Invulnerability analysis is one of the core contents of power grid security research. Traditional analysis methods cannot effectively analyze the process of failure generation, and have limitations in the research of invulnerability analysis. This paper studies the invulnerability analysis of power networks based on complex network theory, and conducts an empirical analysis using Chongming District of Shanghai as an example. For random attacks and selective attacks on the power network, the changes in the network efficiency and the maximum number of connected subgraphs after the attack are obtained, and the network efficiency change rate is proposed as a parameter to evaluate the invulnerability. According to the simulation results, a segmented protection scheme based on real-time centrality closeness priority attack strategy is proposed to improve the invulnerability of the power network and ensure the safety of the power grid.

To reveal the characteristics and influencing factors of cross-platform information diffusion in social media, this paper took the Legitimate Defense Case in Kunshan as an example and studied the characteristics and related factors of information diffusion from other platforms to Sina Weibo using statistical inference and regression analysis methods. We found that users are more inclined to the latter when balancing the high amount of information in microblogs with the convenience of obtaining information, and the transmissibility, basic reproductive number and diffusion depth of cross-platform information are significantly lower than those of non-cross-platform information. Information from WeChat official accounts, Weibo videos, Weibo articles and Sina news has more advantages over other types of information in terms of depth and scale of diffusion. Compared with ordinary users, users who are authenticated as media and government administration spread information from news platforms on a larger scale. A comprehensive consideration of the types and source platforms of information can help us to better understand the spreading of sudden social events in the Internet space, thus helping to effectively guide or control the evolution of public opinion.

In order to improve the accuracy of gender recognition for social network users, the text features and emoticon features of a single user are fused to identify the user's gender, and then the interactive feature information of multiple users is extracted to further improve the accuracy of gender recognition. The experimental results show that the accuracy of user gender recognition is improved by 6.8% after the fusion of multi-user interaction features. It shows that emoticons and multi-user interaction features are very helpful to improve the accuracy of user gender identification, and improve the accuracy of gender information identification of social network users.

The extreme sparsity of data limits the recommendation performance of the model in point-of-interest (POI) recommendation task. And the existing work ignores the differences of users' movement in different time periods. To solve the above problems, this paper proposes a POI recommendation model that incorporates time and geographical information. Firstly, the model learns multiple factors through recurrent neural network. Then the geographical relationship module is used to capture the geographical influence in the trajectory. Finally, through a unified framework, different POIs are recommended according to the different visit needs of users on weekdays and holidays. Experimental results demonstrate that the proposed model achieves better recommendation performances than the state-of-the-art methods.

In order to explore the multidimensional influence of directors and leading actors on the film, we establish the director-co-director network, the leading actor-co-director network and the director-leading double model network to analyse the the film market multi-directional from the perspective of network location. The regression results show that the director and the leading style does not affect the film's score, but co-directing can promotes the box growth. Both degree centrality and structure hole of the director and leading actor can promote the increase of film score and box office. However, they do not have superposition effect, the degree centrality has a greater relative influence. Sharing information is better than exclusive information.

In order to explore the evolution law of innovation in electronic information industry, the collaborative innovation network is constructed from the perspective of social network. Based on the patent data from 1985—2017, the evolution path and spatial pattern trend of collaborative innovation is analyzed. The results show that electronic information industry has entered a stage of rapid development with enhanced network connectivity and obvious core-edge structure. The cooperative mode has gradually changed from institute-oriented to enterprise-oriented collaborative innovation mode with universities and institutes as knowledge partners. The spatial pattern is characterized by unbalanced development, and the inter-regional cooperation shows a radiation-type network form dominated by the core region.

The topological characteristics and resilience analysis of public transportation systems are of great significance in urban management to ensure its safe and sustainable operation. This paper constructs a bus-metro interdependent network model based on the passenger transfer relationship and uses deep learning to identify their network topology attributes. A comprehensive importance indicator of the nodes is established by entropy weight-technique for order preference by similarity to ideal solution (EWM-TOPSIS), and the resilience of the network under different recovery strategies are analyzed. In order to verify the applicability and accuracy of the method, this study takes Wuhan's public transportation network as an example, which has practical guiding significance for the post-disaster recovery and operation management of the urban public transportation system.

In order to understand the process of individual extremism in the context of terrorism and effectively prevent the spread of extremist ideology. Based on the Hegselmann-Krause model of opinion dynamics, this paper proposes a dual-threshold opinion dynamics model for the evolution of individual extreme thoughts in the context of terrorism. Simulation experiments found that the interaction scope, different population structures and social network structures have significant influence on the evolution process of extreme thoughts. The model integrates three factors: individual characteristics, social relations and social environment in the process of extreme opinion dissemination,and the simulation calculation experiment reveals to a certain extent the law of the evolution of individual extreme thoughts in the context of terrorism.

The coexistence of attractors in chaotic systems can enhance the security of synchronous communication. Therefore, it is meaningful to establish a chaotic system with the coexistence of attractors. The hyperbolic tangent function is added to the new chaotic system, and the coexistence with infinite attractors is generated by expanding the equilibrium point of the system. The number of attractors generated by this method is adjustable. In addition, the global feedback control law is designed to realize the synchronization of the system. Theoretical research and numerical simulation results verify the effectiveness of the synchronization method. The chaotic system with adjustable attractors has more complex dynamic behavior, so it has good application value in the field of synchronous communication.

This paper studies the group formation control problem of nonlinear and two-integrator coupling leading the following multi-agent systems. In dealing with the grouping of multi-agent systems, a new control protocol is designed, which does not based on the conventional conservative assumption that the sum of adjacent weights of all nodes from each node in one group to all nodes in the other group is zero or another constant. Then, based on Lyapunov stability theory and algebraic graph theory, sufficient conditions for formation control problems of nonlinear and double integrator second-order multi-agent systems are given respectively. So that the agents in the multi-agent system can reach and maintain the designed formation over time. Finally, two numerical simulations are presented to verify the effectiveness of the results.

The purpose of studying the task scheduling problem of the system is to improve the efficiency of the tier-to-tier multi-shuttle warehouse system with double lifts.The energy consumption of shuttle system is considered in the warehousing operation,and the dual objective model is established, including two objectives: operation time and energy consumption of shuttle system.The method of remove scalarization is used to change the dual objective model into a single objective model. An self-adaption genetic simulated annealing algorithm is proposed, and example is given to verify the effectiveness of the model and algorithm.The results show that compared with the traditional genetic algorithm, the adaptive genetic simulated annealing algorithm has higher solution accuracy, the optimization rate of time is increased by 20.7%, and the optimization rate of energy consumption is increased by 15.5%.The experimental results show that, through the warehousing operation model of the tier-to-tier multi-shuttle warehouse system with double lifts established in this paper and its solution algorithm, it can effectively reduce the system energy consumption and time, so as to improve the warehousing efficiency.

In order to generate chaotic dynamic behavior, aiming at a class of chaotic system, the chaos anti control method is improved, and the symmetrical piecewise linear state feedback controller is designed. The chaos of the linear system can be controlled by adjusting the parameters of the controller. The chaos is proved by combining the maximum Lyapunov index. The stability of the equilibrium point is analyzed according to the Routh Hurwitz criterion, and the dynamic behavior of the chaotic system is analyzed in detail. The new system is used to encrypt the image. The simulation results show that the histogram distribution is uniform and the correlation between adjacent pixels is small. It shows that the designed chaos system has high security performance in image encryption system.

This paper mainly studies the problem of intrusion detection, prevention and repair in discrete event systems. In this paper, we use formal language and automata to model a system that is under actuator-enablement attack. In this paper, we use formal language and automata to model a system that is under actuator-enablement attack.First, this paper uses a diagnoser algorithm to analyze the model and judge its safety after being attacked. Then, based on this, an algorithm is proposed to repair the system when it is not safe after being attacked, so that it can meet the requirements of system safety. Finally, the effectiveness of this method is verified by a traffic system.