**3. Structural health monitoring**

The desire to move from current periodic structural maintenance to a more cost-effective condition-based maintenance (CBM) philosophy to ensure integrity of critical structures has fostered research and development activities to develop SHM solutions. SHM using UGW has found a variety of practical applications for elongated engineering structures including pipes, plates, ship hulls, rails and cables, because of its inherent long range propagation [42].

#### **3.1 Monitoring system design and architecture**

The operational requirements of SHM systems for pipelines and storage tanks are tabulated in **Table 1**. Currently, costly acquisition of SHM data is only justifiable for structures with significantly high failure consequences. Transducer technologies play a critical role in the design of SHM system as they are permanently installed on the structure and required to repeatedly transmit excitation signals and analyse the received responses.

The transducers may need to be attached in environmentally hostile, safetycritical or difficult-to-access areas and therefore they should be designed to perform reliably under prolonged exposure to harsh environmental and operational conditions (EOCs). Therefore, low cost and reliability are the two main factors to consider when designing a SHM sensor system for pipelines and storage tanks. One

**63**

*Monitoring of Critical Metallic Assets in Oil and Gas Industry Using Ultrasonic Guided Waves*

**Operational requirement Pipelines Storage tank floor** Operating temperature −10 to 150°C −10 to 60°C Signal to noise <6 dB <6 dB Operating frequency range 20–100 kHz Resonant frequency Transmission range Up to 100 m 30–100 m

Frequency of data collection Once a week Once a week (depending on the

Baseline subtraction Pattern recognition

Wave mode selection T(0,1) S0 and SH0

Data acquisition Pulse-echo/pitch-catch Pitch-catch

Signal processing Thresholding/outlier analysis

condition)

Tomography Pattern recognition Neural networking Baseline subtraction

cost-effective approach is to use a single pulser-receiver and PC to collect monitoring data from multiple sensor locations at junction points, which can be located in easily accessible location. This significantly reduces cost of repeated access and of

Current state-of-the-art in pipeline monitoring solutions includes corrosion coupons, acoustic emission and magnetostrictive sensors, flexible eddy current arrays, flexible ultrasonic transducers, guided wave sensors, impedance spectroscopy, microwave backscattering and fibre optic sensors. A review of these monitoring technologies can be found in [43]. Corrosion sensors based on electrical resistance and electromechanical impedance spectroscopy can only provide coverage over a small area and are not suitable for non-uniform corrosion artefacts such as pits. Recent advances in acoustic emission (AE) sensor technology [44, 45] have led to corrosion detection and monitoring solutions where acoustics signals from micro-fractures and delamination of the oxide are analysed. These emissions release much less energy than emission from crack growth where AE has shown great potential. In low noise environments AE could be used to detect signals from corrosion with tens of metres range using monitoring frequencies of tens of kilohertz. However, in a live plant, high process noise requires several hundred kilohertz of monitoring frequencies and coverage is limited <0.5 m and requires complex signal processing. For this reason, AE is limited for this application. Magnetostrictive sensor (MsS) is another technology for pipeline monitoring first developed and patented by SwRI® [46]. They have lower power output compared to piezoelectric transducer, however, recent advancements have reported significant improvements in their power output, sensitivity and flaw characterisation [47]. Piezoelectric sensing offers the most promising solution due to their stability, reliability, and cost-effectiveness as described in Section 2.2. This has enabled the development of several SHM solutions. Guided Ultrasonic Ltd. offers one such monitoring system gPIMS [48] and this system's stability and defect detection capabilities have been demonstrated [49] at temperatures up to 90°C. Another example is the system developed by the authors and its installation,

*DOI: http://dx.doi.org/10.5772/intechopen.83366*

the overall system.

**Table 1.**

**3.2 Monitoring system for pipelines**

*Operational requirements of SHM systems.*

*Monitoring of Critical Metallic Assets in Oil and Gas Industry Using Ultrasonic Guided Waves DOI: http://dx.doi.org/10.5772/intechopen.83366*


#### **Table 1.**

*Advances in Structural Health Monitoring*

over the frequencies in the narrow band.

**3. Structural health monitoring**

because of its inherent long range propagation [42].

**3.1 Monitoring system design and architecture**

and avoid dispersion while propagating long distances. The centre frequency of these signals are chosen based on the desired wave mode to achieve low dispersion

locating both internal and external defects without disrupting operation.

There are two modes of operation: pulse-echo and pitch-catch. Pulse-echo mode is more common and utilises the same transducers to excite the UGW and receive the reflected signals as illustrated in **Figure 7**. Pitch-catch mode uses two sets of transducers, one to excite the UGW and the other to receive, and is only used if high resolution or a high inspection range are required. As the UGW propagates in the structure, a proportion of the energy contained in the propagating wave front will be reflected when an acoustic impedance change occurs at a feature or discontinuity in the structure. This enables full coverage of the cross section of the plate or pipe, detecting and

Since the initial developments of GWT of pipes in late 1990s, several studies have been carried out to understand the interaction of T(0,1) and L(0,2) guided wave modes with pipe features (flanges and pipe supports) [37] and defects [35], and the effect of different defect characteristics and excitation frequencies has also been reported [38, 39]. This has led to definitions and standards for GWT instrumentation, data collection and analysis in ISO 18211:2016 [40]. When an axisymmetric mode is incident on an axisymmetric pipe feature such as a uniform weld or a flange, axisymmetric modes are reflected. With a non-axisymmetric feature such as corrosion, a non-axisymmetric wave will also be reflected back to the transducer array. The presence and axial location of defects can thus be determined by analysing these reflections and their time of arrival. Although the L(0,2) mode has shown ~2.5 times more flaw sensitivity compared to T(0,1) [34], it is difficult to excite in pure form and requires complex signal processing due to its dispersive nature. It is also affected by fluid in the pipe, so the torsional mode is more commonly used in practice. GWT using T(0,1) is most effective on straight sections achieving several tens of metres of inspection range but recent studies have evaluated its performance on bends [41].

The desire to move from current periodic structural maintenance to a more cost-effective condition-based maintenance (CBM) philosophy to ensure integrity of critical structures has fostered research and development activities to develop SHM solutions. SHM using UGW has found a variety of practical applications for elongated engineering structures including pipes, plates, ship hulls, rails and cables,

The operational requirements of SHM systems for pipelines and storage tanks are tabulated in **Table 1**. Currently, costly acquisition of SHM data is only justifiable for structures with significantly high failure consequences. Transducer technologies play a critical role in the design of SHM system as they are permanently installed on the structure and required to repeatedly transmit excitation signals and analyse the

The transducers may need to be attached in environmentally hostile, safetycritical or difficult-to-access areas and therefore they should be designed to perform reliably under prolonged exposure to harsh environmental and operational conditions (EOCs). Therefore, low cost and reliability are the two main factors to consider when designing a SHM sensor system for pipelines and storage tanks. One

**62**

received responses.

*Operational requirements of SHM systems.*

cost-effective approach is to use a single pulser-receiver and PC to collect monitoring data from multiple sensor locations at junction points, which can be located in easily accessible location. This significantly reduces cost of repeated access and of the overall system.

#### **3.2 Monitoring system for pipelines**

Current state-of-the-art in pipeline monitoring solutions includes corrosion coupons, acoustic emission and magnetostrictive sensors, flexible eddy current arrays, flexible ultrasonic transducers, guided wave sensors, impedance spectroscopy, microwave backscattering and fibre optic sensors. A review of these monitoring technologies can be found in [43]. Corrosion sensors based on electrical resistance and electromechanical impedance spectroscopy can only provide coverage over a small area and are not suitable for non-uniform corrosion artefacts such as pits. Recent advances in acoustic emission (AE) sensor technology [44, 45] have led to corrosion detection and monitoring solutions where acoustics signals from micro-fractures and delamination of the oxide are analysed. These emissions release much less energy than emission from crack growth where AE has shown great potential. In low noise environments AE could be used to detect signals from corrosion with tens of metres range using monitoring frequencies of tens of kilohertz. However, in a live plant, high process noise requires several hundred kilohertz of monitoring frequencies and coverage is limited <0.5 m and requires complex signal processing. For this reason, AE is limited for this application. Magnetostrictive sensor (MsS) is another technology for pipeline monitoring first developed and patented by SwRI® [46]. They have lower power output compared to piezoelectric transducer, however, recent advancements have reported significant improvements in their power output, sensitivity and flaw characterisation [47]. Piezoelectric sensing offers the most promising solution due to their stability, reliability, and cost-effectiveness as described in Section 2.2. This has enabled the development of several SHM solutions. Guided Ultrasonic Ltd. offers one such monitoring system gPIMS [48] and this system's stability and defect detection capabilities have been demonstrated [49] at temperatures up to 90°C. Another example is the system developed by the authors and its installation, operation and performance is reported [50]. **Figure 8** shows some of these pipeline monitoring devices.
