DC-10: Eden Ngowi

Contact

E-mail: eden.ngowi@ntnu.no

💼LinkedIn 🆔ORCID

Project

Learning-based health-aware operation and maintenance planning for improved safety

Host Organization

NTNU

Supervisors

Prof. Prof. Johannes Jäschke (Main, NTUN); Prof. Alessandra Russo (co-supervisor, IMPERIAL)

Objectives

  1. To build algorithms for diagnostics (state estimation) and prognostics.
  2. To validate the built algorithms by simulation of selected case studies.
  3. To implement the validated algorithms in prototype plants.

Project Description

Equipment and plants degrade with use thus the concept of maintenance arises naturally. Maintenance strategies have evolved with advancement of science and technology. Strategies such as predictive maintenance are now possible since an equipment’s health condition can be monitored and used to estimate the equipment’s remaining useful life (RUL). In many processes especially subsea operations, unplanned maintenances are costly. To avoid them, health aware control (HAC) is implemented leveraging the condition monitoring techniques to extend the equipment’s RUL until the next planned maintenance. Thus, a trade-off is stricken between production (profitability) and health (reliability).

While this is an excellent method, it is hurdled with many challenges such as huge computational cost due to complexity of the models being handled and the large time scale difference between the fast regulatory control actions and degradation dynamics posing a long prediction horizon; and uncertainty in the models as parameters may change with time (stochasticity). All this leads to obtaining near-optimal solutions which can be improved by leveraging data science techniques.

The AI boom is on the move employing ML techniques which enable trends to be drawn from data (diagnostics). This opens an avenue of creating of simpler dynamic models (process, degradation/prognostic) which accurately reflect the phenomenological processes. Such models when used with first principle models can greatly improve HAC implementation in industry leading to safer and more cost effective processes.

Relevant Background

  1.  BSc Chemical Engineering (2015-2019) and MSc Chemical Engineering (2020-2023) from the University of Dar-es-Salaam.
  2.  Master's Thesis: "Development and Testing of a Small-Scale Flash Dryer: A Case Study of Maize Bran Drying."

Publications 

  1. Ngowi, E., Jeremiah, J. M., Kaale, L., & Elisante, E. (2024). Development and Testing of Small-Scale Flash Dryer for Maize Bran. Tanzania Journal of Engineering and Technology, 43(2), 75-91. https://doi.org/10.52339/tjet.v43i2.908