Aiming at the division roughness of topic classification existing in the most online social networks community, the improved particle swarm optimization algorithm is applied to refine community topics and concepts of ...
详细信息
Aiming at the division roughness of topic classification existing in the most online social networks community, the improved particle swarm optimization algorithm is applied to refine community topics and concepts of community seeds and community topic are also introduced. In this paper, first of all, the explicit links existing in the community are mined, and the basic community structure is constructed, then the community content is deeply analyzed, according to implicit feature between nodes under online community, community topic categories are elaborately refined until structure is stable. Experiments show that this proposed algorithm can accelerate the convergence of the node and greatly improves the topic mining accuracy of online social network compared with the state-of-art CR2NDAS model and PLSA model.
In the ac/dc hybrid micro-grid, the reasonable configuration of the micro power to improve power supply reliability and economy is the most important issue due to the types and combinations of distributed power source...
详细信息
In the ac/dc hybrid micro-grid, the reasonable configuration of the micro power to improve power supply reliability and economy is the most important issue due to the types and combinations of distributed power sources are too many. The paper considers the power installation cost, the operation maintenance cost and the equipment replacement cost, as well as the loss of power supply probability and the excess energy ratio as the multi-objective function. The improved particle swarm optimization algorithm is used to optimize the micro power. Through matlab simulation verification, and the basic particleswarmoptimizationalgorithm for comparative analysis. The results indicate that the improvedparticleswarmalgorithm can effectively save the cost of the system and improve the operation efficiency of the ac/dc hybrid micro-grid.
With the development of smart grid, the power data resources increase rapidly, and the potential value of data is gradually explored. This paper based on clustering algorithms, clustering massive amounts of low-voltag...
详细信息
ISBN:
(纸本)9781450388665
With the development of smart grid, the power data resources increase rapidly, and the potential value of data is gradually explored. This paper based on clustering algorithms, clustering massive amounts of low-voltage station users' electricity data, and analysing each user group by clustered electricity consumption behaviour. Classify the low-voltage station according to the electricity consumption behaviour and extract the characteristic parameters of the low-voltage station. Then comprehensively considering the users' electricity consumption behaviour, and characteristic parameters of the typical low-voltage station, the support vector regression (SVR) is used to construct the typical low-voltage station model. The improvedparticleswarmoptimization is used to optimize the SVR parameters. Finally, different low-voltage stations in Tianjin are selected to verify the effectiveness of the proposed model.
A new nonlinear identification method of the platinum resistance sensor based on radial basis function neural network using a improved particle swarm optimization algorithm is proposed to settle its nonlinear *** part...
详细信息
A new nonlinear identification method of the platinum resistance sensor based on radial basis function neural network using a improved particle swarm optimization algorithm is proposed to settle its nonlinear *** particleswarmoptimizationalgorithm is improved by introducing the shrinkage factor and the particle variation *** function of the particle fitness is achieved based on the distance between the actual neural network output values and the expected output *** the global optimum value in the swarm searching space as the initial value of network *** simulation shows that the new nonlinear identification has better nonlinear identification accuracy and *** is proved that the method is effective and feasible.
To solve the unreasonable thrust distribution,turning safety,control flexibility and excessive energy consumption of hovercrafts.A coordinated thrust distribution method of multiple control surfaces for hovercrafts na...
详细信息
ISBN:
(数字)9789887581581
ISBN:
(纸本)9798350366907
To solve the unreasonable thrust distribution,turning safety,control flexibility and excessive energy consumption of hovercrafts.A coordinated thrust distribution method of multiple control surfaces for hovercrafts named improved adaptive compression factor particleswarmoptimization(PSO) algorithm is proposed.A coordinated thrust distribution model with vector nozzle is established to describe the relationship between each control surface and its generated force(moment).More over,an improved adaptive compression factor PSO algorithm is proposed to distribute the resulting forces(moment) reasonably to each control *** salient features are as ***,an adaptive compression factor is introduced to improve the global search ability in the early algorithm ***,an adaptive function is introduced to enhance the local search and convergence *** results are given to demonstrate the proposed method can effectively distribute the resulting force(moment) to each control surface,improving the turning safety,control performance,and reducing energy consumption for hovercrafts.
暂无评论