Use this URL to cite or link to this record in EThOS:
Title: A method for development and validation of multi-agent systems using accurate communication network modelling
Author: Perkonigg, Fidelis
ISNI:       0000 0004 5349 4841
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
There has been a considerable amount of research on multi-agent system (MAS) technologies for a wide variety of industrial applications. One application domain is the power industry, for which multi-agent systems are widely suggested as a promising method for the realisation of highly distributed, flexible, fault tolerant management and control applications. Multi-agent systems make extensive use of digital communication, which can significantly influence the overall system performance. However, no general solutions have been proposed for the difficult tasks of multi-agent system development and validation that would fully account for the underlying communication network performance, before it is first deployed on the target system. This work proposes a new method for this purpose and presents a novel platform that consists of a federation of a standardised multi-agent system development framework (JADE) and an industry standard network simulator (OPNET Modeler). It was realized through generic extensions of the JADE framework to provide discrete event scheduling capabilities, while the OPNET Modeler was extended to provide a generic method of relating network nodes with agents running in JADE. The federation adheres to the High Level Architecture standard. The multi-agent systems analysed using this platform may be deployed on the target system without manual modifications. An example of a time-critical, agent-based protection system for the Smart Grid is presented and its performance analysed with respect to candidate agent behaviours and different communication scenarios. The results clearly show that the feasibility of the multi-agent system critically depends on the application design as well as the communication infrastructure. The developed multi-agent system was shown to be directly deployable on target hardware, which proves that the proposed method not only supports analysis through simulation but also subsequent deployment. The new platform can be used to rapidly develop a wide range of agent-based applications and validate them for different communication technologies before deployment.
Supervisor: Ristic, Mike; Brujic, Djordje Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available