K-Means Clustering Based Urban Rail Train Operation Condition Identification Method

The configuration of the stationary energy storage system (ESS) can effectively utilize the remaining regenerative braking energy in the traction power supply system of urban rail, thus achieving energy saving and emission reduction. Energy management strategy (EMS), as the top-level control of ESS, is crucially influenced by the train status in its optimal control. The traditional method of reflecting train traction or braking conditions based on the relationship between traction network voltage and substation no-load voltage is difficult to achieve accurate identification due to the differences of no-load voltage between multiple substations. Moreover, the method of directly obtaining train status requires high-performance communication between the train and the ESS. This paper proposes an operation condition identification method based on K-means clustering, which only needs to collect some ground information to achieve online identification of the overall operation conditions of trains in the power supply section. The method proposed in this paper can provide important guidance for the EMS design and optimization of stationary ESS for urban rail transit.