Acoustic Emission (AE) Testing for Online Monitoring of Storage Tank Floors

Large storage tanks at petroleum and chemical plants and liquid cargo terminals suffer primarily from floor-plate corrosion, which has historically required tank-opening (internal) inspection to confirm integrity. Thanks to the maturation of Acoustic Emission (AE) testing, however, corrosion activity can now be monitored in real time without taking the tank out of service. This article walks through the fundamentals of AE, sensor layout, signal discrimination, and its role as an alternative to internal inspections.

AE Fundamentals and Application to Storage Tanks

Acoustic Emission is a non-destructive testing technique that detects elastic stress waves released when a material undergoes deformation, crack growth, or corrosion. In the case of tank-floor corrosion, hydrogen generation from electrochemical reactions and the collapse of corrosion products produce elastic waves in the 20 kHz–1 MHz range that propagate through the plate.

AE sensors attached to the lower shell or floor edge of the tank capture these waves, and computer analysis evaluates the presence, location, and activity of corrosion. International standards such as ASTM E1002 and API 653 recognize AE inspection as a valid screening method for extending internal inspection intervals.

Sensor Layout and Source-Location Accuracy

Locating corrosion with AE relies on source location — computed from differences in wave arrival time across multiple sensors. Typical recommendations for large tanks (diameter 20–60 m) are:

• Sensor spacing: 3–6 m around the outer shell, accounting for wave velocity and attenuation in steel.
• Sensor count: 8–12 minimum for ~30 m diameter tanks, 16+ for larger tanks.
• Location accuracy: Within ±0.5 m when sensor layout and sampling rate (≥10 MHz) are adequate.

Wave attenuation at the floor–ground interface and complex propagation caused by sediment under the floor degrade accuracy, so a Pencil Break (Hsu-Nielsen) test should be performed to confirm each sensor's sensitivity and attenuation characteristics.

Discriminating Noise from Real Corrosion Signals

Field noise is the dominant challenge for AE in service. Improving signal discrimination is the key to practical deployment, and the following methods are typically combined:

• Frequency analysis: Hydrogen bubble collapse from corrosion concentrates in the 100–300 kHz band, separating it from mechanical friction (low frequency) and electrical spikes (high frequency).
• Waveform parameter classification: Rise time, duration, and energy patterns. Corrosion signals typically have longer duration and more continuous energy.
• Location filtering: Events clustered at known external noise sources (pumps, piping) are rejected.
• Guard sensors: Additional sensors near noise sources screen out non-tank events.

Machine-learning classifiers (SVM, neural networks) have also been studied; recent publications report false-positive reductions of 50% or more.

Standards and Practical Use as an Alternative to Internal Inspection

API 653 "Tank Inspection, Repair, Alteration, and Reconstruction" allows interval extensions for tanks that meet AE acceptance criteria, in combination with Risk-Based Inspection (RBI) evaluation. This can dramatically reduce asset-management costs.

• Economic impact: Internal inspection (cleaning, gas freeing, scaffolding) can cost millions of yen. If AE finds no significant corrosion, interval extension saves substantial operating costs.
• Japanese regulatory context: Under the Fire Service Act, AE inspection is increasingly adopted for evaluating floor corrosion at hazardous-material facilities, though prior consultation with the regulator is required.
• Continuous monitoring: IoT-enabled AE loggers installed permanently on tanks, with 24/7 cloud-based monitoring, are becoming more common.

Summary

AE-based online monitoring of tank floors offers three clear advantages: (1) real-time inspection without taking the tank out of service, (2) accurate source location of corrosion, and (3) improving practical accuracy through advanced signal discrimination. Challenges remain — facility-specific noise evaluation and regulator coordination, for example. Urisol Inc. supports technical studies on AE application and integrated evaluation with existing UT data. Please contact us for details.

References

• ASTM International, "ASTM E1976-22 Standard Guide for Use of the Acoustic Emission Method." https://www.astm.org/e1976-22.html
• API, "API 653 – Tank Inspection, Repair, Alteration, and Reconstruction." https://www.api.org/products-and-services/standards/important-standards-announcements/standard/653
• NDT Resource Center, "Acoustic Emission Testing – Source Location." https://www.nde-ed.org/NDETechniques/AE/ae_sourcelocation.xhtml
• Physical Acoustics Corporation, "Acoustic Emission Testing of Above Ground Storage Tank Floors." https://www.pacndt.com/applications/storage-tanks/
• Ha-Van, T. et al., "Machine learning-based classification of acoustic emission signals for corrosion detection in storage tanks," Ultrasonics, Vol. 124, 2022. https://www.sciencedirect.com/science/article/pii/S0041624X22000555