Study On PMU Based Methodologies Of Topology Perception And Section Power Flow Coordinated Control Of Power Networks

In the past two decades,significant changes have occurred in modern power networks.Like,the structure and operation performances of the modern power system are distinctively different from those of traditional power system with the implementation of ultra-high voltage long-distance transmission and AC/DC hybrid transmission,the introduction of large amounts of renewable energy systems,distributed energy resources(DER),distributed generation(DG)and energy storage devices,and the rapid development of electric vehicles,more frequent interactions between users and grid.Therefore,refined monitoring,detection and perception,optimization dispatching and control,have become the major concerns of modern power networks worldwide.Undoubtedly,the solutions to these problems rely on the high-precision data acquisition system.At present,high-precision,high sampling rate,real-time synchronized phasor measurement units(PMUs)data has become an important data source for power grid analysis and calculation,fault detection and location,dynamic process monitoring and analysis,stability and coordination control,etc.Naturally,it is also a key support to solve the aforementioned hot topics.PMU has already extensively been installed in the substations of the transmission system and innovative research on it keeps emerging.As the distribution network being paid more attention,the research on the application of PMU at the distribution network level has become a new hot topic,and the large-scale deployment of distribution-level phasor measurement unit(D-PMU)for the distribution network is underway.Concurrently,the development and promotion of advanced technologies such as advanced metering infrastructure(AMI),artificial intelligence,and 5th generation mobile networks(5G)provide new opportunities for the in-depth application of PMU synchronized data both in the transmission and distribution networks,and it is expected to solve some intractable and imperative issues.In light of the above background and opportunities,this thesis focuses on the urgent demands of refined topology perception and sectional power flow coordinated control,and revolves research on the in-depth application of high-precision PMU synchronous measurement data from the connection topology detection of multiple components to the parameter estimation of a single component(such as a power line),and then from the large-scale coordinated section power flow control at the transmission networks to the small-scale three-phase unbalance section power flow coordinated control at the distribution networks.Moreover,the research will focus on the artificial intelligence method for power network topology detection based on PMU,the data fusion method for power line parameter estimation based on PMU and AMI,the coordinated control method of distributed power flow controller(DPFC)based on PMU and the coordinated control method of three-phase unbalance mitigation unit based on PMU.It is expected to provide new ideas and methodologies for the in-depth application of PMU synchronous measurement data in power grid topology perception and section power flow coordinated control.The major works and achievements of this thesis are summarized as follow:(1)In the aspect of detecting network topology,the concept of "kernel-node-map" is proposed and conducted by integrating the structure features of neural networks with the physical topology of the distribution network.We focus on the challenge of topology detection caused by frequent topology change and multiple light-load branches.In consideration of local connectivity characteristics of power networks,a deep convolution neural network(CNN)is introduced based on the concept of "kernel-node-map" for topology detection,namely,the KNDN(kernel-node-map deep networks)method.The idea is to utilize a neural network to formulate the topology detection problem as a mapping process from the measured data to the distribution network topology.It has excellent local sensing performance and is suitable for radial,weakly looped,and loopy power networks.(2)In the aspect of power line parameter estimation,the shorter length and smaller valued parameter of distribution line,and the limitation of synchronous measurement devices pose more challenges for precise line parameter estimation in the distribution networks than in the transmission networks.To tackle this issue,we present a distribution line parameter estimation methodology based on the data fusion of D-PMU and AMI data.Considering time synchronous error(TSE)of smart meters,a data alignment algorithm between D-PMU and AMI is proposed to obtain TSE data set.To solve the problem of the unavailable prior distribution of TSE in smart meters,a nonparametric estimation method based on a Gaussian kernel is introduced to estimate the probability model of TSE.Moreover,the“acceptance-rejection” sampling method can be used to generate TSE data of other meters with the assumption that all smart meters in the same local area follow the same probability model.Thereby,based on the voltage drop optimization model,the study has calculated the estimated value of line resistance and reactance of the distribution system for multiple runs by the Monte Carlo method.Eventually,a satisfactory level of accuracy is obtained.(3)In the aspect of DPFC based sectional power flow coordinated control of the transmission network,to solve the contradiction between the fast response requirements of DPFC series units and the available communication link bandwidth limitation,we proposed a DPFC coordinated control architecture based on PMU/WAMS and power line carrier communication and a two-level power flow control strategy.In the DPFC series unit,it is presented to apply the local collection of the transmission line current phase and relevant calculation to track the referenced voltage phase.Additionally,the amplitude command correction and reference phase correction control measures are adopted to achieve accurate control of fundamental frequency compensation voltage,thereby effectively regulate the section power flow of the transmission line.The proposed DPFC coordinated control system architecture does not need high-speed communication links,which provides a novel and practical coordinated control methodology for DPFC engineering applications.(4)Concerning the coordinated unbalance mitigation of three-phase sectional power flow of low-voltage distribution network(LVDN),this paper suggested a novel concept of "Active Asymmetry Energy-Absorbing"(AAEA)of three-phase electrical equipment in the LVDN.Two back-to-back AC/DC converters are utilized to conduct three-phase unbalanced energy-absorbing by three-phase consumers in the LVDN.Thus,the negative and zero sequence components of the three-phase unbalanced current are compensated.The simulation results validate the correctness and effectiveness of this new concept.Based on the above contents,D-PMU and 5G network-based hierarchical coordinated control architecture and control strategy for three-phase unbalance alleviation are then proposed.Multiple AAEA units or with topology modification measures provide coordinated compensation for the three-phase imbalance of the network with dynamic adaptability to network changes.