Guoqing Wang

Increasing the proportion of renewable energy sources, such as photovoltaic power, in the grid can reduce fossil fuel consumption and build a low-carbon power system. However, the inherent instability of the photovoltaic power output makes it difficult to predict, thus increasing the cost of grid operation. Therefore, to improve the accuracy of power prediction and promote the development of the grid, a four-stage short-term photovoltaic power prediction framework, namely, CDT, is proposed, which includes decomposition, classification, reconstruction and forecasting. The initial power data are decomposed using complete ensemble empirical mode decomposition with adaptive noise. Next, an improved data classification and reconstruction method based on dynamic time warping is developed to process the data, which reduces the dimensionality of the data while preserving trend information. Finally, the reconstructed components are predicted using the improved TCN model. The results of the empirical study show that the proposed CDT has higher precision and scalability in processing and predicting the trend of photovoltaic power generation, compared to the other benchmark models.

With the increasing penetration of renewable energy, the efficient dispatch of integrated energy systems (IESs) is facing severe challenges. Addressing the uncertainty of renewable energy output and designing efficient market mechanisms are crucial for achieving economical and low-carbon operation of IES. To this end, this paper unveils a comprehensive modeling and optimization framework: Firstly, a Spatio-Temporal Diffusion Model (STDM) is proposed, which generates high-quality wind power forecasting data by accurately capturing its spatio-temporal correlations, thereby providing reliable input for IES dispatch. Subsequently, a stochastic optimal scheduling model for electricity–heat–carbon coupled IES is established, comprehensively considering carbon capture equipment and a carbon quota mechanism. Finally, a multi-IES Nash bargaining cooperative game model is developed, encompassing bilateral energy trading and bilateral carbon trading, to equitably distribute cooperative benefits. Simulation results demonstrate that the STDM model significantly outperforms baseline models in both forecasting accuracy and scenario quality, while the designed bilateral market mechanism enhances system economics by reducing the total operating cost by 19.63% and lowering the total carbon emissions by 4.09%.

Finding a stable community in dynamically developing networks is crucial. In this study, we introduce an innovative concept of resilience, originally rooted in infrastructure networks, and merge it with the distinctive features of k-core graphs to define a novel community structure in dynamic networks-referred to as the k-block. Leveraging the Independent Cascade model, we evaluate communities based on their capacity to sustain activity under adversarial conditions, where an attack is characterized by the blocking of nodes responsible for information propagation. The proposed k-block demonstrates exceptional robustness against such disruptions. We develop a non-submodular resilience block identification model to identify these resilient communities. Addressing the computational complexity inherent in this task, we propose a groundbreaking two-stage greedy algorithm with a time complexity lower than $O(V \cdot E)$ . To validate the algorithm’s efficacy, we conduct extensive experiments on four real-world datasets. The results are meticulously analyzed to evaluate both the algorithm’s performance and its practical applicability. Our findings not only underscore the algorithm’s efficiency but also highlight its potential for real-world deployment in dynamic network analysis.

Personal cognition, product advertising and social recommendations are important factors to develop brand preferences, which directly influence consumer purchasing behaviors. A diversified brand preference landscape is conductive to preventing market monopolization, thus promoting healthier market competition. The entry of a new brand enhances market diversity and contributes to equilibrium of consumers' brand preferences. The new entrant stimulates competitive responses from incumbent brands because of catfish effect. Then market shares undergo redistribution as all brands increase their operational vitality in response to the competitive pressure. To select $k$ users in a social network as advertisers of a new brand, this paper proposes the information equilibrium maximization (IEM) problem, and proves that the IEM is NP-hard, computing the objective function is #P-hard, and the objective function is neither modular nor monotonic. Then the entropy-based equilibrium degree maximization (EEDM) algorithm is proposed. In experiments, based on three methods of selecting seed nodes, E_qedm shows its superiority. It has strong robustness when the size of seedsets is large enough, and activation probability and update probability are more than 0.5. Besides, the number of initial preferences has little influence on the performance of E_qedm.

The long-term production capacity planning of semiconductor supply chain manufacturers has a series of characteristics, such as large capital investment, fast technology upgrading, long lead-time of manufacturing equipment, and unstable market environment, which leads to the uncertainty of demand. In addition, blockchain technology is widely used to build consortium chains for information sharing across upstream and downstream enterprises, thus having the potential ability to help finance semiconductor manufacturers. This paper combines two uncertainty-oriented methods (stochastic programming and robust optimization) to examine the conversion of capacity between different product types and construct a two-stage mathematical planning model maximizing the net profit of manufacturers. Through the introduction of blockchain technology and information sharing among enterprises, we improve the effectiveness of our model to realize the optimal allocation of long-term capacity planning. Finally, we reformulate the model into a tractable MINLP, construct numerical examples to verify the solvability, and carry out sensitivity analysis.

With the in-depth development of intelligent media technology, online and offline fusion, reality and virtual entanglement, information content generalization, the boundary between positive and negative information is blurred, all kinds of misinformation in the social network fission spread, and cyberspace governance has become a global consensus. In this article, we comprehensively consider the spread of misinformation in location-based interpersonal social network and online social network, and systematically tackle the novel problem of minimizing the influence of misinformation under individual protection strategies. We first analyze the complexity and modularity of the problem. Then, we leverage the Lovász extension to devise a nonsubmodular set function continuity approximate convex relaxation method, and develop an approximate projected subgradient procedure to obtain a solution with a factor approximate guarantee. Finally, experiments on three assembled real-world datasets demonstrate the effectiveness and feasibility of our designed method and developed the algorithm.