Papers by Keyword: Multi-Agent System (MAS)

Abstract: This project uses modern control technology to reflect the actual work situation, the MAS model is constructed, the intelligence of power dispatching operation system (PDOS) is improved. By using this system, the security of the power net is enhanced, and the working stress of the staff is lightened. This paper analyzes and researches two different aspects of PDOS. On the one hand, a scheme of operating rules’ dynamic management is designed and implemented, which improves the flexibility of PDOS effectively. On the other hand, the PDOS is integrated with management information system (MIS) to realize the web-based broadcast of DOOS. The efficiency and applicability of this implementing scheme have been proved by practice.

Abstract: This article presents the analysis of a possibility of application of agent-based technology in the e-manufacturing environment. The philosophy of e-manufacturing is a response to the demand of gaining the competitive advantage as a result of a simultaneous reduction of goods/materials in stock and the increase of the quality level of customer service. It consists in the complete integration of the company’s environment, i.e. the areas connected with production and the areas related to company’s administration and management. It may be done on the basis of communication applying cutting-edge IT technology with the use of intelligent IT techniques. Agent-based systems constitute a relatively new approach to analysing, designing as well as implementing scattered applications meant to be used in the dynamic environment of e-manufacturing.

Abstract: The work is devoted to solve allocation task problem in multi agents systems using multi-objective genetic algorithms and comparing the technique with methods used in game theories. The paper shows the main advantages of genetic algorithms and the way to apply a parallel approach dividing the population in sub-populations saving time in the search and expanding the coverage of the solution in the Pareto optimal space.

Abstract: Multi-agent system (MAS) contains many intelligent agents, which makes decision for many practical scenarios. However, due to complex decision environment, many data required from sensors are fuzzy. Although a lot of legend solutions address this problem, the decision making result is not very accurate. To solve the problem, we present IFMOD, an algorithm using fuzzy multi-objective decision based on interval-valued intuitionistic fuzzy set. The algorithm adopts iteration way to close the precise value. To make IFMOD algorithm effectively, MAS makes hierarchical structure to complete fuzzy decision. In MAS, most intelligent agents may process the data individually, while other higher intelligent agents make a final decision entirely. The experiment shows that, it is effective for MAS to solve the problem of fuzzy decision in distributed networks.

Abstract: Under the constraints of distribution transformer capacity and customer charging needs, a UML model for EMS of EV Charging Station is proposed to analyze energy dispatching functions and integrate control strategies of Charging Station. Due to the fact that EV charging system is a complex adaptive system, Multi-Agent System method is utilized in this model. The UML development process provides an extensible and integrated framework, which lays the foundations for the further development on EMS of EV Charging station.

Abstract: The cooperative output regulation problem for a class of nonlinear uncertain multi-agent systems is considered. Based on the distributed internal model, the problem is firstly transformed into a global stabilization problem of the augmented system. Then, using the backstepping design method, a distributed control law with its stability analysis is proposed to solve the global stabilization problem of the augmented system. Finally, a numerical simulation is made to show the efficacy of the analytic results.

Abstract: Microgrid reconfiguration plays important role in improving continuity of power supply especially in the event of fault within the power grid or microgrid connected to the medium voltage distribution network. Due to the fault in power grid, intentional islanding through static switch is one possible solution while, fault within the microgrid emphasizes microgrid reconfiguration to recover maximum affected loads in each isolated zone. This paper presents multiagent system based (MAS) approach in order to withstand stably and autonomously in the event of fault within microgrid or power grid. MAS approach includes negotiation agent (NA) for the entire microgrid and bus agent (BA) for each bus within each zone. NA is responsible for selecting a Dominant-BA from each zone and prioritizing the zones by using region weight coefficient (RWC) method. Zone having excessive power will recombine with the zone having lesser power for maintaining power quality of sensitive and less sensitive loads, while normal loads are shed.

Abstract: In power operation, dispatching operation order sheet (DOOS) provides reliable guarantee for safe of power dispatching. The intelligence of power dispatching operation system (PDOS) is improved a lot through the effort of scientists during recent years. By using this system, the security of the power net is enhanced, and the working stress of the staff is lightened. However, such systems can not be fully popularized and applied as yet due to some reason.. This paper analyzes and researches two different aspects in order to enhance the universality of PDOS. On the one hand, a scheme of operating rules’ dynamic management is designed and implemented through analyzing the difference on operating requirement and content between itemized order tickets and comprehensive order tickets, which improves the flexibility of PDOS effectively. On the other hand, the PDOS is integrated with management information system (MIS) to realize the web-based broadcast of DOOS. The efficiency and applicability of this implementing scheme have been proved by practice.

Abstract: The paper considers the solution to the problem of distributed voltage regulation in Smart Grid System. It takes into consideration the conflict of interests of companies included in the generation network and consumers of electricity. The paper presents an innovative method of demand side management which can be used for controlling the quality of electricity optimization of consumption and conservation of electric power in distribution grids of low voltage class; to start a service of voltage regulation to the consumers in the parts of the grid adjoining to the nodes of generation; to develop the grid and realize the regulation of the voltage by the network companies during load and generation active behavior. The decentralized regulation of voltage is supposed to be the basic organization principal which uses local information and control actions. The control of the facilities of voltage regulation is performed via the artificial intelligence (expert systems).

Abstract: Firstly, associative-sets-based collaborative filtering algorithm is proposed. During the process of personalized recommendation, some items evaluated by users are performed by accident, in other words, they have little correlation with users' real preferences. These irrelevant items are equal to noise data, and often interfere with the effectiveness of collaborative filtering. A personalized recommendation algorithm based on Associative Sets is proposed in this paper to solve this problem. It uses frequent it sets to get associative sets, and makes recommendations according to users' real preferences, so as to enhance the accuracy of recommending results. Test results show that the new algorithm is more accurate than the traditional. Secondly, a flexible E-Commerce recommendation system is built. Traditional recommendation system is a sole tool with only one recommendation model. In e-commerce environment, commodities are very rich, personal demands are diversification; E-Commerce systems in different occasions require different types of recommended strategies. For that, we analysis the recommendation system with flexible theory, and proposed a flexible e-commerce recommendation system. It maps the implementation and demand through strategy module, and the whole system would be design as standard parts to adapt to the change of the recommendation strategy.