In the existing work on the robustness of the hypernetworks, researchers have not considered the effect of internal structure on the robustness of the hypernetworks. Aiming at this problem, this paper proposes a capacity-load model that can describe the relationship between the internal structure and the robustness of the hypernetworks. By simulation experiments, we obtain the robustness of the k-uniform scale-free hypernetwork under three modes: preferential connection, random connection, and completed connection within hyperedges. Analysis of the comparison experiments reveals that the robustness of the scale-free hypernetwork is related to the ways of nodes' connection, the size k of the nodes, and the number of ordinary edges m_k within the hyperedges.The results show that the internal structure of the hyperedges has a large impact on the overall robustness of the scale-free hypernetworks.

The campus is a special crowded place. If a patient with covid-19 appears on campus, there is often the risk of large-scale infection, and it may also become an important source of family transmission. This paper proposes a class-based contact reduction strategy for campus epidemic prevention and control. This strategy can be understood as reducing human contact between and within communities after the campus population is divided into groups according to the class structure. The research results show that, on the premise of reducing the same number of contacts or contact time, differential contact reduction strategies based on community structure can achieve better epidemic prevention and control effects, making the infection peak and the total number of disease-causing population fell further on campus.

This paper proposes a high-low-order coupled network model to gain a deeper understanding of the interdependent relationships in real-world network systems and address the cascade failure issues faced by complex networks and their higher-order networks. This model describes the interdependencies between complex networks (lower-order networks) and their higher-order organizations (higher-order networks). Their vulnerability is analyzed by subjecting the high-low-order coupled networks to random attacks. The study reveals that high-low-order coupled networks exhibit greater vulnerability to random attacks than standalone lower-order networks. This finding underscores the importance of considering the interdependencies between high and low-order networks in designing and managing complex network systems, particularly in preventing cascade failures, where special attention should be paid to the vulnerabilities of these interdependent structures.

To explore the population migration characteristics and evolution patterns of BRI countries, this paper constructs a network based on population migration data from 1960 to 2020 from the perspective of complex systems science, and conducts analysis by extracting backbone networks, community divisions, and application of PageRank importance evaluation method. The results demonstrate that the migration volume of inflows and outflows within BRI countries has tended to be balanced, and the degree of openness is increasing year by year. And the migration process generally has "cohesion" and "aggregation". The influence of most BRI countries in the global network is higher than the world average. This study provides evidence and basis for the population policies of various countries.

To cope with the negative impact of the information cocoon effect in the network public opinion environment, we take the individual′s acceptance of guiding opinion into consideration, propose a moderate opinion guidance strategy based on psychological suggestion, and propose a moderate opinion dynamic model. We analyze the influence of guidance strength and guidance ratio through simulation experiments, and the influence of moderate guidance strategy on the evolution process of public opinion is studied. Then, we verify the stability of the bootstrapping strategy under multiple networks and compares it with traditional degree booting. The results show the superiority of moderate guidance in guiding effect, especially in the long term.

To analyze the interactive dissemination mechanism of multiple information on the same public opinion event in social networks, this article divides the interaction between multiple information into three types: cooperation, independence, and competition. Based on the interaction situation of multiple information dissemination, an information dissemination model is constructed. Research has found that information attributes, information interaction factors, and network topology have different impacts on the dissemination of public opinion multi information, and different individuals have different characteristics of state changes during the interactive dissemination of multi information. The research results can help scholars better understand the mechanism of multi information interaction and dissemination of public opinion in social networks, and provide certain theoretical guidance for maximizing or controlling the dissemination of public opinion information.

To obtain a better topology and identify the key nodes and lines of distribution networks, this paper presents a multi-attribute method for optimizing network structure and identifying key nodes and lines. Firstly, multiple distribution network topologies are obtained by the branch-exchange algorithm. Secondly, the invulnerability indices are constructed based on complex network theory. Then, the improved TOPSIS method is used to find the optimal network. Finally, the importance indices are constructed to identify key nodes and lines. Taking the 33-bus and 69-bus systems as examples, it verified that the optimized distribution network has stronger anti-attack ability through random attack. Random attacks and deliberate attacks are used on the distribution network. The results show that the network parameters significantly decrease under deliberate attacks, which proves that this method can effectively identify the key nodes and lines.

The subprime mortgage crisis in the United States shows that CDS significantly impacts banking systemic risk, but the mechanism of how CDS affects banking systemic risk is still unclear. This paper first constructs a dynamic multi-layer banking network model with CDS interactions to study the dual impact of CDS on the banking system in both volatile and stable economic environments. The results show that when the economy is stable, CDS has a positive absorption effect, which successfully transfers the risk and reduces the banking systemic risk; When the economy is volatile, the excess risk assets released by banks due to CDS are transformed into new systemic risk; the size of CDS is negatively correlated with the banking systemic risk and there is a critical value of size.

In order to explore the dynamic evolution of innovator collaborative innovation network, networks are constructed based on the data of joint invention patent in electronic information industry. The paper reveals the characteristics of networks evolution, and constructs random actor models to identify the influencing factors of networks evolution. The results show that the relationship between innovators and innovation in electronic information industry increases in the early stage and becomes stable in the later stage. The network scale tends to be stable and the connections between nodes become closer. Network evolution is influenced by network structure characteristics, innovator characteristics and proximity mechanism. Among them, cooperation breadth and geographical proximity have the greatest impact.

To improve pipeline safety monitoring and maintenance, the optimal risk transmission path of the natural gas pipeline network is explored. Firstly, the network topology is constructed based on complex network theory, and the importance of network nodes is ranked by entropy weight-TOPSIS method. Secondly, the risk propagation model of the natural gas pipeline network is constructed, the failure rate and vulnerability of network nodes are defined, and the risk propagation degree and optimal risk propagation path of nodes under deliberate and random failure strategies are obtained. Finally, based on the empirical analysis of the Shanghai natural gas pipeline network, the results show that the total risk of intentional damage propagation is greater than that of random damage in the case of cascade risk, which provides a basis for pipeline topology optimization and maintenance.

In order to improve the small key space of the encryption algorithm based on low-dimensional system design, a new five-dimensional complex chaotic system is constructed, and its chaotic dynamics properties are studied by using phase diagram, equilibrium point, Lyapunov exponential spectrum, bifurcation diagram and spectral entropy complexity. It is found that the dynamic behavior of this system is rich, and it is more suitable for image encryption algorithm than low-dimensional system. Furthermore, based on this system, a new encryption scheme is proposed in this paper. The scheme includes three diffusion and one scrambling, which can increase the security of image transmission. Finally, the performance of the encryption scheme is analyzed by calculating key space, information entropy, correlation of adjacent pixels and so on. The analysis results show that the encryption scheme designed in this paper has a large key space, good encryption effect and the ability to resist external attacks.

In order to make the chaotic sequence show more complex characteristics, a piecewise linear function is introduced into the typical Chua’s system to produce a controllable number of multi-scroll attractors. The dynamic analysis of this system is performed, and the results illustrate that the chaotic characteristics of the system are widely distributed and have rich dynamic behaviors. On this basis, an encryption algorithm based on Zigzag scrambling and DNA coding is proposed, and its security is analyzed from multiple aspects. The simulation results indicate that this method can not only encrypt the image effectively, but also resist a variety of common attacks effectively, which provides a theoretical basis for image encryption.

This paper proposes an improved ant colony algorithm to solve the problem of slow and poor convergence. First, a correction strategy is introduced, which includes two local correction methods to reduce invalid paths. Second, an adaptive pheromone updating mechanism is developed to distinguish and volatilize the initial pheromone from the pheromone released. For the pheromone released in each iteration, a change law of time-varying volatilization factor is designed to volatilize independently and obtain pheromone volatilization mechanism with adaptive volatilization intensity. Finally, the proposed algorithm is applied to mobile robot path planning. Compared with the existing improved ant colony algorithms, the results show that the improved algorithm is excellent in terms of effective time, average distance and shortest distance.

In order to address the problems of high empty rate and long waiting time of cross-level four-way shuttle during its operations, it is necessary to fully optimize the level of cooperation between shuttles and hoisters. Considering the variable speed of the four-way shuttle and the hoister in the movement process, with the objective of minimizing the total operation time of the inbound and outbound order tasks, a mixed integer planning model for the joint scheduling of the four-way shuttle and the hoister is established. An improved whale optimization algorithm is proposed in combination with the actual constraints of the four-way shuttle and hoister in the warehouse order picking process, and compared with simulated annealing, genetic algorithm, and an improved chimpanzee optimization algorithm. The results show that the designed algorithm has much better solution quality and effectively reduces the time in completing given order picking task.

With the continuous promotion of the "Health China" strategy, the public's demand for sports and fitness is increasing. However, unclear governance modes and operation strategies seriously hinder the efficient operation and public service of China's public stadiums. Based on the evolutionary game theory, this study explored the government-enterprise cooperative governance mode and operation strategy of domestic public stadiums from the perspective of cooperation benefits, cooperation costs, cooperation risks, financial subsidies, default punishment, supervision, and other factors. The results showed that: in terms of governance mode, increasing the benefits of cooperation between government and enterprise, reducing the risks of cooperation, increasing the financial subsidies for both parties and strengthening the punishment for breach of contract will promote the government-enterprise governance mode from independent management mode to cooperative management mode for collaborative governance. In terms of operation strategy, reducing the cost of government supervision and enterprise cooperation, and increasing the punishment for breach of contract between the government and enterprise will promote the transformation from (no supervision, negative) strategy to (supervision, positive) strategy, so as to better and efficiently realize the government-enterprise cooperative governance of the stadiums.

In order to explore the behavior tendency of multi-stakeholders on community waste classification, firstly, an evolutionary game model of residents, community neighborhood committees and local government under the reward and punishment mechanism is constructed. Secondly, the condition of tripartite equilibrium evolution strategy is discussed by solving the model and analyzing the stability. Finally, MATLAB is used for simulation verification and result analysis. The final results show that community neighborhood committee plays an important role in community waste classification; residents are important participants in the classification of urban domestic waste. Moreover, residents’ waste classification behavior will be directly affected by the supervision and management of neighborhood committees; Reasonable incentives and punishment measures of local governments can effectively mobilize their enthusiasm for participation; Compared with incentives, residents and residents' committees are more sensitive to punishment measures.

In order to study the evolutionary mechanism of lane changing for intelligent connected vehicles on urban traffic flow frequently changing sections and reduce the decision conflict of vehicle lane changing, the human-vehicle interaction was realized and the evolutionary dynamic equations were established by quantifying the vehicle revenue and weight coefficient. According to the Jacobi matrix of the game system, the stability of the equilibrium points were analyzed. Finally, the numerical simulation was used to determine the evolution path and the sensitivity analysis of the influencing factors was carried out. The results show that the initial ratio of different driving decisions will affect the evolution direction of the decision, and the left-turn green time will enhance the evolution of vehicle strategy in the direction of changing lane and giving way, and the evolution of vehicle strategy in the direction of changing lane and giving way as the position of decision time decreases.

This paper investigates the adaptive model predictive control (MPC) for a class of constrained linear multiple-input multiple-output (MIMO) systems with parametric uncertainty and input delay. An adaptive update law based on time-varying updating rate is proposed, which enables the update of uncertain parameters in the presence of input delay. Consequently, to deal with the constraints, we convert the optimization problem into a solvable simple structure, which originates from the min-max optimization. Furthermore, theoretically, it is shown that the closed-loop system is asymptotically stable and the proposed adaptive MPC strategy is proved to be recursively feasible. Finally, numerical simulation is given to illustrate the efficacy of the proposed method.

In order to solve the real-time scheduling problem of virtual eldly service personnel, this paper constructs a cost-optimal scheduling optimization model and a disturbance-minimization management model based on the perspective of demand variation, by improving the location update formula of grey wolf optimization algorithm, the non-dominated ranking design multi-objective genetic grey wolf optimization algorithm is introduced. The superiority of the algorithm is verified by solving the comparison index of the standard example, and the feasibility of the model is verified by designing and solving the example. The results show that, compared with the rescheduling method, the disturbance management model can significantly reduce the influence of disturbance events on the agents, generate more abundant decision sets, and is more suitable for the scheduling problem of virtual elderly service personnel.

To address the problems of low accuracy and poor robustness of traditional data-driven battery state of health prediction model, this paper develops a time series prediction model fusing both nonlinear and linear branches. The nonlinear time series prediction branch is formed by a one-dimensional convolutional neural network with a multi-size parallel structure in series with a bidirectional gated recurrent neural network, and the linear branch is constructed by an autoregressive model. Two branches in parallel output prediction results through a fully connected layer. The above prediction model has the generalization ability of the nonlinear part and the memory ability of the linear part, and is more sensitive to the change of input amplitude. The whale optimization algorithm is used to effectively search the optimal model hyper-parameters. The effectiveness and superiority of the linear nonlinear fusion prediction model proposed in this paper are verified by comparing existing models and ablation experiments.