Aiming at the problem that the information distribution of power trading platform is chaotic,resulting in great challenges in the processing and application of power sale information,a research on integrated clustering algorithm of power sale information of power trading platform based on improved CNN is proposed.Based on the in-depth analysis of the supply process and complexity of electricity information sold on the power trading platform,this paper constructs an integrated model of electricity information sold to test the consistency of electricity information sold.The Convolutional Neural Network model(CNN)is improved by Adam algorithm to speed up the learning and training speed of CNN.Based on the constructed information model,the improved CNN is applied to determine the information clustering center and the optimal distance threshold of the selling point,so as to realize the integrated clustering of the selling information of the power trading platform.The experimental data show that the clustering accuracy,normalization degree and adjusted Rand index obtained by the proposed algorithm are greater than those of the comparison algorithm,which fully proves that the proposed algorithm has better clustering effect.
关键词
改进CNN/电力交易平台/发售电信息/聚类/一体化
Key words
improve CNN/power trading platform/selling power information/clustering/integrated
维普
万方数据