Research On Fault Diagnosis Method Of Intelligent Distribution Network Considering D-PMU Measurement Information

As power grid upgrading and transformation,the power grid also develops in the direction of larger scale and more complex structure.However,the real-time performance of data communication is poor and the communication architecture is incomplete in the existing distribution network,resulting in the information collection is not perfect,and fault diagnosis of distribution network is easily affected by the type,quantity and collection interval of measurement data.In order to make up for the disadvantages of the existing research,this thesis analyzes the information communication and fault diagnosis in the distribution network based on the existing technical architecture of power Internet of Things,the fault diagnosis and fault location method of distribution network considering D-PMU measurement information is studied.The main detail of the completion is divided into:(1)Based on ubiquitous power Internet of things,the D-PMU intelligent measure device is applied in smart grid to realize real-time communication.The measurement data is collected in distribution system and uploaded to the main station of the system in layers and partitions.It also to meet the data technology foundation for many applications of power distribution network in the future.(2)Aiming at the fault diagnosis problem of distribution network,the compact wavelet neural network diagnosis model is established.The signal after frequency band decomposition is trained and analyzed according to the wavelet transform and neural network model.The genetic algorithm is introduced to improve and adjust the weight of wavelet neural network.On the premise of ensuring the diagnosis accuracy,the convergence speed is improved.Finally,the effectiveness of the proposed method is verified.(3)The fault location method of distribution network is studied based on D-PMU real-time information.Based on load symmetry,fault location methods based on line distribution parameters and fault location methods based on line impedance are proposed respectively.The proposed fault location method has high accuracy based on the measurement accuracy,fault resistance,distance information and T-connection line,it guarantees the safe and stable operation of the distribution network.In addition,it also gives the suggestion of D-PMU configuration considering the problem of fault diagnosis.Finally,the proposed method is verified in the example of IEEE33 node.