Demand Factor Forecasting Method of Design for Communication and Signal Building Electric Systems by Kernel Density Estimation

Abstract

Abstract With the expanding high-speed railway construction scale, the loads along the railway are gradually transformed into electrification, and capacity level of power distribution system of high-speed railway is increasing. The power system for communication and signal loads along the railway is one of the important parts of guaranteeing the reliable operation of the high-speed railway. However, at present, there is no unified standard for transformer capacity design in the electrical design of communication signal building., resulting in the theoretical value of transformer capacity design inconsistent with the actual engineering requirement. To address the problems and formulate a more accurate design standard for the communication and signal building, the load demand factor forecasting of communication and signal building based on based on kernel density estimation with optimal window width is proposed in this paper. Firstly, the topology of the high-speed railway power system is introduced and the principle of demand factor is proposed. Then, the kernel function selection and the optimal window width of the algorithm is proposed to adapt to the estimate the demand factor based on massive samples. Lastly, the experiment is conducted in the trial operating-high-speed railway, the correctness of the forecasting demand factor is verified. And the results are applicable as the reference for electric design of the power loads in high-speed railway in the future.