Ontology-driven complex event processing for real time algal bloom detection

AOW Dec 2011 Jonathan YuKerry Taylor and Brad Sherman

Chaffey dam

CSIRO.

Chaffey dam

CSIRO.

Chaffey Dam affected by algal blooms

CSIRO.

Proof-of-concept andExample Complex Event Definition

• Given a set of data streams which monitor a range of parameters –

• Total Phosphorus (TP),

• Total Nitrates (TN),

• Meteorological conditions,

• Temperature range in the water column;

• Able to define some complex high-level events (e.g. potential algal bloom), given the aggregation of low-level factors (e.g. Rise in TP, TN)

• Working from a database curated from a study on Chaffey Dam in the 1990s

CSIRO.

Ontology-driven complex event system

CSIRO.

1. Composes CE

Sensor Network

Ontology-enabledUser Interface

Semantic Mediator

GSN

VSensors

Ontologies

SSN Ontology

DomainOntology

7.Updates UI with notification

3. Deploys CE to GSN as VSensor via translation

4. Interface with sensors

capture VSensor entitiescapture sensor / data sources

6. Matching event notification generated

2. Submits CE definition

captures notification

captures CE definition

8. Views notifications

5. Sensor streams

data

Users

Reasoner

Future work

• Still work to be done to finish developing domain ontologies to allow for complex events to be defined

• Mapping the SSN-XG ontology with the relevant ‘sensors’ – in this case we simulate sensors using the Chaffey Dam historical dataset

• Deploying on a live sensor network• Application of this approach to other water quality monitoring

challenges – in this domain, driving force is in determining what kinds of response required.

CSIRO.

Questions

Jonathan Yu

Research Engineer

Environmental Information Systems Group

CSIRO Land and Water

phone: (03) 9252 6440

email: jonathan.yu@csiro.au

CSIRO.