The existing search engines are lack of the consideration of personalization and display the same search results for different users despite their differences in interesting and purpose. By analyzing user＇s dynamic s...The existing search engines are lack of the consideration of personalization and display the same search results for different users despite their differences in interesting and purpose. By analyzing user＇s dynamic search behavior, the paper introduces a new method of using a keyword query graph to express user＇s dynamic search behavior, and uses Bayesian network to construct the prior probability of keyword selection and the migration probability between keywords for each user. To reflect the dynamic changes of the user＇s preference, the paper introduces non-lineal gradual forgetting collaborative filtering strategy into the personalized search recommendation model. By calculating the similarity between each two users, the model can do the recommendation based on neighbors and be used to construct the personalized search engine.展开更多
Flexible AC transmission systems(FACTS)devices can effectively optimize the distribution of power flow.Power flow entropy can be applied as a measure of load distribution.In this paper,a method is proposed to optimize...Flexible AC transmission systems(FACTS)devices can effectively optimize the distribution of power flow.Power flow entropy can be applied as a measure of load distribution.In this paper,a method is proposed to optimize the distribution of power flow with the coordination of multi-type FACTS devices and establishes the corresponding mathematical models.The modified group searcher optimization(GSO)algorithm is proposed,in which the angle search is combined with chaotic search model to avoid jumping into local optimization.Compared with the different optimal allocation of multi-FACTS devices,the optimal allocation of multi-FACTS devices is achieved under the economic constraints.The locations obtained by this method can achieve the purpose of balancing power flow and enhancing the system performances.The simulations are demonstrated in an IEEE 118-bus power system with two classical types of FACTS,namely static var compensator(SVC)and thyristor controlled series Compensator(TCSC).The simulation results show that the proposed method is feasible and effective.展开更多
基金supported by the National Natural Science Foundation of China (60432010)the National Basic Research Program of China (2007CB307103)+1 种基金the Fundamental Research Funds for the Central Universities (2009RC0507)Important Science & Technology Specific Project of Guizhou Province (【2007】6017)
文摘The existing search engines are lack of the consideration of personalization and display the same search results for different users despite their differences in interesting and purpose. By analyzing user＇s dynamic search behavior, the paper introduces a new method of using a keyword query graph to express user＇s dynamic search behavior, and uses Bayesian network to construct the prior probability of keyword selection and the migration probability between keywords for each user. To reflect the dynamic changes of the user＇s preference, the paper introduces non-lineal gradual forgetting collaborative filtering strategy into the personalized search recommendation model. By calculating the similarity between each two users, the model can do the recommendation based on neighbors and be used to construct the personalized search engine.
基金This work was funded by National Science and Technology Support Program of China(2010BAE00816).
文摘Flexible AC transmission systems(FACTS)devices can effectively optimize the distribution of power flow.Power flow entropy can be applied as a measure of load distribution.In this paper,a method is proposed to optimize the distribution of power flow with the coordination of multi-type FACTS devices and establishes the corresponding mathematical models.The modified group searcher optimization(GSO)algorithm is proposed,in which the angle search is combined with chaotic search model to avoid jumping into local optimization.Compared with the different optimal allocation of multi-FACTS devices,the optimal allocation of multi-FACTS devices is achieved under the economic constraints.The locations obtained by this method can achieve the purpose of balancing power flow and enhancing the system performances.The simulations are demonstrated in an IEEE 118-bus power system with two classical types of FACTS,namely static var compensator(SVC)and thyristor controlled series Compensator(TCSC).The simulation results show that the proposed method is feasible and effective.
