IVHM Approach - The HAHN Spring IVHM Approach covers the following areas:

  • Sensing Technology: To accelerate the approval process of a proposed IVHM sensing technology, HAHN Spring adopts an approach leading to “approval by equivalence” as follows: (a) evaluate the technology effectiveness; (b) establish a methodology to determine the capability of the technology according to industry accepted practices, e.g. to determine the technology Probability Of Detection; (c) establish an approach to reduce the costs of the tests required to determine the capability; (d) demonstrate how the determined capability, when implemented, can maintain the level of safety risks as low as the level achieved by existing approved technologies.
  • Data Analysis: By analysing IVHM data, HAHN Spring would deliver independent recommendations for system improvements; the company would tailor software tools to extend analysis capabilities covering fusion, anomaly detection, advanced usage, diagnostics and prognostics. The tools include model based algorithms, data driven models, tailored statistical methods, artificial intelligence techniques, and simulations targeted at the development and evaluation of advanced IVHM techniques. The tools include advanced signal processing algorithms to: (a) extract time and frequency domain features not only from sensed signals but also from normalised, enhanced, virtual, envelope and average signals; (b) select optimum set of significant diagnostic features that can significantly reduce false alarm rates and no-fault-found events.
  • State Detection: Algorithms can be configured to (a) automatically build a model for normal baseline operations using a number of significant diagnostic features, (b) detect abnormal deviations from the normal model, and (c) evaluate the severity of detected deviations.
  • Health Assessment: A process can be configured to label the abnormal events with the most likely faults, failure modes or abnormal conditions.
  • Prognostics Assessment: HAHN Spring can implement prognostic models showing the progression of faults to failure as a function of operational conditions.
  • Actionable Decisions: The information generated by the above algorithms can be transferred into actionable maintenance and management instructions.
HAHN Spring provides recommendations for system improvements using advanced algorithms operating on huge datasets.
© Copyright 2015 HAHN Spring Ltd.