-phase network,through analyzing the current phase angle difference of the 1-mode fault component after the phase-mode transformation before and after the fault,deriving the criteria of sector fault location.Simulation results show that the method can't be influenced by transition resistance and can quickly and accurately locate the fault section in non-displacement distribution network.Fault recovery of distribution network is a large-scale nonlinear programming problem to find the best objectives under several constraints.Firefly algorithm is a new simulating biology intelligent algorithm which converges quickly and has simple algorithm structure.On this basis,this paper proposed a new inertia weight multi-objective firefly algorithm: importing the crowding distance and non-dominated sort of NSGA2,so that the algorithm can solve multi-objective optimization problem;constructing elite records mechanism to decrease the loses of elite;importing inertia weight,through adjusting the numerical value of inertia weight to improve the rate of convergence.This paper uses the inertia weight multi-objective firefly algorithm to optimize the fault recovery path of active distribution network,aimed at reducing the system's active loss and the number of switch operation,and uses the Monte Carlo method to simulate the uncertainty of DG output.At last,the fault recovery method is exerted on 33-node feeder.The results show the effectiveness of the fault recovery strategy.