ST-17AB : Optimal sensor placement for nonlinear dynamic model updating and response prediction of civil engineering structures subjected to future uncertain dynamic loadings
This project is undertaken to ascertain the precise points where sensors, to collect system vibration data, can be placed on a structure so that the greatest and most reliable yield of information can be extracted. A statistical methodology is adopted to compute inevitable uncertainty present in the...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2016
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/67422 |
| Summary: | This project is undertaken to ascertain the precise points where sensors, to collect system vibration data, can be placed on a structure so that the greatest and most reliable yield of information can be extracted. A statistical methodology is adopted to compute inevitable uncertainty present in the model parameters as well as any uncertainty arising due to model assumptions. This is done by estimating a range of values approximately close to the ‘true’ value using Bayesian statistical procedure. Information entropy theory dictates that a measure of these uncertainties can predict the most optimal location for the sensor. Thus, information entropy is calibrated for a heuristic model to determine a minimal error from a unique combination across a set of sensor placement combinations. The calibration is used to handle small uncertainty in the initial inputs fed into the dynamic model. The amended model can then be used accurately to make reasonable predictions of the behaviour of the system when it is subjected to a future uncertain dynamic loading. Moreover, the project intends to explore if output errors are reduced for placing more than one sensor at its optimal location |
|---|