Date of Award


Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy


Department of Mechanical and Manufacturing

First Advisor

Dr. Ger Kelly


Structural Health Monitoring (SHM) is a rapidly growing sector of the construction industry. Automated monitoring methods arc mainly limited to global monitoring and the application of local monitoring methods has not gained widespread application due to the fact that currently technology requires that the systems are wired to data acquisition devices. This thesis presents a fully embeddable wireless sensor node for the local monitoring of structural concrete. It presents the first reported wireless embedded system implementation of the electromechanical impedance method using both resonant and antiresonant sensors for this purpose.

The first section of this thesis examines the feasibility of embedding wireless sensors into concrete. Wireless sensors (motes) with temperature and humidity measurement capabilities were designed with onboard communication using the 433 MHz ISM band. A package was designed to allow the sensor to be embedded into concrete to protect the sensor and its electronics from the aggressive environment within concrete. A humidity chamber was used to optimise the design of the package to ensure that it did not affect the sensor's accuracy. An extensive testing procedure was designed replicating onsite conditions. Test results suggest that it is possible to deploy an embedded sensor and transmit live data wirelessly to a data acquisition system located outside the concrete. A temperature and moisture sensor applied to the mote r gave an indication of the ability of the sensor to transfer live data f om within the concrete. Preliminary results show that steel backed formwork reduces the transmission distance of the sensors to 3.5m which extended to over Sm when the formwork was removed. It was also found that the moisture content in the early stages had a large impact on the transmission of data. Further analysis of the transmission of data from within the concrete was carried out by increasing the reinforcement mesh density and depth of embedment and it was found that the transmission distance during the early stages of curing was reduced to less than I m.

An extensive review of the current monitoring methods and their applicability to the designed sensing platfonn resulted in the decision to advance the electromechanical impedance (EMl) method to allow it to be applied to the designed mote.

The advancement of the EMI method allows it to be embedded into freshly poured concrete. The designed sensing system was embedded into a concrete test cube at first pour and testing has shown that the antiresonant frequency shift is the most suitable parameter for monitoring the condition of the concrete. The antiresonant shift development was shown to correlate well with the compressive strength development. Results also show that the embedded EMI method is sensitive to changes in the condition of the concrete including the removal of formwork. The ability of the sensor to sense loads was examined under two conditions - continuous loading and variable loading. In conjunction with the antiresonant frequency shift, monitoring of maximum resistance was found to be a good indicator of the condition of the cement aggregate interface. Tests also demonstrated that the packaging process did not significantly degrade the piezoelectric performance of the sensors and the orientation of the sensor with relation to its polarization during testing did not significantly affect its response.

The existence of an impedance analysis IC (AD5933 from Analog Devices) removes the requirement for an external impedance analyser which allows the EMI method be incorporated into the wireless mote. The piezoelectric sensors were specifically designed to be applicable with the AD5933 and an extensive testing procedure was designed to determine its ability to be incorporated into the wireless sensor. It was determined that the AD5933 requires continuous calibration and a method of automatically calibrating the chip is presented. This was a critical development in the design of a fully automated system. The accuracy of the impedance readings taken using the AD5933 was compared with a HP4 I 92A impedance analyser and it was found that the AD5933 can be used as a replacement for the HP4 l 92A. The results demonstrate that the designed wireless system is sufficiently reliable to not only monitor and transmit the evolution of strength but can also continue to monitor in-service condition after the concrete has cured.

Access Level