Ph. D. Project
Title:
Data-driven Robust Detection of Operational Drifts in Pilot Units
Dates:
2024/10/01 - 2027/09/30
Supervisor(s):
Description:
IFPEN aims to become a key player in the triple energy, ecological and digital transition by offering differentiating technological solutions in response
to the societal and industrial challenges of energy and climate. The implementation of new methodological approaches that combine data science and
experimentation is among the solutions under study to accelerate progress and reduce R&D costs. Learning algorithms based on time series data, which
constitute most generated data, are currently gaining momentum in the literature, whether supervised or unsupervised (ARIMA, LSTM...).
The field of Prognostics and Health Management (PHM) is of particular interest. It is a discipline that focuses on the degradation mechanisms of
systems to estimate their health status, anticipate failures, and optimize maintenance. For instance, on a pilot plant, it is very challenging to obtain data
during operational problems, which are becoming more frequent due to the high variability of loads to process and operational conditions to explore.
This thesis aims to overcome these challenges by exploring data-driven methodologies. The choice of this approach is justified by the extensive
diversity of data available at IFPEN. Additionally, new experimental trials may be conducted as a complementary measure.
About IFP Energies nouvelles
IFP Energies nouvelles is a French public-sector research, innovation and training center. Its mission is to develop efficient, economical, clean and
sustainable technologies in the fields of energy, transport and the environment. For more information, see our WEB site
(https://www.ifpenergiesnouvelles.fr/ifpen/presentation).
IFPEN offers a stimulating research environment, with access to first in class laboratory infrastructures and computing facilities. IFPEN offers
competitive salary and benefits packages. All PhD students have access to dedicated seminars and training sessions.
to the societal and industrial challenges of energy and climate. The implementation of new methodological approaches that combine data science and
experimentation is among the solutions under study to accelerate progress and reduce R&D costs. Learning algorithms based on time series data, which
constitute most generated data, are currently gaining momentum in the literature, whether supervised or unsupervised (ARIMA, LSTM...).
The field of Prognostics and Health Management (PHM) is of particular interest. It is a discipline that focuses on the degradation mechanisms of
systems to estimate their health status, anticipate failures, and optimize maintenance. For instance, on a pilot plant, it is very challenging to obtain data
during operational problems, which are becoming more frequent due to the high variability of loads to process and operational conditions to explore.
This thesis aims to overcome these challenges by exploring data-driven methodologies. The choice of this approach is justified by the extensive
diversity of data available at IFPEN. Additionally, new experimental trials may be conducted as a complementary measure.
About IFP Energies nouvelles
IFP Energies nouvelles is a French public-sector research, innovation and training center. Its mission is to develop efficient, economical, clean and
sustainable technologies in the fields of energy, transport and the environment. For more information, see our WEB site
(https://www.ifpenergiesnouvelles.fr/ifpen/presentation).
IFPEN offers a stimulating research environment, with access to first in class laboratory infrastructures and computing facilities. IFPEN offers
competitive salary and benefits packages. All PhD students have access to dedicated seminars and training sessions.
Keywords:
Data science, machine learning, prognostics and health management PHM, deep learning, Process Syste
Department(s):
| Modeling and Control of Industrial Systems |