A Method of Constructing Power System Dispatcher Personas Based on Convolutional Neural Network
Hujun Shi, Yunju Zhang · Journal of Physics Conference Series · 2022
Abstract The recommendation algorithm has been applied in many fields and achieved pretty results in recent years. The personas for a large number of dispatchers in the power system are an indispensable step in the process of intelligent recommendation. In order to recommend dispatchers of different levels according to different operation scenarios, the aim of dispatcher personas is the classification of business level. In this paper, we focused on constructing dispatcher personas of Chinese power system. Unlike of the previous research work, we used convolutional neural network(CNN) to create dispatcher personas. The contributions of this paper are reflected in the following aspects: 1. dataset were built automatically based on internal data; 2. a CNN text classifier were trained to construct the dispatcher personas; 3. the effectiveness of the model is proved by experiments on real data. The experimental results show that the method proposed in the paper is superior to the baseline method and the average accuracy is 96%.