Speaker’s name: Peter Aalberts
Organization: ROSEN Group
Job title: Sales Manager
Biography:
Peter Aalberts started in the industry at ROSEN and is working for the company for more than 2 decades. Starting as a survey engineer in 2004 he performed storage tank inspections worldwide. After a few years he became a certified EEMUA 159 and API 570 (pipeline) inspector. During these 2 decades he served several positions within the ROSEN company where he worked on developing new equipment, train new staff, and build up a ROSEN service hub in the Amsterdam region. Since 2024 he has become a sales manager responsible for the sales within the BENELUX area.
Presentation title: Innovating Tank Bottom Inspections with MFL Ultra Sizing and Machine Learning
Presentation abstract/summary:
This paper presents the application of Ultra high-resolution magnetic flux leakage (MFL) technology for the inspection of tank bottoms, focusing on its effectiveness in identifying corrosion, structural defects, and other anomalies that could compromise the integrity of storage tanks, as well as minimize time of personnel inside the tank, by leveraging machine learning algorithms.
Traditional inspection methods often fall short in detecting subtle flaws, particularly in older tanks where corrosion may be localized and difficult to assess. By leveraging advancements in Ultra high resolution MFL, this study demonstrates enhanced sensitivity and accuracy in detecting and sizing these critical issues.
The methodology involves the use of advanced sensor arrays and ML based data processing algorithms that refine the magnetic field and eddy current measurements, enabling the automated detection and sizing of minute changes in material properties.
Results from field tests illustrate a significant improvement in defect characterization and localization compared to conventional techniques. Additionally, the paper discusses automated post-inspection repair planning in compliance with EEMUA and API codes.
This breakthrough, due to its data consistency and reliability, underscores the potential of Ultra high resolution MFL as a transformative tool in data-driven asset integrity management, enabling AI assisted integrity planning.