This paper presents an experimental study on the bond durability of basalt fiber–reinforced polymer (BFRP) bars, steel–fiber reinforced polymer (FRP) composite bars (SFCB), and steel bars embedded in seawater sea-sand concrete (SWSSC) and subject to a simulated ocean environment. The effects of the environment type, exposure period (i.e., 3, 6, and 9 months), and reinforcement type on bond durability were investigated. The degradation mechanism was analyzed using scanning electron microscopy (SEM). The test results showed that after being conditioned, the failure mode changed from shear failure at the BFRP rib-concrete occlusal interface to the peeling of the BFRP surface. For the BFRP bars, after 9 months of aging, the bond stress retention was 92% in the wet-dry cycling environment and 78% in the immersion environment. For the SFCBs, after 9 months of aging, the retention in the wet-dry cycling environment was 100% and that in the immersion environment was 90%. For the steel bars, although there was no reduction in bond strength, pitting corrosion was observed at the bar surface. The predicted bond stress retention of BFRP bars embedded in SWSSC after 50 years of service life in various environmental conditions ranged from 47% to 83%.

A new type of large-diameter iron-based shape memory alloy (Fe-SMA) stranded wire was developed, and its stress recovery behaviour was experimentally investigated in this study. The produced Fe-SMA stranded wires had a diameter of 15.2 mm, which were the same as the most commonly used steel-stranded wires. This kind of large-diameter Fe-SMA stranded wire matched the existing prestressed anchorage system to reduce application cost and accelerate the field application of the Fe-SMA in civil engineering. The effects of prestrain levels (4%, 6%, 8%, 10%), activation temperatures (200°C, 300°C, 400°C, 500°C), and secondary activation effect on the recovery stress of Fe-SMA stranded wires were analysed. Test results showed that the recovery stresses of the Fe-SMA stranded wire were 223–357 MPa. The maximum recovery stress was measured for the specimen with a prestrain of 8% and an activation temperature of 500°C. With increasing activation temperature, the recovery stress increased, but the increment decreased gradually. An increased prestrain level resulted in a first increase and a second decrease in the recovery stress, which was highest at the 8% prestrain level. If the shape memory effect (SME) was not fully used during the first activation, then the remaining martensitic phase changed on the second activation. The recovery stress generated by the first activation was essentially the same as those generated by the second activation at the same activation temperature. Taking the Nanshao River concrete box girder bridge in China as an example, the effect of using Fe-SMA stranded wires to compensate the prestressing loss was calculated. The results showed that the Fe-SMA stranded wire developed was able to compensate the vertical prestress loss of the prestressed box girder bridge web well and maintain the main tensile stresses at a low level during the service life of the bridge.

This paper proposes a novel approach for monitoring multi-bolt looseness using guided waves and the cross-correlation of the wavelet energy envelope. By assessing variations in the wave packet, the looseness in multi-bolt assemblies can be estimated. First, the dispersion effects of Lamb waves were theoretically analyzed using the Rayleigh–Lamb equation. Next, the wavelet energy was derived through wavelet transform, and the Lamb wave envelope was obtained as a criterion for accurately separating the wave packet. Cross-correlation analysis was employed to quantitatively evaluate the dispersion of wave packets for varying levels of bolt looseness. A looseness index, termed the normalized decorrelation coefficient of wavelet energy (NDCWE), was defined. Then, validation experiments were conducted using a joint with five M8 bolts, each tightened to a standard torque of 42 N·m. Two piezoelectric transducers were attached to the periphery of the bolt group. Three preload conditions were tested for each bolt: fully tightened, 80% of the standard torque, and 10% of the standard torque, corresponding to no looseness, minor looseness, and significant looseness, respectively. Results showed that when significant looseness occurs, the NDCWE value exceeds 0.4, confirming the effectiveness of NDCWE in detecting substantial reductions in bolt preload. Experiments assessing the effect of temperature revealed that temperature has a negligible effect on the waveforms of the S ₀ and A ₀ mode waves. Finally, to quantitatively evaluate the efficiency of the ultrasonic transducers, the bolt-to-sensor ratio (BSR) was introduced. In this study, the BSR reached 2.5, indicating that a single piezoelectric transducer can monitor the preload of 2.5 bolts. The proposed approach shows great potential for multi-bolt looseness monitoring.

The root mean square deviation is a common way to measure the electromechanical impedance of piezoceramic transducers that are used for traditional preload monitoring. However, due to the impedance signal’s high sensitivity to temperature, most current research is carried out under the same temperature conditions to avoid its effects. Even so, it is impossible to ignore the temperature factor in practical engineering, which hinders the application of impedance technology. Therefore, this paper proposes a novel approach for preload monitoring in actual engineering: using a wireless impedance signal acquisition platform to collect long-distance impedance signals and then establishing a predicted model based on machine learning (ML) considering the temperature. The wireless impedance signal acquisition platform includes a smart washer, a piezoelectric impedance wireless sensing device, and the host computer software. This test established a dataset under various operating conditions and developed a machine-learning-based predictive model. After comparing five typical ML algorithms, it was discovered that XGBoost performed better in prediction accuracy and generalization ability. Moreover, the Shapley additive explanation method is utilized to further analyze and interpret the XGBoost model. It indicates that ML primarily relies on numerical features (such as the area of each subinterval) to identify the impedance signal and predict the prestress, whereas information features (such as temperature value, peak, etc.) have little influence on the model’s output. Finally, the results above demonstrate that the ML-based models can predict the preload at different temperatures, effectively reducing temperature interference.