A Method to Inspect the Implementation of Electricity Price Based on Deep Learning Variational Autoencoder
Xiying Gao, Ning Ye, Jinchun Song, Haomiao Wang, Ye Zhang, Yan Guan, Xiaowen Song, Yingkai Cai, Wenshu Zhang, Qian Hui, Dan Li · 2018
In this paper, we propose a method for performing electricity price execution inspection by using a variational autoencoder technology in deep learning. The variational auto encoder based anomaly detection algorithm(VABAD) can be used both as a discriminant model and as a feature of the generation model, which effectively solves the calculation problem of multiple heterogeneous parameters of current electricity price inspection implementation. The reconstruction probability is a probabilistic measure that takes into account the variability of the distribution of variables. It is used by autoencoder based anomaly detection methods. Experimental results show that the proposed method has been validated and compared to the existing approaches. The databases used in this paper come from Power Marketing System that occurred in Liaoning, China in 2015.