Semantic sewer pipe failure detection: Linked data approaches for discovering events Jonathan Yu | Research software engineer Environmental Information.

Slides:



Advertisements
Similar presentations
Dr. Leo Obrst MITRE Information Semantics Information Discovery & Understanding Command & Control Center February 6, 2014February 6, 2014February 6, 2014.
Advertisements

Event detection using ontologies CSIRO LAND AND WATER Jonathan Yu 13 Feb 2013.
From Ontology Design to Deployment Semantic Application Development with TopBraid Holger Knublauch
ASCR Data Science Centers Infrastructure Demonstration S. Canon, N. Desai, M. Ernst, K. Kleese-Van Dam, G. Shipman, B. Tierney.
Detecting sewer rising main events using an ontology-driven event processing system CSIRO LAND AND WATER Jonathan Yu | Research software engineer Paul.
1 Publishing Linked Sensor Data Semantic Sensor Networks Workshop 2010 In conjunction with the 9th International Semantic Web Conference (ISWC 2010), 7-11.
Event dashboard: Capturing user-defined semantics for event detection over real-time sensor data CSIRO LAND AND WATER Jonathan Yu | Research engineer Environmental.
JSI Sensor Middleware. Slide 2 of x Embedded vs. Midleware based Architecture for Sensor Metadata Management Embedded approach assign an IP address to.
Jennifer A. Dunne Santa Fe Institute Pacific Ecoinformatics & Computational Ecology Lab Rich William, Neo Martinez, et al. Challenges.
A Semantically Enabled Service Architecture for Mashups over Streaming and Stored Data Alasdair J G Gray University of Manchester Extended Semantic Web.
Semantic Web and Web Mining: Networking with Industry and Academia İsmail Hakkı Toroslu IST EVENT 2006.
Ambient Intelligence for the networked home environment VantagePoint Tutorial June Darmstadt.
IST NeOn-project.org The Semantic Web is growing… #SW Pages Lee, J., Goodwin, R. (2004) The Semantic.
Performing event detection over real-time sensor data using ontology-driven approaches CSIRO LAND AND WATER Jonathan Yu | Research software engineer Environmental.
Web-based Portal for Discovery, Retrieval and Visualization of Earth Science Datasets in Grid Environment Zhenping (Jane) Liu.
Audumbar Chormale Advisor: Dr. Anupam Joshi M.S. Thesis Defense
Copyright © 2012, Oracle and/or its affiliates. All rights reserved. 1.
Presented to: By: Date: Federal Aviation Administration Enterprise Information Management SOA Brown Bag #2 Sam Ceccola – SOA Architect November 17, 2010.
Vocabulary Services “Huuh - what is it good for…” (in WDTS anyway…) 4 th September 2009 Jonathan Yu CSIRO Land and Water.
©Ian Sommerville 2004Software Engineering, 7th edition. Chapter 18 Slide 1 Software Reuse 2.
February Semantion Privately owned, founded in 2000 First commercial implementation of OASIS ebXML Registry and Repository.
What Can Do for You! Fabian Christ
Discussion and conclusion The OGC SOS describes a global standard for storing and recalling sensor data and the associated metadata. The standard covers.
Information Integration Intelligence with TopBraid Suite SemTech, San Jose, Holger Knublauch
Towards validating observation data in WaterML 2.0 WATER FOR A HEALTHY COUNTRY You can change this image to be appropriate for your topic by inserting.
An approach to Intelligent Information Fusion in Sensor Saturated Urban Environments Charalampos Doulaverakis Centre for Research and Technology Hellas.
Using Vocabulary Services in Validation of Water Data May 2010 Simon Cox, JRC Jonathan Yu & David Ratcliffe, CSIRO.
Knowledge based Learning Experience Management on the Semantic Web Feng (Barry) TAO, Hugh Davis Learning Society Lab University of Southampton.
TWIRL Twinning virtual World (on- line) Information with Real world (off-Line) data sources Kick-Off Meeting Cassidian 08 & 09 October 2012, Paris - France.
November 2003 Presented to “Commercializing RDF” Semantic Software Solutions for Enterprise Web Management International World Wide Web Conference 2004.
The GRIMOIRES Service Registry Weijian Fang and Luc Moreau School of Electronics and Computer Science University of Southampton.
Digital Enterprise Research Institute HADA – An Access Controlled Application for Publishing and Discovering Linked Government Data Owen Sacco.
1 Foundations V: Infrastructure and Architecture, Middleware Deborah McGuinness TA Weijing Chen Semantic eScience Week 10, November 7, 2011.
Microsoft SharePoint Server 2010 for the Microsoft ASP.NET Developer Yaroslav Pentsarskyy
Linked-data and the Internet of Things Payam Barnaghi Centre for Communication Systems Research University of Surrey March 2012.
FI-CORE Data Context Media Management Chapter Release 4.1 & Sprint Review.
Metadata. Generally speaking, metadata are data and information that describe and model data and information For example, a database schema is the metadata.
Chapter 10 Analysis and Design Discipline. 2 Purpose The purpose is to translate the requirements into a specification that describes how to implement.
Oracle Database 11g Semantics Overview Xavier Lopez, Ph.D., Dir. Of Product Mgt., Spatial & Semantic Technologies Souripriya Das, Ph.D., Consultant Member.
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
D2.5 Proof-of-Concept Evaluation for Modelling Time and Space.
Ontology-driven complex event processing for real time algal bloom detection AOW Dec 2011 Jonathan Yu Kerry Taylor and Brad Sherman.
© Geodise Project, University of Southampton, Knowledge Management in Geodise Geodise Knowledge Management Team Barry Tao, Colin Puleston, Liming.
TWC-SWQP: A Semantically-Enabled Provenance-Aware Water Quality Portal Ping Wang, Jin Guang Zheng, Linyun Fu, Evan W. Patton, Timothy Lebo, Li Ding, Joanne.
Presented by Jens Schwidder Tara D. Gibson James D. Myers Computing & Computational Sciences Directorate Oak Ridge National Laboratory Scientific Annotation.
N NESSTAR: A Semantic Web Application for Statistical Data and Metadata Pasqualino “Titto” Assini Nesstar Ltd - UK.
User Profiling using Semantic Web Group members: Ashwin Somaiah Asha Stephen Charlie Sudharshan Reddy.
MyGrid/Taverna Provenance Daniele Turi University of Manchester OMII f2f Meeting, London, 19-20/4/06.
Issues in Ontology-based Information integration By Zhan Cui, Dean Jones and Paul O’Brien.
1 Registry Services Overview J. Steven Hughes (Deputy Chair) Principal Computer Scientist NASA/JPL 17 December 2015.
Application Ontology Manager for Hydra IST Ján Hreňo Martin Sarnovský Peter Kostelník TU Košice.
THE SEMANTIC WEB By Conrad Williams. Contents  What is the Semantic Web?  Technologies  XML  RDF  OWL  Implementations  Social Networking  Scholarly.
Knowledge Modeling and Discovery. About Thetus Thetus develops knowledge modeling and discovery infrastructure software for customers who: Have high-value.
A Technical Overview Bill Branan DuraCloud Technical Lead.
Selected Semantic Web UMBC CoBrA – Context Broker Architecture  Using OWL to define ontologies for context modeling and reasoning  Taking.
Module 5: Managing Content. Overview Publishing Content Executing Reports Creating Cached Instances Creating Snapshots and Report History Creating Subscriptions.
Linked Open Data for European Earth Observation Products Carlo Matteo Scalzo CTO, Epistematica epistematica.
Integrating and Extending Workflow 8 AA301 Carl Sykes Ed Heaney.
EBI is an Outstation of the European Molecular Biology Laboratory. Semantic Interoperability Framework Sarala M. Wimalaratne (RICORDO project)
MEKON & HOBO Java Frameworks for building Ontology-Driven Applications Current use cases:  Almost (!) products:  Knowledge-driven clinical documentation.
Data Grids, Digital Libraries and Persistent Archives: An Integrated Approach to Publishing, Sharing and Archiving Data. Written By: R. Moore, A. Rajasekar,
Linking Ontologies to Spatial Databases
Giuseppina Inserra INFN Catania
Middleware independent Information Service
Knowledge Based Workflow Building Architecture
Geospatial and Problem Specific Semantics Danielle Forsyth, CEO and Co-Founder Thetus Corporation 20 June, 2006.
Session 2: Metadata and Catalogues
LOD reference architecture
Semantic Markup for Semantic Web Tools:
About Thetus Thetus develops knowledge discovery and modeling infrastructure software for customers who: Have high value data that does not neatly fit.
Presentation transcript:

Semantic sewer pipe failure detection: Linked data approaches for discovering events Jonathan Yu | Research software engineer Environmental Information Systems, CLW Highett 21 October 2013 CSIRO LAND AND WATER

Undetected sewer rising mains pipe failures... Direct costs: water service providers ($ mil. per event) Indirect costs: social, environmental ($10k - $1 mil. per event) We can apply event detection over sensor networks for addressing issues in urban contexts such as detecting pipe failures There is an extensive network of pipes each with varied material compositions, age, and surrounding soil properties, which makes prediction of pipe failure a little unpredictable. Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Sewer rising mains case study We got hold of data from a recent pipe burst in Victoria, where a pipe failure occurred close to a creek which fed the port phillip bay. It was undetected for more than a week. - only detected when a passer-by walking his dog noticed a significant change in the colour of the creek. The ramification for the water service provider was for them to send trucks in to pump out sewerage until the pipe was fixed, which wasn’t cheap. - the EPA threatened with hefty fines. The graph in this slide show the pipe failure event documented with the red line. The blue line shows the flow rate in litres per second. - notice an upward trend as it built up to the failure event and then sustained a higher than normal flow rate... for a long time. Example event: Flow rate > 100 l/s Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

RDF Triple Store contains ontologies Semantic Sensor Network These ontologies provide semantics and constructs to describe sensors and observations generally General model extended with domain semantics and knowledge Allows definitions to be captured explicitly and used consistently. Extensible – able to capture more domain knowledge/rules E.g. PVC pipe feature definitions and domain rules Event-detection SSN Extensions - quantity values Water domain – flow, pressure, units of measure Pipe features, observations, sensors defs We use a set of ontology modules shown here in this slide. Ontologies offer the ability to import one another, so we can modularise our definitions – separating sensor definitions from the domain and the application definitions. The beauty of this is that we can potentially leverage existing ontologies published through the web, if they exist and reuse them and keep our application definitions fairly separate. We can also separate our user definitions from domain definitions, which enhances the ability to reuse both application and domain semantics. Pipe domain rules – MCA, risk levels, PVC pipes, asbestos concrete pipes

Event Notification Interaction Internal network Event rules deployed in GSN send notifications to web service Web service adds metadata to notification and sends to RDF Triple Store RDF Triple Store persists the sensor observations and event notifications like a semantic knowledge base Web server Sensor Middleware (GSN) Observation and Notification REST Web service Event Rules Virtual Sensors Real-time sensor data Sensor Network SPARQL Event Dashboard – maps sensors to ontologies RDF Triple Store

Event Detection Linked Data API Public accessible network Internal network Web Server Web Server Event Detection Linked Data SPARQL RDF Triple Store List notifications, list sensor observations, view semantic descriptions of pipes, pumps, observed properties allows users to browse contents of a RDF triple store via standard web browser configured to view sensor observations, event notifications, semantic definitions, domain knowledge base Also enables software clients to retrieve JSON/XML/RDF/TXT formats of the same information for mashups and data fusion activities Event Dashboard – maps sensors to ontologies

Event Detection Linked Data Visualization client Public accessible network Internal network Web Server Web Server Viz SPARQL RDF Triple Store Event Detection Linked Data Event Dashboard – maps sensors to ontologies Example of a visualization client querying the RDF triple store for sensor observations and event notifications Identifiers from the RDF triple store resolve to metadata and semantic definitions delivered via the Event Detection Linked Data API

Overall architecture schematic Public accessible network Internal network Web Server Reverse Proxy / auth Web Server Sensor Middleware (GSN) Observation and Notification REST Web service Viz Event Rules Virtual Sensors Real-time sensor data Event Detection Linked Data Sensor Network SPARQL Event Dashboard – maps sensors to ontologies http://waterinformatics-ext1-cdc.it.csiro.au/elda/event/list/observation http://waterinformatics-ext1-cdc.it.csiro.au/elda/event/list/notification RDF Triple Store Event Dashboard

Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Summary Domain & event ontologies: Defining and capturing sewer pipe event descriptions extending SSN ontology and Linked Data APIs and RESTful services Publish and discovery of sewer pipe event notifications and observations Demo visualization client Preliminary work to demo real-time events can be combined with domain knowledge for context sensitive event detection using ontologies Availability of real-time sensor data presents many potential applications Barriers for allowing user-access to event definition over observations barrier to precise definition of event semantics Ontologies offer a means to capture semantics of a domain of discourse - semantics are treated as first class citizens - not implicit in the syntax or code of the user interface, middleware services, databases, documents, text files - Resolving issue of data heterogeneity - facilitates interoperability between systems - semantics captured in ontologies are independent of software components - separation of concerns and greater flexibility and reuse of software components -- e.g interchange domains   Ontology-driven approaches can assist user-definition of events over a given sensor network - using familiar user interfaces - browser based Demonstrated a prototype browser-based user interface Lowers the barrie for users like domain scientists and managers - general purpose tool for exploring sensor data in consistent fashion - Ontology tools allow consistency checking, and inferrencing capability - Allows Capture semantics of user defined events of interest precisely and consistently Combining real-time events (dynamic) with domain knowledge (static) using ontologies Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Event Detection Ontology def’s: Event Rule, Value Constraints, Units Slide shows some application ontology definitions. Value Constraint event rules – which is a simple construct to specify that an event rule may have -- a feature of interest, -- a sensor of interest, -- on an observed property -- with a value constraint on the observed value. E.g. Flow rate observations on a pipe flow sensor, on a pipe at Clunies Ross st, greater than 100 l/s. Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Domain ontologies (uwda:) - Sensors In our system, we have defined some sensors specific to the domain. In the urban water domain, we define Pipe flow sensors, and pump pressure sensors. Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Sewer rising mains case study We got hold of data from a recent pipe burst in Victoria, where a pipe failure occurred close to a creek which fed the port phillip bay. It was undetected for more than a week. - only detected when a passer-by walking his dog noticed a significant change in the colour of the creek. The ramification for the water service provider was for them to send trucks in to pump out sewerage until the pipe was fixed, which wasn’t cheap. - the EPA threatened with hefty fines. The graph in this slide show the pipe failure event documented with the red line. The blue line shows the flow rate in litres per second. - notice an upward trend as it built up to the failure event and then sustained a higher than normal flow rate... for a long time. Example event: Flow rate > 100 l/s Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Event rule definition instances Rule ID Observed property Value constraint Feature of interest Observed By (Sensor) 1 Flow 2 > 100 l/s 3 Pipe A 4 Pipe Sensor A-1 5 Using the definition of a ValueConstraint Event rule class, we can start to describe event rule instances with varying levels of detail. This allows us to describe event rule instances with varying levels of detail. We can describe an event of interest as all observations observing flow. Flow greater than 100 l/s Flow greater than 100 l/s on Feature of interest Flow greater than 100 l/s on a sensor Flow greater than 100 l/s on both sensor and feature. Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Fusing real-time events with domain knowledge Real-time data Sensor Network Notifications Event of Interest Query knowledge base (domain knowledge) e.g. Populate knowledge base with parameterised historical pipe failure data. Infer likelihood of pipe failure based on physical attributes and known operating environment Knowledge Base In many domains, there exists a large amount of expert knowledge, which is quite static compared to the dynamic real-time observation data. So in our work, we considered incorporating such expert knowledge as an extension to the event detection based on real-time sensor data. Fusing real-time event notifications with domain knowledge. - Allows us to perform context-sensitive event detection. Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Modelling the feature of interest – pipe materials Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Modelling the burst events Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Event detections using dynamic and static info > 200 PSI Notification: Location: Pipe X Risk of burst: LOW + (Static) Pipe material is PVC and Risk level of pipe is A (good) Start to create event rules that leverage static domain knowledge captured in the domain rules For example, fusing real-time pump pressure sensor observations, with knowledge about the state of the pipe that it is pumping into. Event here : pressure exceeds 200 p.s.i. && pipe is relatively ok Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Event detections using dynamic and static info > 200 PSI Notification: Location: Pipe X Risk of burst: HIGH + (Static) Pipe material is PVC and Risk level of pipe is E (bad) Event here : pressure exceeds 200 p.s.i. && pipe is in bad shape Start to create event rules that leverage static domain knowledge captured in the domain rules Context-sensitive event detection Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Sensor Infrastructure Event Dashboard Notification clients Initialise sensor network and sensor middleware Map sensors Query Notifications Define event constraint Visualise Notifications Deploy event constraint Integrate with Notification systems Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Ontology-enabled User Interface Future work Email / SMS Database Notification handling Messaging queue systems Attaching metadata based on event rule semantics More complex events Event semantics Incorporate processing-filters User studies to evaluate the user interface Deployments on actual sensor networks Existing alert systems Execute workflow A Smoothing function Event B Ontology-enabled User Interface Sensor Network Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Questions? Performing event detection over real-time sensor data using ontology-driven approaches | Jonathan Yu

Thank you Land and Water Jonathan Yu Paul Davis Brad Sherman Research Software Engineer Research Scientist t +61 3 9252 6440 t +61 3 9252 6310 e jonathan.yu@csiro.au e paul.davis@csiro.au e Brad.Sherman@csiro.au w www.csiro.au/clw Land and Water ICT Centre Scott Gould Kerry Taylor Donavan Marney Research Projects Officer Principal Research Scientist Research team leader t +61 3 9252 6103 t +61  2 6216 7038 t +61 3 9252 6585 e scott.gould@csiro.au e kerry.taylor@csiro.au e donavan.marney@csiro.au w www.csiro.au/clw w www.csiro.au/ict Land and Water