APPLICATION DES SYSTEMES STRUCTURES A L’ETUDE DU

Marie-Therese

Description

Master Research Proposal Decentralized approach to fault detection and isolation in a solar power-plant Supervisors: Alain Kibangou and Federica Garin E-mail: [e-mail], [e-mail]: NeCS ( http://necs.inrialpes.fr) Start: February 2011 Duration: 5 months Context: This work will be carried out in the NeCS team (Networked Control Systems), a joint CNRS/INRIA team at GIPSA-Lab laboratory in Grenoble, France, within the project SMART ENERGY (MINALOGIC project, under revision). The objective of the team is to propose an innovative approach to fault detection and isolation in an electrical network, by formulating and solving such problems as distributed estimation problems over a sensor network. Topic description: Stimulated by increasing energy demand and ecological concerns, clean energy production with renewable resources is a key research topic that presents a largely unexplored potential of development. Solar farms constitute power plants of the future. In such systems, electricity is produced thanks to the combined action of a large number of interconnected modules (solar panels). Each module individually produces energy, but only their interconnection allows reaching the global task of a relevant energy production. Due to the interconnection topology, a local fault on a given module can induce damageable effects on the whole network. In order to detect and ...

Informations

• Publié par: Assa
• Nombre de lectures: 17
• Langue: English

Extrait

Département d’Automatique – GIPSA-Lab – B.P. 46 – 38402 – Saint-Martin-d’Hères-Cedex

Master Research Proposal

Decentralized approach to fault detection and isolation in a solar power-plant

Supervisors:

Alain Kibangou and Federica Garin

E-mail:

[e-mail]

,

[e-mail]

Team: NeCS (

http://necs.inrialpes.fr

)

Start: February 2011

Duration: 5 months

Context:

This work will be carried out in the NeCS team (Networked Control Systems), a joint CNRS/INRIA

team at GIPSA-Lab laboratory in Grenoble, France, within the project SMART ENERGY (MINALOGIC project,

under revision). The objective of the team is to propose an innovative approach to fault detection and isolation

in an electrical network, by formulating and solving such problems as distributed estimation problems over a

sensor network.

Topic description:

Stimulated by increasing energy demand and ecological concerns, clean energy production

with renewable resources is a key research topic that presents a largely unexplored potential of development.

Solar farms constitute power plants of the future. In such systems, electricity is produced thanks to the

combined action of a large number of interconnected modules (solar panels). Each module individually

produces energy, but only their interconnection allows reaching the global task of a relevant energy production.

Due to the interconnection topology, a local fault on a given module can induce damageable effects on the

whole network.

In order to detect and localize a fault, a sensor network can be deployed over the farm. Thanks to the recent

advances in wireless communications, the sensors can be equipped with wireless devices, creating a network of

communicating sensors. A classical way to exploit such a network would be to create a hierarchical (tree-like)

structure which conveys all measurements to a centralized computer which would analyze all data. However, a

failure in a communication link or in the centralized computer would result in breakdown of the whole fault

detection system, which is an unacceptable risk in an application domain of strategic importance such as a

power plant. Therefore we propose a decentralized approach that relies on local data aggregation using the

computing and communicating resources of the sensor nodes. As a consequence, nodes cooperation produces

a global decision, available at each point of the network, and computable even in the case where a few sensors

or links are unavailable.

Usually, a diagnosis procedure can be performed in two main steps:

-

Multi-model estimation, where each model characterizes the nominal or the faulty functioning of the system;

-

Fault detection, by means of the analysis of the innovation.

The overall procedure can also be achieved in a distributed way, by combining local filtering (e.g., Kalman filter)

with data exchange and aggregation (e.g., consensus algorithm). A first part of the research work will be a

survey of the state of the art in decentralized estimation and specifically on distributed Kalman filter. The second

part will be devoted to our specific problem, where, instead of a multi-model approach, we will model the faults

with jump processes. Thus, the state of the system will be augmented with some discrete-valued indicators of

faults, to be estimated. Under such a model, the process is no longer Gaussian, and in this case the Kalman

filter is not optimal. Our goal is to develop an optimal decentralized estimator for this scenario.

Candidate profile:

This work requires strong skills in systems theory (estimation, Observers synthesis, Non-

linear filtering, and Random processes).

Bibliography

•

Olfati-Saber R., “Distributed Kalman Filter with Embedded Consensus Filters”, Proc. 44

th

IEEE CDC, pp. 8179-

8184, Sevilla, Spain, 2005.

•

Franco E., Olfati-Saber R., Parisini T., and Polycarpou M., “Distributed Fault Diagnosis using Sensor Networks

and Consensus-based Filters”, Proc. 45

th

IEEE CDC, pp.386-391, San Diego, CA,USA, 2006.

•

Kibangou A.Y. and Monin A. “GPS based Land Vehicle Positioning using Gaussian Sum Filters.”

IEEE ICASSP

2008, pp. 3653-3656, Las Vegas, Ne, USA, 2008.