multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

25
Multi-sensor data fusion system for enhanced analysis of deterioration in concrete structures Othman Sidek and S.A.Quadri Collaborative µ-electronic Design Excellence Centre Universiti Sains Malaysia Paper presented at PIERS Conference: Progress In Electromagnetic Research Symposium Proceedings, Suzhou, China, Sept. 12-16, 2011, 637

Upload: sayed-abulhasan-quadri

Post on 18-Dec-2014

999 views

Category:

Education


0 download

DESCRIPTION

 

TRANSCRIPT

Page 1: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Multi-sensor data fusion system for enhanced analysis of deterioration in concrete structures

Othman Sidek and S.A.Quadri

Collaborative µ-electronic Design Excellence CentreUniversiti Sains Malaysia

Paper presented at PIERS Conference:Progress In Electromagnetic Research Symposium Proceedings, Suzhou, China, Sept. 12-16, 2011, 637

Page 2: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Data fusion Data-fusion is a problem-solving technique based on the idea of integrating many answers to a question into a single; best answer.

Data fusion is defined as process of combining inputs from various sensors to provide a robust and complete description of an environment or process of interest.

It is multilevel, multifaceted process dealing with the automatic detection, association, correlation , estimation, and combination of data and information from single and multiple sources.

Page 3: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Data fusion

“Properly said, fusion is neither a theory nor a technology in its own. It is a concept which uses various techniques pertaining to information theory, artificial intelligence and statistics”

[1] Dave L. Hall and James Llinas, “Introduction to Multisensor Data Fusion”, IEEE , Vol. 85, No. 1, pp. 6 – 23, Jan 1997.

1

Page 4: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Multisensor data fusion provides significant advantages over single source data. The use of multiple types of sensors plays an important role in achieving reasonable accuracy and precision.

Page 5: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

A novel integrated multi sensor data fusion approach in structural health monitoring is proposed.

The study concerns to find a simple and affordable monitoring strategy for Alkali-aggregate reaction (AAR), which is one of the root causes for structural deterioration in concrete.

Page 6: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

AAR Expansion in Concrete Structure

Alkali-aggregate reaction is a term mainly referring to a reaction which occurs over time in concrete between the highly alkaline cement paste and non-crystalline silicon dioxide.

It produces a gel that expands when in contact with the moisture in concrete, and can lead to the development of high tensile stresses and cracking of concrete.

AAR is a serious problem that adversely affects the durability and integrity of concrete structures.

Page 7: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

AAR Expansion in Concrete Structure

AAR GEL

Page 8: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

AAR Expansion in Concrete Structure

AAR GEL

Page 9: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

AAR Expansion in Concrete Structure

AAR GEL

Page 10: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

AAR Expansion in Concrete Structure

AAR GEL

Page 11: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Simulating AAR expansion within reasonable laboratory timescale.

Because of the lengthy lead time required to evaluate adequately aggregate sources for potential alkali-aggregate reaction and expansion we use standard accelerated method to accomplish evaluation process.

National Building Research Institute (NBRI) standard testing method is employed to accelerate AAR expansion on four samples which are prepared with different level of alkali concentrations.

Page 12: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Four samples are with different alkali concentrations :

1. Non reactive

2. Marginal reactive

3. Moderately reactive and

4. Very reactive

Hypothetical graph showing expansions (in percentage) in various samples

Page 13: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Hypothetical graph showing expansions (in percentage) in various samples

02468

101214161820

1 2 3 4

Non Reactive Marginal ReactiveModerate ReactiveVery Reactive

Page 14: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Heterogeneous sensor system

Different sensor systems are used at surface and internal level.

• Acoustic sensor system (Pulse-echo Method)

Electro-mechanical system(LVDT Sensors)

• Optical systems

(CCD Camera)

• Embedded sensors (PZT Piezo electric sensors)

Surface Damage Level

Internal Damage Level

Page 15: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Figure showing experimental setup

Page 16: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Features extraction

Data obtained from the various sensors is subjected to feature extraction process.

Feature extraction is a process of removing redundant data and extracting informative data from large set of data.

Page 17: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Multi sensor Data Fusion

The challenge for data fusion is to merge heterogeneous data from acoustic system, electro-mechanical system, optical system and embedded sensors in an efficient way to increase the accuracy and consistency of the acquired data.

Features extracted from heterogeneous sensors are fed to Decentralized Kalman filter.

Decentralized Kalman filter has long been regarded as the optimal solution to many tracking and data prediction tasks and a robust means to fuse heterogeneous data.

Page 18: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Decentralized Kalman filter .The Kalman filter uses a system's dynamics model (i.e., physical laws of motion), known control inputs to that system, and measurements (such as from sensors) to form an estimate of the system's varying quantities (its state) that is better than the estimate obtained by using any one measurement alone. As such, it is a common sensor fusion algorithm.

If raw data obtained sensors is fused it is Centralized Kalman filter.

If processed data (features) are fused is called Decentralized Kalman filter .

If raw data and features are fused it is called Hybrid Kalman filter

Page 19: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Data flow diagram

Page 20: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Artificial neural network (ANN)

The fused global estimates and individual source estimates are fed to artificial neural network (ANN), which characterize and quantify the level of damage.

Page 21: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Multi-layer perceptron (MLP) network is used which is having three layers:

• Input layer

• Hidden layer

• Output layer

ANN is ideally suited to identify non linear system dynamics

A well trained ANN network characterizes and quantifies level of deterioration of the specimen under study.

Page 22: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Expected results :Establishing correlation among surface damage level, internal damage level and the amount of gel concentration in the structure.

Improved accuracy using Multi sensor data fusion

Hypothetical graph showing damage level using single data source sensor system and data fusion system

Hypothetical graph showing correlation between external and internal damage levels.

Page 23: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

0

2

4

6

8

10

12

14

16

1 2 3 4

Acoustic Sensor

LVDT Sensor

CCD Camera

PZT EmbeddedSensorData fusion

The root-mean-square deviation (RMSD) is used as the damage indexing

Hypothetical graph showing damage level using single data source sensor system and data fusion system

Page 24: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Hypothetical graph showing correlation between external and internal damage level

0

2

4

6

8

10

12

14

Extern

al 10 20 30 40 50 60 70 80 90

Line 1

Page 25: Multi sensor data fusion system for enhanced analysis of deterioration in concrete structures

Thanks