This paper proposes an Prediction of Multivariate Spatial-Temporal Series Data Based on Adaptive Spatial-Temporal Information(ASTCN), which learns complex spatio-temporal information from multivariate spatio-temporal series through trainable temporal embeddings and graph adjacency matrices. A gated fusion mechanism is employed to control the proportion of different temporal embeddings to improve prediction accuracy. Visualization of the temporal embeddings reveals that weekly temporal embeddings have the greatest impact on prediction accuracy, followed by daily temporal embeddings, while monthly temporal embeddings have the least impact. Additionally, a novel method for constructing graph adjacency matrices is introduced. Ablation experiments demonstrate that the two types of graph adjacency matrices proposed in this method have varying degrees of influence on improving the prediction accuracy of the dataset. Consequently, this paper integrates the two graph adjacency matrices, enabling ASTCN to achieve superior prediction accuracy on traffic speed, traffic flow, and air quality datasets compared to when either matrix is used alone. In comparative experiments, ASTCN ultimately achieves excellent prediction performance with relatively low training costs.

Maintaining the reasonable distribution of load current while regulating the bus voltage balance among them is a challenging control issue in DC microgrids. Many cases have proposed a variety of schemes to realize load current sharing and voltage balance for DC microgrid systems with single load or constant current load, but few have considered the problem of voltage regulation and load distribution in the face of multiple distributed generation units and non-constant current loads. For a type of DC microgrid system with n groups of distributed power generation units and non-constant current load units, a new strategy for indirectly adjusting the bus voltage by compensating the reference voltage of the converter is proposed. This strategy realizes load current sharing and bus voltage balance by using the proportion of micro-source output in the total load and the regulation of bus voltage feedback signal, so as to improve the accuracy of load distribution and the ability of voltage regulation. The Hurwitz stability criterion proves theoretically that the proposed control strategy can stabilize the system. Finally, the simulation and experimental analysis of the DC microgrid under different conditions are carried out through MATLAB and the intelligent microgrid experimental platform to verify the effectiveness of the proposed control scheme.

This paper investigates cooperative adaptive cruise control (CACC) of vehicles, focusing on the effect of medium access control (MAC) protocol and the inaccurate acceleration measurement. A Markov chain is used to describe the randomness in vehicular network access under a MAC protocol; a reduced-order observer is proposed to estimate the relative acceleration of neighboring vehicles. Based on stochastic system techniques, a series of sufficient conditions are given in the form of backward recursive Riccati Difference Equations (RDE) to guarantee the stable tracking error. Numerical simulations are given to verify the effectiveness of the proposed approach.