2022 HBK Technology Days

Prof. Elizabeth Cross, Dynamics Research Group, University of Sheffield

Structural Health Monitoring as a research field began by studying physical models for systems, comparing modelled and monitored response to understand observed behaviours. As sensing technology advanced and monitoring data became more readily available, many practitioners and researchers adopted a purely data-driven approach capitalising also on technology enhancement from the machine learning field.

Today, where we would like to be able to automatically assess the health of our structures across their operational envelope, we find that despite these advances, we often lack data that represent all behaviours of interest, thereby precluding an entirely data-driven approach.

In this talk, we will discuss the development of a physics-informed approach to machine learning for structural health monitoring. This fast-growing area of research attempts to build our engineering knowledge of a system, alleviating some burden on data acquisition and increasing model interpretability.

Haroun El Mir, PhD Researcher, Transport Systems, Cranfield University

Landing gear systems on Aircraft undergo a multitude of forces during their life cycle, leading to the eventual replacement of this system based on a ‘safe life’ approach that certain circumstances underestimate the component’s remaining useful life.

The efficacy of fatigue life approximation methodologies is studied and compared to the ongoing Structural Health Monitoring techniques being researched, which will forecast failures based on the system’s specific life and withstanding abilities, ranging from creating a digital twin to applying neural network technologies, in order to simulate and approximate locations and levels of failure along the structure.

Explainable Artificial Intelligence (AI) allows for the ease of integration of Deep Neural Network (DNN) data into predictive maintenance, which is a procedure focused on the health of a system and its efficient upkeep via the use of sensor-based data.

Test data from a flight includes a multitude of conditions and varying parameters such as the surface of the landing strip as well as the aircraft itself, requiring the use of DNN models for damage assessment and failure anticipation, where compliance to standards is a major question raised, as the EASA AI roadmap is followed, as well as the ICAO and FAA.

This presentation additionally discusses the challenges faced with respect to standardizing the explainable AI methodologies and their parameters specifically for the case of landing gear.

Dietmar Maicz, Railway and Infrastructure Monitoring Expert, HBK

With the help of HBK's TSI-Spot® (Single Point Of Truth) concept, railroad companies are enabled to obtain a holistic view of vehicles and infrastructure. New measurement methods and data processing methods allow the early detection of unnecessary wear and deviations on vehicles (wheels and pantograph) and infrastructure (superstructure and overhead line) in real-time and, thanks to the high data quality, to derive reliable, fully automated forecasts for future plannable maintenance interventions.

Maintenance planning receives daily updated information about the track condition and the individual reaction of a vehicle type in real-time. HBK's high-precision, compact and cost-effective measurement technology, combined with state-of-the-art software solutions, enables fully autonomous measurement of the relevant permanent way and overhead line parameters in real-time.

Mr. Chris Wynn-Jones, Application Engineer – Reliability, HBK

A fault tree is used to identify the critical path/s in a process, system or service. This introduction will focus on two different views, top down and bottom up.

Top Down

• This view is read downwards, from undesirable events down to the lowest known fault/cause branch item.
• All the undesirable events that can result from your Process, System or Service, are driven by something that is no longer working correctly. 
• Engineers use this mentality to discover weaknesses in their Process, System or Service to calculate the probability of failure and occurrence. 

Bottom Up

• This view is read upwards, from the fault/cause up to their end effect. 
• All known faults/causes have an undesirable end effect on the Process, System or Service. 
• Using mixed modelling it is possible to combine faults/causes, build in redundancy and/or identify standbys (a mitigating action) enabling reliability and probability of failure to be calculated for known end effects.

Mr. Sam Eisenberg, Product Manager, HBK ReliaSoft Products

This presentation will explore how best to use fault tree analyses to uncover and surface your primary unreliability drivers in your system.

We will explore the methodologies and next steps once you have identified the bad actors in your system.

Mr. Adi Dhora, Engineering Solutions, Reliability Solutions Consultant, HBK & 
Mr. Eric Fritts, Engineering Solutions, Project Engineer, HBK

This presentation describes a practical approach to use automation for creating reliability models, simulations, and insights for a large number of assets. Most industries are looking to pursue more data-driven reliability practices. Most reliability practitioners highly value the insights they gain from fault tree analysis and reliability block diagrams.

However, often a limiting factor is data access, quality, and personnel time to process large quantities of data. This is where automation can help to enable reliability practitioners to focus majority of their time on analysis and decisions.

This presentation will showcase how best to merge automation with reliability-know-how to scale in-depth analysis like reliability predictions and event probability for tens of thousands of assets in a short timeframe.

HBK Prenscia

To be confirmed