Overhead Crane Steel Structure Fatigue Monitoring with FBG Sensors

Following the fatigue life assessment methodology for metal structures outlined in ISO 4301 Crane Design Standard, crane steel structures accumulate fatigue damage under long-term alternating loads. Traditional periodic non-destructive testing (MT/UT) cannot capture the full crack initiation and propagation process. Fiber Bragg Grating (FBG) sensing technology offers a distributed, high-precision, long-term online monitoring solution for crane structural health monitoring (SHM). The KL-SHM-FBG system, which integrates fiber-optic strain sensing with fatigue life assessment algorithms, has been deployed across multiple metallurgical and port applications.

FBG Sensing System: Components and Operating Principle

FBG sensors use UV laser to inscribe a periodic refractive index grating into the fiber core. When broadband light is launched, the specific wavelength satisfying the Bragg condition λ_B=2n_effΛ is reflected back. When the fiber experiences axial strain or temperature changes, both the grating period Λ and the effective refractive index n_eff shift, causing the reflected wavelength to drift. For strain measurement, typical sensitivity is 1.2 pm/με with a resolution of 0.1 με — significantly finer than the 1 με resolution of resistance strain gauges.

The KL-SHM-FBG system comprises three main components:

①Fiber-optic light source module — ASE broadband source (C-band 1525–1565 nm, output power ≥10 mW);

②Wavelength demodulator — based on tunable Fabry–Pérot filter principle, scan frequency 1 kHz, wavelength resolution 1 pm;

③Sensor network array — each channel supports up to 16 FBG sensors in series, differentiated using wavelength division multiplexing (WDM).

Data is transmitted via a 4G/5G gateway to the cloud-based SHM platform for remote real-time viewing.

Strain Sensor Network Deployment Strategy

The crane main girder, end carriages, and connection zones are fatigue-sensitive areas that require prioritized monitoring. Based on finite element analysis (FEA) results, bi-directional FBG strain sensors (longitudinal and transverse) are installed on the bottom flange plate at the mid-span section of the main girder. Uni-directional sensors are placed at both ends of the main girder (approximately one-quarter of the span from the end carriage), and multi-point sensors are arranged around the High-Strength Bolt groups at the end carriage connections.

Sensor installation uses a pre-tensioned adhesive fixing method with a cyanoacrylate protective coating. Each sensor is individually calibrated during workshop prefabrication to ensure an initial wavelength accuracy of ±0.05 nm. During field deployment, temperature-compensating FBGs (installed strain-free and suspended) eliminate temperature cross-sensitivity effects. Each main girder is recommended to have 16–24 measurement points, forming a complete strain field monitoring network. Signal cables run along cable trays on the main girder web plate and terminate at the data acquisition cabinet beneath the operator cab.

Sensor spacing involves a trade-off between detection accuracy and system cost. A practical guideline: for general purpose bridge cranes with spans of 22.5–31.5 m, bi-directional FBG points are placed every 2.5–3 m along the longitudinal direction of the main girder bottom flange, with mid-span zones densified to 1.5 m intervals. For end carriage connection areas, priority is given to the first and last rows of High-Strength Bolt groups (stress concentration zones), with at least 2 measurement points per connection node (one active, one redundant). For metallurgical cranes (Work Duty A7/A8), all regions with FEA stress levels above 60 MPa should be instrumented to ensure 100% coverage of critical cross-sections. The WDM capability of the KL-SHM-FBG system allows measurement points to be added or removed flexibly along the main trunk cable without re-laying the fiber-optic backbone.

Online Fatigue Life Assessment Methodology

Fatigue life assessment based on real-time strain-time history data follows a standard three-step procedure. First, rainflow counting is applied to the strain-time waveform at each measurement point to extract full and half cycles as amplitude-mean pairs. Next, the damage contribution of each cycle is calculated using the Palmgren-Miner Linear Cumulative Damage Rule (D = Σ n_i/N_i ≤ 1.0 for Design Life Verification), based on the material S-N curve (fatigue class FAT 125–160 for Q235B (≈S235JR) and Q355B (≈S355JR) steels commonly used in overhead cranes). Finally, the cumulative damage D = ΣD_i is summed; when D ≥ 1, the location is deemed to have reached its fatigue life limit.

To improve assessment accuracy, the system incorporates the Goodman mean stress correction model, which converts the stress amplitude of non-symmetric cycles into an equivalent zero-mean stress amplitude: σ_a,e = σ_a × σ_b/(σ_b−σ_m). A Markov matrix stores the rainflow counting results, with each measurement point maintaining a 128×128 amplitude-mean two-dimensional matrix, achieving a data compression ratio of 1000:1—suitable for long-term online recording. The system updates the cumulative damage curve every 10 minutes and automatically pushes a warning notification when the daily damage rate exceeds the threshold (e.g., single-day damage > 0.01%).

In engineering practice, uncertainty in fatigue life assessment primarily stems from the scatter in S-N curves and the handling of low-amplitude cycles during rainflow counting. For low-amplitude cycles (amplitude < 5με), the International Institute of Welding (IIW) recommends a cut-off limit approach, directly ignoring them since their fatigue damage contribution falls below the fatigue limit threshold. For medium-amplitude cycles (5–30με), the Haibach correction formula adjusts the S-N curve slope from m = 3 to m' = 2m−1 = 5, providing a more conservative evaluation of cumulative effects from low-stress cycles. The KL-SHM-FBG system defaults to the IIW standard dual-slope S-N curve method; users can also select different standards (e.g., Eurocode 3, BS 7608, or AS 4100) in the software interface based on the specific steel grade and joint type.

FBG vs. Resistance Strain Gauge: Six Key Comparisons

To clearly demonstrate the technical advantages of fiber Bragg grating (FBG) sensing over traditional resistance strain gauges, the following comparison examines six critical dimensions. These parameters directly determine the long-term reliability and maintenance costs of a structural health monitoring (SHM) system in demanding industrial environments such as steel mills and ports, making them the core decision factors during system selection.

From an engineering application perspective, the key trade-offs in solution selection include initial capital investment, long-term maintenance costs, data reliability, and system scalability. While resistance strain gauge solutions offer lower per-point procurement costs, in the high-vibration, dust-laden, and electromagnetically noisy environment of overhead crane operations, issues such as signal cable aging and fracture, amplifier zero drift, and radio-frequency interference drive the total cost of ownership (TCO) to 1.8–2.5 times that of an FBG-based solution. The six-dimensional comparison below helps enterprises make an informed choice based on their specific operating conditions and budget constraints.

Real-World SHM Case Study: Steel Plant Overhead Crane

Take a 32t/50t overhead crane in the steelmaking bay of a steel plant as an example. The crane operates at work duty A7, completes an average of 220 duty cycles per day, and has been in service for 8 years. In June 2025, a KL-SHM-FBG system was installed, deploying 24 FBG sensing points across the main girder mid-span, end carriage connection zones, and weld seam heat-affected areas. On day 47 after commissioning, sensing point #7 at the main girder mid-span detected an 18% sudden increase in strain amplitude. Rainflow counting analysis indicated that the daily damage rate at this point jumped from a baseline of 0.003% to 0.024%.

Magnetic Particle Inspection (MPI) performed during a scheduled shutdown confirmed a transverse fatigue crack approximately 12mm in length in that region. After timely weld repair, the crane returned to normal operation, preventing a potential sudden fracture. This case demonstrates the engineering value of the FBG-SHM system's early warning capability.

The success of this deployment attracted attention across the industry, leading 3 additional steel plants and 2 port operators to install similar SHM systems. From an economic standpoint, the average cost of repairing a main girder fracture caused by undetected cracks is approximately ¥450,000–800,000 (including structural replacement, production downtime, and hoisting costs), while the per-crane deployment cost of the KL-SHM-FBG system is approximately ¥60,000–100,000, yielding a cost-benefit ratio of roughly 1:5 to 1:8. Within 12 months of system operation, the steel plant identified 3 fatigue cracks and 7 loose connection components in advance, directly preventing 2 potential main girder fracture incidents. The plant has since standardized on FBG sensing for SHM upgrades across 6 additional cranes, with a cumulative 192 FBG sensors deployed across all metallurgical bay cranes.

Frequently Asked Questions