**2. Electromechanical impedance based concrete strength estimation**

As already described, in order to manage effectively the construction process of the concrete structures, online monitoring of the concrete strength-development is strongly required. To end this, this study employs the electromechanical impedance-based structural health monitoring (SHM) methodology because the electromechanical impedance can represent the mechanical properties of a host structure. The strength variation developed during the curing process can be observed throughout the electromechanical impedance measurements.

#### **2.1 Strength development due to concrete curing process**

Concrete achieves its strength through a hydraulic process known as hydration. With the addition of the correct amount of water, the cement gels into a paste that glues sand and aggregates together to form hardened concrete. The curing of concrete involves maintaining a proper moisture vapor transmission rate (2% mvtr) immediately after concrete placement

accessibility of the impedance method for in-field measurements. Lynch et al. (2004) designed a wireless active sensing unit to monitor civil structures, which was constructed of off-the-shelf components and had the ability to command active sensors and actuators from a computational core combined with wireless transmission and sensing circuits, embedded algorithm to process the acquired data, and structural status broadcasting. Grisso and Inman (2005) designed a DSP (Digital Signal Processor) based prototype to provide wireless assessment of thermal protection systems. It was able to directly detect damages by analyzing variations of the electrical impedance of PZT sensors bonded to the structure. The obtained impedance signals were compared with the pre-stored baseline and a statistical damage index was calculated. Mascarenas et al. (2007) proposed a wireless sensor node which consists of a miniaturized impedance measuring chip, a microprocessor, and a radiofrequency identification (RFID) module. Low cost impedance measuring chip actuated the structure through a PZT and measured the structural impedance response, and RFID module delivered the diagnostic result to a base station. Park S. et al. (2009) improved the wireless sensor node of Mascarenas et al. (2007) by adding two multiplexer IC chips for 16 channels for the cases of multiple sensors in a small region and by embedding signal processing algorithms in the microcontroller unit (MCU) for both structural damage identification and sensor self-diagnosis. Research groups at Los Alamos National Lab recently developed a compact impedance-based wireless sensing device (WID3) for low power operation (Overly et al. 2007, 2008), which requires around 60 mW of power to operate. Here, a wake-up capability was combined for low power operation. Taylor et al. (2009a, b) extended the capability of the WID3 by implementing a module with lowfrequency A/D and D/A converters to measure low-frequency vibration data for multiple SHM techniques. It is wirelessly triggered by a mobile agent for use in the mobile-hostbased wireless sensing network. Min et al. (2010) developed a multi-functional wireless sensor node integrating piezoelectric actuating, sensing, signal processing, temperature

In this context, this study presents a series of efforts to confirm the applicability of the electromechanical impedance technique using both wired and wireless systems for online

As already described, in order to manage effectively the construction process of the concrete structures, online monitoring of the concrete strength-development is strongly required. To end this, this study employs the electromechanical impedance-based structural health monitoring (SHM) methodology because the electromechanical impedance can represent the mechanical properties of a host structure. The strength variation developed during the curing process can be observed throughout the electromechanical impedance

Concrete achieves its strength through a hydraulic process known as hydration. With the addition of the correct amount of water, the cement gels into a paste that glues sand and aggregates together to form hardened concrete. The curing of concrete involves maintaining a proper moisture vapor transmission rate (2% mvtr) immediately after concrete placement

monitoring on the strength-development during the curing process of concrete.

**2.1 Strength development due to concrete curing process** 

**2. Electromechanical impedance based concrete strength estimation** 

compensating, and energy harvesting modules.

measurements.

as well as throughout the ensuing period of approximately 28 days. Fig. 1 shows a typical strength development curve (Shariq et al., 2010).

Fig. 1. Typical Strength Development Curve

#### **2.2 Electromechanical impedance modeling**

The electromechanical impedance-based SHM techniques have been developed as a promising tool for real-time structural damage assessment on critical members of large structural systems (Park G. et al., 2003, Koo et al., 2009, Taylor et al., 2009a, 2009b, Mascarenas et al., 2009). They make use of piezoelectric sensors such as piezoceramic (PZT) and macro-fiber composite (MFC) patches, which form a collocated sensor and actuator, often referred to as a self-sensing actuator (Giurgiutiu, 2007). The basis of this active sensing technology is the energy transfer between the actuator and the host mechanical system. If a PZT attached on a structure is driven with a sinusoidal voltage, it causes the local area of the structure to vibrate (the converse piezoelectric effect). And the structural response causes an electrical response in the PZT (the direct piezoelectric effect). Liang et al. (1996) first proposed a one-dimensional analytical model of this setup as in Fig. 2, and showed that the electrical admittance (inverse of the electrical impedance), Y(ω), of a PZT is directly correlated to the local mechanical impedance of the host structure, Zs(ω), and that of a PZT patch, Za(ω), in most applications as

$$Y(\alpha) = G(\alpha) + jB(\alpha) = j\alpha \mathbb{C} \left( 1 - \kappa\_{31}^2 \frac{Z\_s(\alpha)}{Z\_s(\alpha) + Z\_a(\alpha)} \right) \tag{1}$$

where G is the conductance (real part); B is the susceptance (imaginary part); C is the zeroload capacitance of a PZT; and κ31 is the electromechanical coupling coefficient of a PZT. Given that the mechanical impedance and the material properties of the PZT stay constant, the equation shows that a change in the structure's mechanical impedance directly results in a change in the electrical impedance measured by the PZT. Since damages cause a change in the structure's local mass, stiffness, or damping properties and consequently its mechanical impedance, the structure's mechanical integrity can be assessed by monitoring the PZT's electrical impedance. It should be noted that the admittance function, Y(ω), is a complex number. Bhalla et al. (2002) demonstrated that the real part of the measured admittance is more sensitively changed due to the structural damage condition as compared to the

Ubiquitous Piezoelectric Sensor Network

Fig. 3. Schematic diagram of a self-sensing circuit

**3. Development of wireless impedance sensor nodes** 

**3.1 Subsystems of wireless impedance sensor nodes** 

functional wireless sensor node developed by Min et al. (2010) was utilized.

(UPSN)-Based Concrete Curing Monitoring for u-Construction 79

The recent advances in wireless online monitoring integrating actuation and sensing, onboard computing, and radio-frequency (RF) telemetry improved the accessibility of the electromechanical impedance method for in-field measurements. In this study, the multi-

To measure the electromechanical impedances, impedance analyzers such as HP4294A are conventionally used. However, they are not quite suitable for field applications to online SHM because they are bulky (approximately 25 kg) and expensive (approximately 40,000 USD). Thus, research on the impedance based SHM technique trends toward development of self-contained sensors and wireless active sensor nodes with all required functions including actuating/sensing, data processing, damage assessments and sensor selfdiagnostics on the sensor board as well as power management with energy harvesters. Recently, Analog Devices© developed an integrated impedance converters, AD5933 (www.analog.com). It is equipped with a 12-bit analog-to-digital converter (ADC), a digitalto-analog converter (DAC) and a discrete Fourier transform (DFT) functionality. The frequency generator allows an external complex impedance with range of 100 Ω to 10MΩ to be excited with a known frequency of up to 100 kHz. AD5933 is just of a penny size, thus it provides a solution for self-contained miniaturized impedance measuring. Therefore, AD5933 has been used as a core component in developing a wireless impedance sensor node for SHM applications (Mascarenas et al. 2007, 2009, Overly et al. 2007, 2008, Taylor et al. 2009a, b, Park S. et al. 2009, Min et al. 2010). The wireless sensor node, proposed in this study, has extended the previous researches for multifunctional and environment-friendly uses in the impedance-based SHM (Mascarenas et al. 2007, Park S. et al. 2009). It was designed by adding: (1) optimal arrangement of each chip for low power consumption, (2) energy harvester equipped with solar panels, (3) peer-to-peer communication by using a RF transceiver of CC2420, which enables to construct the ubiquitous sensor network, (4) internal algorithms for operations, which are optimized by using microcontroller-dependent

imaginary part. On the other hand, Park G. et al. (2006) found out that the imaginary part can be more effectively used for piezoelectric sensor self-diagnosis.

Fig. 2. 1-D Model used to derive electromechanical admittance of a PZT bonded to a structure (Liang et al. 1996)

#### **2.3 Self-sensing based wired impedance measurement method**

In this study, impedance measurement systems are based on a self-sensing technique, as shown in Fig. 3. A self-sensing circuit as described in Fig. 3 is suitable for use in cast-in-site concrete because it is inexpensive and has sufficient accuracy to measure the development of strength, even though the impedance signal is less accurate than other impedance measurement methods. The self-sensing circuit board consists of a single PZT patch, and a voltage divider, such as a resister or capacitor to acquire the output voltage.

The impedance is measured in three steps based on the self-sensing circuit as follows: (1) the input voltage (Vi ) generated from an arbitrary waveform generator (AWG) is applied to the free surface of the PZT sensor; (2) the output voltage from the self-sensing circuit (Vo) is measured using a digitizer (DIG); and (3) the admittance, which is the inverse of impedance, is derived from the input voltage, output voltage, and reference capacitance (Cr). The output voltage from the self-sensing circuit consists of the input voltage and mechanical response of the structure (Vr). Although the amplitude of the mechanical responses of the structure is small enough to ignore, the output voltage is dominated by the input voltage and can be approximated as follows (Lee and Sohn, 2006)

$$V\_o(t) = \frac{C\_p}{C\_p + C\_r} \left[ V\_i(t) + V\_p(t) \right] = \frac{C\_p}{C\_p + C\_r} V\_i(t) \tag{2}$$

where Cp is the PZT capacitance and Cr is the reference capacitance of the self-sensing circuit. From Eq. (2), the impedance of structure (*Z*) is derived as follows:

$$Z(o) = \frac{V\_p(o)}{I(o)} = [i\nu C\_p]^{-1} = \left[ i\alpha C\_r \left( \frac{V\_o(o)}{V\_i(o) - V\_o(o)} \right) \right]^{-1} \tag{3}$$

Thus, the impedance explained at Eq. (1) could be measured by Eq. (3) using the self-sensing circuit displayed in Fig. 3.

imaginary part. On the other hand, Park G. et al. (2006) found out that the imaginary part

Fig. 2. 1-D Model used to derive electromechanical admittance of a PZT bonded to a

In this study, impedance measurement systems are based on a self-sensing technique, as shown in Fig. 3. A self-sensing circuit as described in Fig. 3 is suitable for use in cast-in-site concrete because it is inexpensive and has sufficient accuracy to measure the development of strength, even though the impedance signal is less accurate than other impedance measurement methods. The self-sensing circuit board consists of a single PZT patch, and a

The impedance is measured in three steps based on the self-sensing circuit as follows: (1) the input voltage (Vi ) generated from an arbitrary waveform generator (AWG) is applied to the free surface of the PZT sensor; (2) the output voltage from the self-sensing circuit (Vo) is measured using a digitizer (DIG); and (3) the admittance, which is the inverse of impedance, is derived from the input voltage, output voltage, and reference capacitance (Cr). The output voltage from the self-sensing circuit consists of the input voltage and mechanical response of the structure (Vr). Although the amplitude of the mechanical responses of the structure is small enough to ignore, the output voltage is dominated by the input voltage and can be

> ( ) () () ( ) *p p o i p i p r p r C C Vt Vt V t Vt C C C C*

where Cp is the PZT capacitance and Cr is the reference capacitance of the self-sensing

<sup>1</sup> ( ) ( ) () [ ] ( ) () () *<sup>p</sup> <sup>o</sup> p r*

*I V V*

Thus, the impedance explained at Eq. (1) could be measured by Eq. (3) using the self-sensing

 ω

<sup>−</sup> = = <sup>−</sup>

+ + (2)

*i o*

ω

ω

 ω 1

−

(3)

= +

*<sup>V</sup> <sup>V</sup> <sup>Z</sup> iwC i C*

circuit. From Eq. (2), the impedance of structure (*Z*) is derived as follows:

ω

ω

**2.3 Self-sensing based wired impedance measurement method** 

voltage divider, such as a resister or capacitor to acquire the output voltage.

can be more effectively used for piezoelectric sensor self-diagnosis.

structure (Liang et al. 1996)

approximated as follows (Lee and Sohn, 2006)

ω

circuit displayed in Fig. 3.

Fig. 3. Schematic diagram of a self-sensing circuit
