No abstract

No abstract

The paper interprets a smart grid as an enhanced service environment. For a reliable, autonomic integration of services without permanent actions of experts, the Service Discovery has to agree on the semantic interpretation of nonfunctional and technical criteria. The proposed service environment for smart grids applies our Service Discovery framework ACTAS (Adaptive Composition and Trading with Agents for Services), which is based on software agents and enhances SOC with ontologies for the classification and quasistandardization of criteria and their constraints for a commonly agreed Service Discovery, Composition, and Selection taking advantage of established algorithms.

In distributed systems, traders mediate between clients and service providers. This paper introduces a trading model, which supports multiagent systems (MAS) and goes beyond simple trading in three ways: (a) Service composition -The trader composes complex services of the current service offers. During the composition, it checks the availability of the service offers. (b) Use of group agents -Group agents represent a group of agents with their individual policies and other context information. The trader can use the group agent's information for a pre-selection of service offers. (c) Adaptability -The trading model uses the notion of clients' trust into services and adapts to the clients' preferences and system policies. The trading model is used in a Computer-Supported Cooperative Work (CSCW) application, in which the trader finds adequate communication services for project teams with geographically distributed members.