Multi-scenario Capacity Optimization Design of Railway Power Conditioner Considering Load Uncertainty

With the development of power scale of railway traction load, power quality problems of traction power supply system have also been widely concerned. Railway Power Conditioner (RPC) can comprehensively solve power quality problems and has become a current hot research topic. How to reduce the RPC capacity while satisfying the power quality simultaneously is the urgent problem to be settled. The existing method is to set a fixed load value when calculating the compensation capacity in the optimization algorithm, while the randomness and uncertainty of real load fluctuations were not considered. It will probably cause a bad compensation consequence with an insufficient compensation capacity or a poor cost performance with a huge compensation capacity. To deal with this problem, a Multi-scenario Analysis method considering the uncertainty of traction load is introduced while calculating the RPC capacity in this paper. Optimization algorithm combined with K-means Clustering Algorithm is used to cluster the sampled massive data to achieve several operational scenarios and to calculate optimization compensation capacity of each scenario, and finally, the compensation capacity suitable for every scenario is obtained. This method can improve the economy of RPC while meeting the compensation requirements. The method is calculated and verified in MATLAB and further comparison and analysis have made, results show the feasibility and validity of the method used in this paper.