Autonomous Optimal Operation Of AC/DC Hybrid Distribution Network Based On Power Electronic Transformer

With the rapid growth of distributed renewable energy,large-scale Distributed Renewable Energy is connected to the distribution power grid.At present,there are too many AC-DC converter in the access technology of renewable energy,which reduces the efficiency and affects the coveniences.In addition,insufficient interconnection and inflexible regulation of distribution network also limit the full consumption of DRE.The AC/DC hybrid distribution network is constructed by bidirectional multi-port Power Electronic Transformer,which can integrage DRE at multi AC/DC voltage level,reduce transformation,absorb more DRE in a wider range of interconnected network and improve energy efficiency.In order to realize the full consmption and efficient utilization of DRE,this paper proposes a day-ahead optimal scheduling model with PET,establishes an intra-day rolling optimization model and proposes improved active set algotithm that can quickly solve the model.The PSCAD software is used to simulate the distribution network based on PET.In the day-ahead optimal scheduling,in order to realize the full aborbtion of DRE,a three-port linear steady-state model of PET is constructed.Considering the uncertainty of renewable energy,the linear robust optimal scheduling model with uncertain variables is transformed into an optimal scheduling model with only deterministic optimization variables by using the dual optimization theory,so that the model can be solved efficiently and quickly with the solution software.According to the schduling instruction value from day-ahead shceduling,the control paramters of the droop control nodes in the network are set to achieve autonomous operation.In the intra-day real-time running stage,in order to realize the efficient uilization of DRE and aiming at the problems of low network inertia caused by the access of a large number of power electronic devices,which will greatly affected network by the unbalanced power,based on Model Predictive Control,a rolling optimization model that solve the control increment and the corresponding improved active set algorithm which can solve the model quickly are proposed.By setting different scenarios,the effectiveness of the proposed model and algorithm are verified by comparing the solution results and solution time of the optimization model in the different scenarios.Eventually,aiming at AC power network based on PET,with the intra-day control strategy,the real-time simulation model is established by the PSCAD software.In the case of different output fluctuations of DRE,the real-time simulation of AC power network is realized through PSCAD.The simulation results verify the effectiveness of the intra-day scheduling strategy proposed in this paper.