Complex Network Construction of Univariate Chaotic Time Series Based on Maximum Mean Discrepancy

A multi-factor combination prediction model of carbon emissions based on improved CEEMDAN

As the global greenhouse effect intensifies, carbon emissions are gradually becoming a hot topic of discussion. Accurate carbon emissions prediction is an important foundation to realize carbon neutrality and peak carbon dioxide emissions. To accurately predict carbon emissions, a multi-factor combination prediction model based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), bidirectional long short-term memory optimized by lemurs optimizer (LOBiLSTM) and least squares support vector machine optimized by lemurs optimizer (LOLSSVM), named ICEEMDAN-LOBiLSTM-LOLSSVM, is proposed. Firstly, the influencing factors of carbon emissions are selected by Spearman correlation coefficient, and carbon emissions are decomposed into intrinsic mode functions (IMFs) by ICEEMDAN. Secondly, the influencing factors and IMFs are input into LOBiLSTM and LOLSSVM respectively for prediction. Then, the point prediction results are obtained by weighting the prediction results of LOBiLSTM and LOLSSVM. Finally, probability density function of point prediction error is calculated by kernel density estimation, and the interval prediction results are calculated according to different confidence intervals. Carbon emissions of China and Germany are selected to verify the superiority of ICEEMDAN-LOBiLSTM-LOLSSVM. The experiment shows that RMSE, MAE, MAPE, and R2 of the proposed model are 0.4468, 0.3612, 0.0120, and 0.9839 respectively for China, which is the best among the nine models, as well as for Germany.

Characterizing gas-liquid two-phase flow behavior using complex network and deep learning

Gas-liquid two-phase flow is polymorphic and unstable, and characterizing its flow behavior is a major challenge in the study of multiphase flow. We first conduct dynamic experiments on gas-liquid two-phase flow in a vertical tube and obtain multi-channel signals using a self-designed four-sector distributed conductivity sensor. In order to characterize the evolution of gas-liquid two-phase flow, we transform the obtained signals using the adaptive optimal kernel time-frequency representation and build a complex network based on the time-frequency energy distribution. As quantitative indicators, global clustering coefficients of the complex network at various sparsity levels are computed to analyze the dynamic behavior of various flow structures. The results demonstrate that the proposed approach enables effective analysis of multi-channel measurement information for revealing the evolutionary mechanisms of gas-liquid two-phase flow. Furthermore, for the purpose of flow structure recognition, we propose a temporal-spatio convolutional neural network and achieve a classification accuracy of 95.83%.