Infrastructure for Smart Cities Bridging research and production - - PowerPoint PPT Presentation

www.data61.csiro.au

Infrastructure for Smart Cities

Bridging research and production

Sandy Taylor | Software Engineer Ben Barnes | Software Engineer September 2016

Section 1: Background The need for a solution

The Sydney Harbour Bridge needs:

• $15,000,000 maintenance p/a
• Almost 100 full-time workers
• 15 full-time painters
• 30,000 L of paint per coat

Where do we come in?

• Provide early warning of maintenance needs for the road deck

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Bridges Are Needy

3 |

Harbour Bridge Paint

• 800 jack arches support the

road deck

• Aging infrastructure, needs to

be replaced eventually

• Need to extend the life of the

deck without a significant increase in maintenance costs

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

The Road Deck

4 |

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Inspection & Maintenance

5 |

• Two year visual

inspection cycle

• Inspection manual is

exhaustive

• Similar for metalwork,

paintwork…

• Access is difficult. Crawl

spaces, gantry crane + scaffold required in places

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Inspection & Maintenance

6 |

• Road surface cracks often the

first indication of damage

• But too late, cracks are a

result of underlying structural damage

• Need to detect damage

(fractures) in jack arch before it becomes serious

Mortal risk, but also economic:

• A crash in March shut down three lanes for around two hours
• Estimated cost to the economy: $16,000,000

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Failure Is Costly

7 |

Jan 2016 Sept 2016 Mar 2016

Section 2: The Solution How did we get there?

• Continuously monitor 800 structural joints?
• Provide early warning of maintenance needs?
• Augment and direct inspection routines?
• Extend service life of bridge deck?
• Without a significant increase in maintenance costs?

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Can We Do Better With Data?

9 |

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Structural Health Monitoring

10 |

I I I I I I

2400 sensors installed

Data Collection Data Processing Presentation, Alerting & Aggregation

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Structural Health Monitoring

11 |

• A problem in three parts:

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor

Data Challenges & Constraints

12 |

• A production, real time system that also caters for future research

and analysis techniques

• Large data volumes (100s of Gb per day)
• Limited uplink capacity from site – reduce network storage and

load (store and forward approach)

• Real time decisions and visualisation – immediate notification of

structural defects

• Timeframe: milliseconds to decades

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 13 |

Data Collection

Sensor Node Sensor Node … Sensor Node Sensor Node … Gateway Gateway ADSL Modem Server

Designed For Research

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 14 |

• The SHM field changes rapidly
• Structures don’t (few chances to modify the system)
• Long-running project, have to adapt

Data Processor Data Acquisition Data Processor Data Processor Storage

Research To Production

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 15 |

Raw data MATLAB/ R

Collaboration with Engineering

C++ Data Processor Packaged by Ops

Live data to frontend

Data Analysis Methods

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 16 |

Several methods are either in the process of being developed or have been deployed on the bridge:

• Defect/Damage Detection
• Comparative Vibration
• SVM classifiers
• Time Series Analysis
• Population clustering analysis
• Condition Assessment
• Multi-scale (localised vs global)
• Operational modal analysis
• Classifiers derived from dynamic FEA models
• Vehicle Tracking
• Tracking and classification
• Impact force reconstruction from sensor data
• Predictive Analysis
• Predicting relative likelihood of failure based on multiple factors
• Data fusion technique

We will cover a few of these damage detection methods

First Step -Detecting Events

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 17 |

• Only keep data based on

traffic events

• Threshold based detection
• Reduces data volume to 1/30th
• Better signal to noise ratio,

more consistent samples

• Filter out smaller vehicles

Heuristic Approach

• Intuition: structural member should behave as a rigid body
• Sensors should show same response to disturbance
• Cracks allow sections to move independently
• Can use cross-correlation to measure similarity
• Scaled based on vibration energy: more vibration, more damage

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 18 |

Joint 1 Joint 2 Joint 3 Joint 4 Joint 5 Joint 6

Heuristic Approach Problems

• Vibration varies significantly over the bridge
• Proof-of-concept nodes weren’t representative of this
• Could tune parameters per sensor, but not a scalable solution
• There must be a better way!

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 19 |

A C B

SVM & FFT

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 20 |

• Unsupervised single class

Support Vector Machine (SVM)

• Model-free, not tied to

structural details

• Looks for any change, not just

damage

• Training is a challenge
• Still some threshold

adjustment

50 100 150 200 250 300 350 400 450

• 2
• 1.5
• 1
• 0.5

0.5 Decision values Test event index Joint 5 Joint 4

Degraded Normal traffic “events”

Dynamic Event Threshold

• Looks for a change in variance
• Threshold is now a ratio: event amplitude

noise amplitude

• Event detection is decoupled from units of measurement
• No more manual tweaking

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 21 |

Sensor Faults

• Sudden, unexplained health warnings
• How do we reduce false positives?

Consider environmental factors:

• Sensor detachment
• Water ingress
• Temperature sensitivity

Build in measures of confidence

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 22 |

Missing potting

Visual Presentation

• How to clearly convey status?
• Historical analysis & aggregation
• Simple up front, more

information if desired

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 23 |

Presentation & Aggregation

Grouped Aggregation

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 24 |

Grouped Aggregation & Alerting

Alerting

Sensors ML Decision Values Heuristic Score Overall Jack Arch Health H M L M L M L S 1 S 2 S 3 Structural Component System Component Virtual Sensor Bay (Group)

Validation Of Our Method

• Whilst there hasn’t been any actual damage:

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 25 |

Section 3: Future work, takeaways

• Encourage collaboration between research and engineering

throughout the project

• Divide the problem into collection, processing and presentation
• Separate logical processing topology from physical network
• Build in confidence metrics from the beginning

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 27 |

What have we learned?

• Decouple processing from platform – node, gateway, cloud
• Improve confidence estimates
• Strain gauges - fatigue / life cycle
• FEA model updating

Infrastructure for Smart Cities | Ben Barnes, Sandy Taylor 28 |

Future Development