Uncertain Scheduling Based On The Markov Chains-Multiple Scenarios Technique In Active Distribution Network

With more and more renewable resources and controllable units integrating to the active distribution network,it's topology and scheduling mode become more complicated.With the rapid development of power electronic technology and new energy technology,AC/DC hybrid active distribution network contained flexible DC device is a development trend.Through the flexible DC device,the power exchange is achieved between AC network and DC network,which adjuste the controllable resource in a wider range,and improve the consumptive ability of the renewable resource.In this context,the paper aimed at maximum consumption of the renewable resources and optimal dispatch of several kinds of the controllable units,taking a research on the dispatch of the active distribution network.Considering the uncertainty and the volatility of the renewable resource,the paper proposed a multi time scale scheduling stratagy that combined markov chain and multiple senarios technique.The temporal correlation of the errors changes is considered trough markov chains to guarantee the availability of original scenario set.A large number of scenarios are clustered by fuzzy C-means algorithm to gain representative scenarios.In the long time scale,the typical scenarios are employed to take uncertainty into account,and the outputs of the connecting line and flexible DC devices are optimized with the goal of minimizing the total cost of the power grid.There are also reference results of other controllable units in the long time scale to transmit to the short time scale.In the short-time scale,the output of adjustable resources is corrected on the basis of long-time scale results to minimize the difference,while the results of the connecting line and flexible DC remain unchanged.The paper takes active distribution network involved three terminal flexible DC device as an example to simulate.According to the clustering validity and similarity degree,the representative scenarioes are obtained.Through Matlab program,the results verify that multiple scenarios technique with the Markov chain can describe the uncertainty of original problem effectively and correct the prediction deviation which increases over time.It is also beneficial to reducing the pressure of short-time scale dispatch.The paper also illustrates that the model is contributed to deal with the uncertain fluctuations and improve the consumptive ability of intermittent resource.