Papers by Author: De Jia Shi

Abstract: A novel multi-agent particle swarm optimization algorithm (MAI'SO) is proposed for optimal reactive power dispatch and voltage control of power system. The method integrates multi-agent system (MAS) and particle swarm optimization algorithm (PSO). An agent in MAI.SO represents a particle to PSO and a candidate solution to the optimization problem. All agents live in a lattice-like environment, with each agent fixed on a lattice-point. In order to decrease fitness value, quickly, agents compete and cooperate with their neighbors. and they can also use knowledge. Making use of these agent interactions and evolution mechanism of I.SO. MAPSO realizes the purpose of' minimizing the value of' objective function. MAPSO applied for optimal reactive power is evaluated on an IEEE 30-bus power system. It is shown that the proposed approach converges to better solutions much faster than the earlier reported approaches

Abstract: Knowledge management modeling is foundation of knowledge management practice. Previous work of knowledge management modeling is mainly about concept architecture. Its tools and methods are not strong. This paper suggested a new knowledge management meta-model based on the analyzing and integration of knowledge management model. The core of the meta-model includes organizational structure, business process, resource and knowledge. A knowledge management definition language (KMDL) is proposed, which defines a least set of knowledge management modeling components and properties. It also offers a shareable exchange format in common sense. Based on the meta-model and KMDL, we developed a knowledge management modeling tool.

Abstract: With MAS(Multi Agent System)coordinated technology and market bidding game rules, a grid resource allocation model based on market economy was introduced, which could show the relation between supply and demand. The utility function of consumer was given; The results of emulation test showed that the resource allocation algorithm could be a reference to resources of consumers and the approach significantly improves resource utilization, which made the allocation of the whole network resource tend to be more reasonable.

Abstract: In order to automatically determine the optimal threshold in image segmentation, this paper presented a new method of image segmentation based on improved genetic algorithm combined with mutual information; it used this improved genetic algorithm to globally optimize infrared image segmentation functions. This method could automatically adjust the parameters of genetic algorithm according to the fitness values of individuals and the decentralizing degree of individuals of the population, and kept the variety of population for rapidly converging to get the optimal thresholds in image segmentation, it overcame the shortcomings including worse convergent speed, easy to premature that exist in traditional genetic algorithm etc.

Abstract: Mulit-agent system[MAS] research on learning has been in the area of negotiation, and learning strategies of other agents.This paper presents an agent learning approach in multi-agent system based on Bayesian learning, it researches to develop agents that learn free-text queries and keyword searches in MAS. The MAS learns to identify an appropriate agent to answer free-text and natural language queries as well as keyword searches submitted by users. The paper describes how Bayesian learning is implemented in MAS, and analyzes the effectiveness of MAS learning based on the Bayesian learning approach by analyzing the accuracy and degree of learning.