Eoghan Chelmiah

ORCID ID: 0000-0003-3199-4696

Research Project Title: Intelligent Prognostics of Electric Propulsion Systems for Sustainable Eco-Friendly Vehicles

Supervisors/s: Dr Darren Kavanagh and Dr Violeta McLoone

Project Funding: Government of Ireland Scholarship from the Irish Research Council



  • Biography
  • Research Project Description
  • Publications and Outputs


Eoghan Chelmiah received the First Class B.Eng(Hons) degree in Electronic Systems Engineering from South East Technological University Ireland (formerly known as  Institute of Technology Carlow) in 2019, achieving an Elite Gold Sports Scholarship for representing both the University and Ireland in Boxing, MMA, Karate and Taekwondo during his studies.

He received the Government of Ireland Scholarship from the Irish Research Council to complete a PhD in Machine Learning Methods for Intelligent Prognostics of Electric Machines at South East Technological University.

Eoghan is a member of the IEEE, Signal Processing, Robotics and Automation, and Industrial Electronics Societies.

His research interests include failure diagnostics and prognostics, condition-based monitoring, machine learning, artificial intelligence, pattern recognition and electric propulsion systems.

Research Project Description

We are witnessing an electric revolution within the transport sector, which has largely been accelerated more recently by the 26th Conference of the Parties (COP26) at United Nations Climate Change Conference and Paris Agreement (UNFCCC) on Climate Change (effective Nov. 2016). The Internal Combustion Engine (ICE) is gradually being replaced by electric machines, which results in high performance vehicles that are energy efficient with zero emission of greenhouse gases.

The electrification of the powertrain includes battery cells, power electronics and electric machines for both traction/propulsion as well as for kinetic energy recovery during brake events. However, major reliability and sustainability challenges exist which is severely hampering their widespread adoption and effectiveness.

My applied research project focuses on novel electronic sensing and machine learning (ML) algorithms for advanced Condition-based Monitoring (CbM) of the electric machines used for the propulsion of electric propulsion systems. The impact of this research will contribute towards developing new knowledge and understanding towards significantly advancing the design of robust electric machines that prevents unexpected catastrophic failure modes and premature aging occurring.

Publications and Outputs

  • Chelmiah, E.T., McLoone, V.I. and Kavanagh, D.F., 2022. Remaining Useful Life Estimation of Rotating Machines through Supervised Learning with Non-Linear Approaches. Applied Sciences12(9), p.4136.

Bearings are one of the most common causes of failure for rotating electric machines. Intelligent condition-based monitoring (CbM) can be used to predict rolling element bearing fault modes using non-invasive and inexpensive sensing. Strategically placed accelerometers can acquire bearing vibration signals, which contain salient prognostic information regarding the state of health. Machine learning (ML) algorithms are currently being investigated to accurately predict the health of machines and equipment in real time. This is highly advantageous towards reducing unscheduled maintenance, increasing the operational lifetime, as well as mitigation of the associated health risks caused by catastrophic machine failure. Motivated by this, a robust CbM system is presented for rotating machines that is suitable for various industrial applications. Novel non-linear methods for both feature engineering (one-third octave bands) and wear-state modelling (exponential) are investigated. The paper compares two main types of feature extraction, which are derived from Short-Time Fourier Transform (STFT) and Envelope Analysis (EA). In addition, two types of supervised learning, Support Vector Machines (SVM) and k-Nearest Neighbour (k-NN) are explored. The work is tested and validated on the PRONOSTIA platform dataset, with remaining useful life (RUL) classification results of up to 74.3% and a mean absolute error of 0.08 achieved.


  • Chelmiah, E.T., McLoone, V.I. and Kavanagh, D.F., 2020, October. Remaining useful life estimation of rotating machines using octave spectral features. In IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society (pp. 3031-3036). IEEE.

Bearing failure is one of the most common causes of failure for electric machines. Acquiring the vibration data of a machine with suitably placed accelerometers is a non-invasive and widely adopted approach for obtaining information regarding the health condition of the mechanical bearings.

This paper presents a robust condition monitoring method for wear state classification and remaining useful life estimation of the mechanical rolling element bearings using orthogonal vibration signals.

This proposed method uses non-linear signal processing techniques in the frequency domain for feature subset selection of short-time Fourier Transform (STFT) spectra and non-linear temporal class boundaries for classification using Coarse and Weighted k-Nearest Neighbour.

This method has been tested and validated using the IEEE PHM PRONOSTIA challenge dataset.

The signal processing and ML based approach presented here has performed extremely well with correct classification results of up to 75.6% being achieved.

This work is of significant merit and will be highly valuable for the electric machines community allowing for the implementation of a robust condition monitoring system for many industrial applications using vibration sensors.


  • Chelmiah, E.T., McLoone, V.I. and Kavanagh, D.F., 2020, December. Wear state estimation of rolling element bearings using support vector machines. In 2020 15th IEEE International Conference on Signal Processing (ICSP) (Vol. 1, pp. 306-311). IEEE.

Failure of mechanical bearings is extremely problematic with respect to the reliability of electric and rotating machines for various applications and technologies, in transport, energy systems and advanced robotic systems (industry 4.0). The occurrence of catastrophic failure modes gives rise to abrupt unscheduled down time and maintenance, machine overheating (burn-out), as well as significant reliability and health and safety considerations for various mission critical end-applications.

This paper proposes a robust method of remaining useful life (RUL) estimation using Support Vector Machines (SVMs) to classify the wear states of the rolling element bearings.

Different types of feature sets are derived from the Short-time Fourier Transform (STFT) as well as Envelope Analysis; which are compared for both linear and non-linear wear state models.

The methods were optimised for different types of SVM kernelling functions and achieved classification accuracy of up to 67%. Medium and coarse Gaussian kernelling functions achieved the highest level of accuracy. This proposed method proves to be a valuable non-invasive predictive CbM approach for electric and rotating machines, using accelerometer-based vibration signals mounted on the external races of rolling element bearings under test.


  • Chelmiah, E.T. and Kavanagh, D.F., 2021, October. Hilbert Marginal Spectrum for Failure Mode Diagnosis of Rotating Machines. In IECON 2021–47th Annual Conference of the IEEE Industrial Electronics Society (pp. 1-6). IEEE.

Mechanical bearings are a core component for rotating machines. These critical elements suffer from degradation which can be gradual or abrupt which typically results in premature failure modes occurring. In recent years, bearing faults have been reported to be the cause of up to 75% of low voltage motor/generator breakdowns and up to 41% of all rotating machine failures. Abrupt equipment failure in many critical applications such as aircraft, automotive and energy converters is often extremely costly, untimely and catastrophic, as well presenting serious health and safety implications.

The paper investigates a unique approach for bearing fault mode diagnosis through incorporating novel Hilbert Marginal Spectrum derived features that are utilised in Machine Learning (ML) classification algorithms, specifically that of Support Vector Machines (SVM) and k-Nearest Neighbour (k-NN). The techniques and methods proposed are tested and validated on real vibration signals achieving 93.8% classification accuracy.


  • Chelmiah, E.T. and Kavanagh, D.F., 2022, September. Acoustic Sensor Array Topologies for Partial Discharge Localisation in Electric Machines. In 2022 International Conference on Electrical Machines (ICEM) (pp. 1582-1588). IEEE.

The importance of increasing the robustness and reliability of electric machines is paramount for a large number of industrial applications. In particular for electric propulsion systems being developed for the transport sector, such as electric vehicles, aircraft and ships which incorporate high power and torque density electric machines. Insulation failure has been reported to be responsible for up to 66% of high voltage machine breakdowns and proves to be one of the most challenging to accurately diagnose. A major indicator of electrical insulation failure is the occurrence of partial discharge (PD) events, which occur within air pockets in the insulation or flash-over leakage across or along the surface of coil insulation. This paper presents a method for localising PD emission sources using an acoustic sensor arrays and a cross-correlation based time-difference of arrival (TDOA) algorithm.

The results from analytical experiments are reported for a large traction based switched reluctance machine (SRM), which demonstrate the accuracy, effectiveness and robustness of the approach with accuracy errors of less than 1cm Euclidean distance.

The findings of this work is extremely valuable towards the design and realisation of a hardware based experimental test-bed.