Real-time structural health monitoring for concrete beams: a cost-effective ‘Industry 4.0’ solution using piezo sensors
Summary (4 min read)
INTRODUCTION
- Indeed, over 25% of Canadian concrete bridges are deemed to be structurally deficient (Cusson et al., 2011), and 85% of high-rise buildings in New South Wales (NSW) built after 2000 had some form of structural failure (Randolph et al., 2019).
- LVDT sensors will provide insufficient information regarding the cause of the observed displacement (Subramanian and Murugesan 2019), and microscopes can only observe and measure localised surface deformations without any indication of below surface deformations (Bernard 2019).
- Researchers have also suggested the use of several forms of embedded and surface sensors in concrete to assess the concrete quality, economically (Taheri 2019b).
STRUCTURAL HEALTH MONITORING OF CONCRETE MEMBERS
- Various methods for SHM are applied across the industry to observe, record and analyse physical changes to structural members throughout their lifecycle (Lynch et al., 2016).
- Moreover, these methods are proven cumbersome with low efficacy and increasingly, are deemed impractical (Ghodoosi et al., 2018; Oesterreich and Teuteberg 2016).
- Visual-based observation techniques such as the human eye, fibrescope, borescope, hand-held magnifier or stereo microscope are labour intensive and do not offer detailed or quantitative information about interior defects occurring internally within concrete members.
- Acoustic techniques such as the rebound hammer, ultrasonic pulse velocity (UPV), impact echo, spectral wave analysis, crosshole sonic lagging or parallel seismic have various limitations.
- Set against this contextual backdrop, a paradigm shift has occurred in the market, where new low-cost and highly accurate digital methods are designed based on including sensors that can be embedded internally in new structures or on the surface of already existing structures (Li et al., 2016).
Sensor-based methods
- The contemporary concreting industry has progressively moved away from cumbersome SHM tools and techniques (Zinno et al., 2019).
- As illustrated, most existing techniques are complex, expensive and operators require rigorous training to possess competency in these techniques.
- This study proposes a novel technique with the help of a pilot study to address this research and provide industry with a viable accurate solution at an extremely affordable cost.
- Low- cost piezoceramic sensors ($2.22 AUD each) and a raspberry pi model B 3+ ($54 AUD) as a controller are identified as a viable alternative package for monitoring the structural health of concrete members.
Piezoceramic Sensors
- Piezoceramic sensors have been utilised heavily for SHM in the aircraft industry (Chang 2016; Shen et al., 2006), automobile (Martinotto et al., 2016) and manufacturing (Hossain et al., 2016) industries.
- Water solubility and high humidity environments can affect the sensor (Mikulik and Linderman 2019).
- Moreover, Dong et al., (2019) suggest that the use of piezoceramic sensors may affect the mechanical properties of the concrete structure when they are embedded.
- That said, these barriers can readily be overcome with cost-effective techniques.
- Yan et al. (2013), also embedded sensors into smaller sized concrete blocks to form a concrete smart aggregate and avoid physical damage that may occur to the delicate patches (where the latter may be damaged during curing of concrete members).
RESEARCH PHILOSOPHY
- Whilst interpretavism (Roberts et al. 2019) informs the research direction and methods of measurement employed (via qualitative analysis of literature), positivism is employed to conduct quantitative analysis of empirical data (Edwards et al. 2019).
- This combination of philosophies ensures that a scientifically robust research instrument is adopted.
Research Approach
- When piezoceramic sensors are used to measure the mechanical properties of concrete, one of three common methods are often adopted: the impendence based method; the vibration characteristic based method; and the lamb-wave based method (Stojić et al., 2012).
- This data is then correlated to strain displacements measurements collected by a LVDT electric strain gauge to assess crack detection and occurrence in four test sample members under various loading conditions.
- For data analysis, signal processing techniques were adopted including Fourier transform (ul Haq et al., 2017), Hilbert-Huang transform (Wei et al., 2016) and wavelet analysis (Jain et al., 2016).
- Fourier transform, Hilbert-Huang transform and wavelet analysis methods require extensive mathematical computation and signal processing.
- Hence, for the purposes of this study, a simple hybrid analysis technique that correlates vibrational voltage feedback from piezoelectric elements and simple LVDT strain gauge displacements is adopted to facilitate easy adoption by industry practitioners.
Concrete member design
- The beam will have 25 mm cover on all sides .
- The mix-design and material composition of the M25 grade concrete members are provided in Table 1.
- The Raspberry Pi microcontroller is a low cost, credit card-sized computer that plugs into a computer monitor or TV, and uses a standard keyboard and mouse, and uses much lesser power than other equivalent computing units (Raspberry Pi Foundation 2019).
- The analogue to digital converter (ADC), is utilised to convert the analogue data received from the piezoceramic sensor into digital signals that are passed to the Raspberry Pi.
EXPERIMENT AND ANALYSIS DESIGN
- The concrete mix components have been prepared, weighed, and dry mixed before adding the water using a lab scale mixer at Deakin’s concrete laboratory.
- Then, the wet mix has been poured into the prepared moulds.
- The piezoceramic sensors are then attached to the pre-determined locations on the surface of the beams .
- One beam will be tested in flexure under three- point loading set up.
- Furthermore, surface cracks were visually monitored and measured to use as further reference material when compari g several results.
Sensor Setup
- Sensors were attached to each beam externally using adhesive tape to ensure optimised surface contact between the sensors and the concrete beam.
- The final beam included five sensors where there were two on each of the front and rear faces and one on the base of the beam .
- Setting up the sensors involved attaching the wiring by ensuring the male end of a wire was touching the exposed wire from the sensor and securing with tape.
- Where the wire length previously connected to the sensors was not long enough, further extensions are attached ensuring that the length ends with a male connection point.
- First, it was anticipated that because the sensors will collect data within a range of 20 – 50 mm, they were placed in the region of expected large damage on the beam.
Three-point bending test
- The test is performed on the beams to achieve the ultimate flexural load.
- This is the maximum transverse load and the corresponding bending moment that the beam can tolerate before full structural failure.
- All outputs have been connected to a control panel and data acquisitioning system to capture the load-displacement relationship of each test.
- Therefore, the piezoceramic sensors were attached towards the mid-section of the beam.
- The materials have the property of generating an electric charge when subjected to a mechanical strain (direct effect for sensor) and conversely, generate a mechanical strain when subjected to an applied electric field (Taheri 2019b).
FROM EXPERIMENTS TO FINDINGS
- Final testing was carried on once the entire setup was ready.
- The beam was placed on the testing frame and ensuring that the marking were made in such a way that 13 sensors are connected to the correct location.
- The load was gradually applied on the beams and the code was run at the start of loading and the loading on the beams continued till the specimen failed due to excessive deformation and concrete crushing.
- Data streams from the piezoceramic sensors were collected for all four samples in SmartWorks platform with technical support provided by AltAir Solutions Company (Agarwal and Alam 2018).
Test Beam 1 (control test)
- Beam 1 had 13 sensors attached on the surfaces when the load was applied.
- At initial time instant, the voltage fluctuation of the sensors is ignored.
- In fact the voltage spikes occur a few seconds before the surface cracks are observed on camera.
- Also certain micro-cracks and internal cracks which cannot be registered on camera or even on a microscope are easily detected through the piezoceramic sensors through minor voltage spikes.
- Similar tests are repeated for beams 2, 3, 4 respectively with similar displacement graphs obtained for the respective beams with the piezoceramic sensors detecting a spike in voltage at each of the major displacement spikes.
THEORETICAL AND MANAGERIAL IMPLICATIONS
- The multidisciplinary approach (using Industry 4.0 advanced technologies) adopted towards solving an important maintenance issues associated with the construction industry has some significant theoretical and managerial implications.
- Specifically, the work provides an economical and multi-featured addition to extant literature in the area of non-destructive testing (NDT) techniques (as outlined in Appendices 1 and 2).
- The research presented therefore provides a useful case study of Industry 4.0 adoption and thus serves to generate wider polemic debate and discussion within the contemporary construction and civil e gineering management discipline.
- Significant time savings (and by implication, cost savings) can be made in turnaround time required to obtain test results.
- Enhanced transferability of data across the supply chain to better inform practitioners involved in the post-construction stages and assist decision making on maintaining concrete structures.
CONCLUSIONS
- Many of these techniques suffer from the limitations of economic infeasibility or complex signal-processing techniques.
- Presently this method has been used on large scale infrastructure projects or some critical projects (Park et al., 2003; Su et al., 2018).
- The results of this study prove that piezoceramic sensors could detect both internal and external cracks and assist in real-time monitoring of concrete structures.
- This study also serves as a real-life application of Industry 4.0 in the construction sector and consequently, reveals how technology can automated this process moving forwards.
Did you find this useful? Give us your feedback
Citations
81 citations
Cites background from "Real-time structural health monitor..."
...…the significant and plethora of advantages offered by Industry 4.0 for enhancing the construction industry’s performance (Bebelaar et al., 2018; Ghosh et al., 2020), a review of extant literature demonstrates conspicuous deficiencies with existing research undertaken; for example, there is a…...
[...]
78 citations
76 citations
46 citations
29 citations
References
168 citations
144 citations
135 citations
132 citations
Related Papers (5)
Frequently Asked Questions (16)
Q2. What have the authors stated for future works in "Real-time structural health monitoring for concrete beams: a cost-effective ‘industry 4.0’ solution using piezo sensors" ?
However, this approach being a preliminary scoping study had several limitations and some significant lessons for future studies. The use of wired sensors and its complex and sensitive circuit could potentially lead to delays and damage to devices. Studies investigating the exact range and lifespan of piezoceramic sensors could also further assist in fine-tuning this technique for in ustrial use.
Q3. What is the way to protect the sensors?
As cement continually reacts with water and developstrong bonds between mix components to build the final concrete strength, a protectivelayer is necessary to protect the embedded sensors from its boundary, moisture damage,and corrosion (Sanches et al., 2019).
Q4. What is the role of the piezoceramic element in concrete?
Acting as a sensor, actuator, accelerator ortransducer within the concrete member, the piezoceramic sensors detect the electricalenergy converted from mechanical energy and convert it into a voltage output (Ballasand Schoen 2017).
Q5. What are the limitations of conventional methods?
Acoustic techniques such as the rebound hammer, ultrasonic pulse velocity(UPV), impact echo, spectral wave analysis, crosshole sonic lagging or parallel seismichave various limitations.
Q6. What is the design of the testing frame?
The testing frame is self-supported type andprovide a full circle of loading system, while the loading has been introduced through ahydraulic jack with load cell to monitor the actual applied load.
Q7. What is the ultimate load at which the structures failed?
The ultimate load at which the structures failed was recorded as 88.37 kN , 83.31 kN,78.71 kN and 89.61kN for test samples 1, 2, 3 and 4 respectively.
Q8. What are the main uses of piezoceramic sensors?
Piezoceramic sensors have been utilised heavily for SHM in the aircraft industry (Chang2016; Shen et al., 2006), automobile (Martinotto et al., 2016) and manufacturing(Hossain et al., 2016) industries.
Q9. What are the advantages of piezo sensors?
In summary, although piezo elements have limitations of being fragile and non-waterresistant, their economic feasibility and simplicity of usage provide strong arguments forusing them on real-time SHM projects.
Q10. What was the reason for the installation of the sensors?
it was anticipated that because the sensorswill collect data within a range of 20 – 50 mm, they were placed in the region ofexpected large damage on the beam.
Q11. How many Pa/seconds is used to control the beam?
while it has been changed to deflection control of 1 mm/minute at the laterstages to ensure capturing the full load-deflection relationship and to avoid suddenfailure and damages to the instrumentations.
Q12. What is the purpose of this study?
This study investigates the application of low-costpiezoceramic sensors to detect deformations within the concrete structure (i.e., cracksand fractures) due to the member being placed under physical strain.
Q13. What is the significance of the multidisciplinary approach?
The multidisciplinary approach (using Industry 4.0 advanced technologies) adoptedtowards solving an important maintenance issues associated with the constructionindustry has some significant theoretical and managerial implications.
Q14. How many sensors were attached to the beam?
The four beams included 13sensors for each beam with five on the front and rear faces of the beam, one on the baseand two on the top (ref. Fig. 5 and 6).
Q15. What are the main components used in the present study?
The main components used in the present ‘Industry 4.0’ study include: a Raspberry Pi;piezoceramic sensors; a breadboard; analogue to digital converter; and two 16-bitmultiplexers (refer to Figure 2).<Insert Figure 2 about here>
Q16. What are the main limitations of conventional methods?
conventional methods include different tests such as a simple human eyedetection of surface defects (Ghodoosi et al., 2018) or a compressive strength test whichonly provides results after a 28 day curing period (Yildirim et al., 2015).