Fuzzy community structure can effectively improve the network transmission performance. In this paper, a new link rewiring strategy, community combination information link rewiring strategy (CCLS), is proposed based on the community structure and definition of link importance by using the assortativity or dissortativity between nodes and kcorestructure, which can delete links within the community and add links between communities to weaken the community structure and improve the network capacity. In order to verify the effectiveness of the method, we carried out simulation experiments on pseudo-random network, CWS small-world network with community structure, scale-free network with community structure and real network respectively. The simulation results show that CCLS strategy can effectively weaken the network community characteristics, and improve network transmission capacity.

There are various dynamic allocation problems in real life. However, the swarm intelligence division of labor has a natural advantage in solving such dynamic allocation problems because it simulates the division of labor between biological groups.At present, most studies on division of labor with swarm intelligence focus on the stimulation-response principle of ant colonies, while ignores the principle of activating and inhibition of bee colonies. Different from the individual environment interaction mode of the stimulation-response principle, the principle of activating and inhibition adopts interaction mode among individuals. Aiming at the phenomenon of labor division of bee colonies, this paper proposes an activating and inhibition labor division model (AILD). In order to verify the validity of the AILD model, traffic signal timing,a typical time allocation problem,is selected and the corresponding activating and inhibition labor division signal timing algorithm (AILD-ST) based on the evolutionary solution is proposed. This paper uses the real traffic flow data to implement simulation experiment based on AILD-ST algorithm. Compared with Webster algorithm, ant colony algorithm and bee colony algorithm, the results show that the proposed algorithm is of good quality and high calculation efficiency

In order to reveal the evolution mechanism of the taking off and landing safety risk of amphibious seaplane, and effectively prevent the safety risk of seaplane in taking off and landing stage, the right-oriented network topology structure of the amphibious seaplane take-off and landing safety risk evolution was constructed based on the action path between risk factors, and regression analysis was used to verify the scale-free characteristics of the complex network. The node degree centrality, betweenness centrality, closeness centrality and comprehensive value were applied to identify the key risk factors from different perspectives. Matlab was used to analyze the functional robustness and structural robustness of the network under random and deliberate attacks. The robustness effects of degree attack, betweenness centrality attack, closeness centrality attack, comprehensive attack were contrasted, the key risk factors were identified and the chain-breaking control strategy was proposed. The results show that the safety risk network of amphibious seaplane take-off and landing is a scale-free network; the robustness of the network under random attack is stronger than deliberate attack, and the structural robustness of the degree attack is the worst, and the performance robustness of the comprehensive value attack is the worst; nodes with higher comprehensive values are the key risk factors of the network, and priority disposal of key nodes can help prevent taking off and landing accidents.

Complex networks nodes can have multiple attributes, and different attributes or attribute sets will lead to different connections between nodes, thus the network would have different functions.Aiming at the problem that existing complex network models can not construct networks with different functions according to the selected attributes of nodes, a multifunctional complex network model is proposed.The network model is represented only by nodes and their attribute sets. Different network topology and network functions are determined by selecting the attributes of nodes and defining the mapping rules of nodes under the corresponding attributes.By establishing and analyzing a missile defense network, the availability and effectiveness of the multi-functional complex network model are verified

To compare the existing management modes and explore the evolution rules of social network in old residence communities, this paper applies to the complex network theory to constructing the social network evolution model. The model carries out simulation experiments on the evolution of social network under different management modes. And simulation results are verified by the statistical data of 1238 valid questionnaires in Shaanxi province. Results show that: 1)after the housing system reform, the information transmission paths between residents in old residence community have raised sharply, and it is easy to produce social contradictions; 2)with the scale expansion of old residence community, the relationships between residents is more stable and the length of information transmission paths is shorter. But the management burden will be aggravated at same time. The differences in management effects of different management modes will also be reduced; 3)the management mode of property company can fully communicate the relationship between residents and alleviate the social contradictions in old residence community, but it is not always the best management mode; 4)the principal-agent problem should be paid attention to when the large-scale old residence community adopts the market-oriented management mode, and the small-scale old residence community should give full play to the residents’ capability of self-government

This paper investigates the affection of automorphism on controllability for the multi-agent systems with leader-follower framework. By using graph theory and matrix theory, we obtain a criterion for determining the existence of automorphism structure of multi-agent systems. Based on the analyzing of automorphism, we point out the relationship between automorphism and controllability of, which provides a method on the aspect of graphics to identify judging the controllability. In addition, this paper also extends the single leader symmetric systems to the multi-leader symmetric systems, which investigates the controllability of leader selection, and provides the directions and methods for further researches on the controllability of complex topological structures

UAV swarm has greatly broadened the application fields of UAV, and effectively remedied the shortcomings of single UAV in mission execution. In this paper, a novel algorithm based on multi-agent system (MAS) is designed, which integrates the leaders and followers control algorithms in leader-follower method for UAV formation motion consensus and formation control. This algorithm is suitable for both leaders and followers, and can randomly designate UAV as a leader. Leaders can be influenced by navigation feedback from followers, which improves the adaptability and scalability of the system. Theoretical analysis shows that UAV formation will not collide and converge to the desired formation. Numerical simulations are performed to illustrate the theoretical results and verify the velocity and trajectory tracking abilities of the algorithm. The results show that the algorithm can guarantee a swarm of UAVs to fly in the scheduled formation and track

Genetic algorithm (GA) is a natural heuristic algorithm based on biological evolutionism, which has been widely used in numerous fields. At present, studies on GA are mainly focused on theoretical research, improvement and applications of GA. In order to improve the performance of GA, this paper proposes a GA based on an improved BA network and existed research results combining complex networks with GAs, in view that complex networks are powerful model of collective behavior and individual relationships in a group composed of many individuals. The improved GA realizes further improvement of the network structure and improves the selection strategy of the traditional GA and the population size adaptive strategy adopted in response to the incremental use of nodes in the network. The specific steps and methods of the improved algorithm are given in detail and the performance of the improved algorithm is verified by numerical experiments. The results show that the improved algorithm performs better both than the basic GA and GA based on ordinary BA network. The results have a certain guiding role for the improvement of GA

The handwriting digital recognition methods generally are sensitive to the noise and the structure of image, which easily leads to the decrement of recognition accuracy and the increment of computation complexity. In this paper, the objective cluster analysis algorithm is introduced and combined with the template matching mechanism. In order to reduce the effects of the noise and the data distribution, and to improve the recognition accuracy, the one-pass clustering for the template set of the numeral to be identified is proposed. Furthermore, the new clustering centers are used to simplify the primary template dataset, by which the computation efficiency can be enhanced. The simulation about the random handwriting recognition in presence of structural deformation and noise demonstrates the simplicity, practicability and effectiveness of the proposed approach by comparing with the traditional methods

Bearing is one of the most widely used parts in rotating machinery. However, the bearing often fails because of the poor work environment such as high speed and heavy load, which results the equipment stops and even casualties. The fault-induced impulses are too weak to be detected. A novel matched-stable stochastic resonance (MSR) is proposed for bearing fault diagnosis. Unlike the traditional fixed-stable stochastic resonance like the bi-stable stochastic resonance, the potential structure and potential well number of the MSR are changed according to the complicated and diverse vibration signals, which is more benefit for the enhancement of the weak bearing fault-induced impulses. Through the bearing inner ring fault case and rolling element fault case, the proposed method is effective for bearing fault diagnosis and better than the traditional bi-stable stochastic resonance.