An intelligent agent-based distributed architecture for Smart-Grid integrated network management

This paper describes a distributed software architecture designed to manage, in an integrated way, interoperable Smart-Grid networks. It has been designed to address the fact of having massive deployments of AMI, DA, and HAN devices (meters, concentrators, sensors) from multiple vendor technologies, which must be accessed through several communications networks. This network management system is used to deliver the service configurations to the Smart-Grid, retrieve data measurements, and monitor network availability. The proposed architecture is based on autonomous, intelligent, cooperative, distributed agents. The intelligence is achieved by means of data ontology definitions and rule-engines. Data ontologies are retrieved from the different reference data models (smart-grid, communications, proprietary and other models). Each agent is able to specialize in an autonomous way by means of rule sets and the specific data models. Agents are reactive to external events and communicate by means of an event-based real-time middleware. This middleware provides a structured distributed event-bus. Using this architecture, it is possible to monitor the whole network status, deliver complex service definitions to the smart-meters, upgrade device and service configurations, and retrieve measurements with security. In addition, as the intelligence is distributed in agents, it is possible to segregate the different functional roles in the service, associating them to the corresponding agents. This architecture has already been tested on huge device deployments.

[1]  David A Chappell,et al.  Enterprise Service Bus , 2004 .

[2]  Randy Presuhn,et al.  Management Information Base (MIB) for the Simple Network Management Protocol (SNMP) , 2002, RFC.

[3]  Lisandro Zambenedetti Granville,et al.  On the Performance of Web Services Management Standards - An Evaluation of MUWS and WS-Management for Network Management , 2007, 2007 10th IFIP/IEEE International Symposium on Integrated Network Management.

[4]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[5]  Thomas M. Chen,et al.  A model and evaluation of distributed network management approaches , 2002, IEEE J. Sel. Areas Commun..

[6]  Edward Curry,et al.  Message‐Oriented Middleware , 2005 .

[7]  Michael Luck,et al.  Agent-Based Software Development , 2004 .

[8]  Barbara Messing,et al.  An Introduction to MultiAgent Systems , 2002, Künstliche Intell..

[9]  Julie A. Adams,et al.  Multiagent Systems: A Modern Approach to Distributed Artificial Intelligence , 2001, AI Mag..

[10]  Bernhard Fürlinger.  eTOM - Enhanced Telecom Operations Map , 2008 .

[11]  Bert Wijnen,et al.  Coexistence between Version 1, Version 2, and Version 3 of the Internet-standard Network Management Framework , 2000, RFC.

[12]  Bert Wijnen,et al.  An Architecture for Describing Simple Network Management Protocol (SNMP) Management Frameworks , 2002, RFC.

[13]  Stefan Wallin,et al.  Rethinking Network Management Solutions , 2006, IT Professional.

[14]  Gregor Hohpe,et al.  Enterprise Integration Patterns: Designing, Building, and Deploying Messaging Solutions , 2003 .

[15]  Mats Larsson,et al.  Active Management of Distributed Energy Resources Using Standardized Communications and Modern Information Technologies , 2009, IEEE Transactions on Industrial Electronics.

[16]  Falko Menge.  Enterprise Service Bus , 2007 .

[17]  Peter Norvig,et al.  Artificial intelligence - a modern approach, 2nd Edition , 2003, Prentice Hall series in artificial intelligence.

[18]  Nada Golmie,et al.  NIST Framework and Roadmap for Smart Grid Interoperability Standards, Release 1.0 , 2010 .

[19]  Valeriy Vyatkin,et al.  Concept for intelligent distributed power system automation with IEC 61850 and IEC 61499 , 2008, 2008 IEEE International Conference on Systems, Man and Cybernetics.