Advancing System-Level Prognostics with Multimodal Data Integration and Uncertainty Quantification

Advancing System-Level Prognostics with Multimodal Data Integration and Uncertainty Quantification

1. Scientific context

Prognostics and Health Management (PHM) plays a critical role in improving availability, reliability, safety, and reducing maintenance costs of complex engineering systems [1]. By predicting the Remaining Useful Life (RUL) of components and systems, PHM enables proactive maintenance, reduces unplanned downtime, and optimizes resource utilization. As modern industrial systems become more complex, with interconnected components and multi-sensor data streams, System-Level Prognostics (SLP), a pivotal subfield of PHM, has become indispensable across diverse sectors, including aerospace, energy, manufacturing, and critical infrastructure. Tamssaouet et al. provide a comprehensive review of SLP, highlighting its advantages in identifying a wide range of applications and the inherent challenges, particularly the need for robust methodologies to address uncertainty and component interactions [2].

One of the primary challenges in SLP lies in effectively modeling the dependencies and interactions among system components, which significantly influence degradation and failure modes. Conventional methods, such as model-based methods, have addressed these dependencies by leveraging physical prior knowledge of system dynamics. For example, the Inoperability Input-Output Model (IIM) integrates mission profile effects and online parameter estimation, offering dynamic RUL predictions for complex systems like the Tennessee Eastman Process [3]. Similarly, state-space models with diffusion coefficient matrices effectively capture the coupling effects of degradation in multi-component systems, employing techniques like Kalman filtering and Monte Carlo simulations [4]. However, as systems grow increasingly and sensor data become more high-dimensional, these models often face scalability challenges.

To complement model-based methods, data-driven approaches have gained prominence for their ability to integrate diverse data sources. Bayesian Networks (BNs), for example, provide a robust framework for modeling probabilistic relationships while incorporating historical data, real-time sensor inputs, and expert knowledge. This capability is crucial for evaluating the health of complex systems where interactions between components play a critical role [5]. However, BNs also face limitations, including reliance on high-quality historical data, computational complexity, and scalability issues in handling intricate systems with numerous variables.

To address the limitations of model-based and data-driven methods, hybrid approaches have emerged as a promising solution by leveraging the strengths of both. For instance, Eker et al. [6] emphasize that hybrid methodologies enhance robustness in prognostics by accommodating variability and uncertainties in real-world operational conditions. Similarly, Li et al. [7] highlight the integration of deep learning algorithms, such as convolutional neural networks, with physics-based models to improve RUL estimation. However, combining data-driven and model-based methods requires meticulous calibration and validation to ensure the hybrid framework accurately captures system dynamics and degradations. Another critical gap is the lack of comprehensive uncertainty quantification at the system level. While Nguyen et al. [8] proposed a probabilistic deep learning methodology combining probabilistic models with deep recurrent neural networks to predict RUL distributions of components and derive system-level reliability, their study assumes independent component degradation and does not account for interactions. This simplification limits the model’s applicability to systems with complex interdependencies, underscoring the need for further research to address these limitations in hybrid prognostic frameworks.

2. Thesis objectives

This thesis aims to address the critical challenges in SLP by developing advanced hybrid approaches that allow enhancing the robustness, scalability, and reliability of prognostics algorithms, ensuring their effectiveness and adaptability in increasingly intricate engineering systems and dynamic industrial environments. Building on the foundation of prior research, the proposed methods will introduce transformative strategies for robust data integration, efficient modeling of component interactions, and rigorous uncertainty management. Ultimately, the goal is to establish more accurate and scalable prognostic solutions capable of adapting to increasingly complex engineering systems and dynamic industrial environments.

In summary, the outcomes of this thesis will primarily contribute to advancing the theoretical foundation of hybrid prognostic modeling by bridging gaps between data-driven and physics-based approaches. This work aims to establish new paradigms in the scientific understanding of system-level prognostics and to contribute significantly to the broader research community.​

References

[1] Khanh T.P. Nguyen, Kamal Medjaher, Do T. Tran (2023). A review of artificial intelligence methods for engineering prognostics and health management with implementation guidelines. Artificial Intelligence Review, Volume: 56, Issue: 4, Pages: 3659- 3709.

[2] Ferhat Tamssaouet, Khanh T.P. Nguyen, Kamal Medjaher, Marcos Eduardo Orchard (2023). System-level failure prognostics: Literature review and main challenges. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability. Volume: 237, Issue: 3, Pages: 524-545.

[3] Ferhat Tamssaouet, Khanh T.P. Nguyen, Kamal Medjaher, Marcos Orchard (2021). Online joint estimation and prediction for system-level prognostics under component interactions and mission profile eects. ISA transactions 113, 52-63.

[4] Xi, X., Chen, M. & Zhou, D. Remaining useful life prediction for multi-component systems with hidden dependencies. Sci. China Inf. Sci. 62, 22202 (2019). https://doi.org/10.1007/s11432-017-9347-5

[5] Wang, X., Guo, H., Wang, J., & Wang, L. (2018). Predicting the health status of an unmanned aerial vehicles data-link system based on a bayesian network. Sensors, 18(11), 3916. https://doi.org/10.3390/s18113916

[6] Eker, Ö., Camci, F., & Jennions, I. (2019). A new hybrid prognostic methodology. International Journal of Prognostics and Health Management, 10(2). https://doi.org/10.36001/ijphm.2019.v10i2.2727

[7] Li, X., Ding, Q., & Sun, J. (2018). Remaining useful life estimation in prognostics using deep convolution neural networks. Reliability Engineering & System Safety, 172, 1-11. https://doi.org/10.1016/j.ress.2017.11.021

[8] Khanh T.P. Nguyen, Kamal Medjaher, Christian Gogu (2022). Probabilistic deep learning methodology for uncertainty quantification of remaining useful lifetime of multi- component systems. Reliability Engineering & System Safety. Volume 222, June 2022, 108383.

The LGP is a multidisciplinary laboratory that develops research activities in materials, mechanics, automation, computer science, electrical engineering, robotics and production sciences and techniques in the field of Systems Science and Engineering. Given its affiliation with UTTOP, its mission is to conduct and develop research in areas related to the training of ENIT general engineers and thus promote the link between training, research and technology transfer. The LGP is also part of the Université Fédérale de Toulouse Midi Pyrénées and shares a number of activities with the Institut National Polytechnique de Toulouse (Toulouse INP)

The research is organized most often in close connection with real problems of the socio-economic world by following a unit of view of the product/process type throughout the life cycle of the product, from design to dismantling.

The LGP relies on remarkable equipment, consistent with the needs of companies and the profile of engineers trained at ENIT.

Following its restructuring, in July 2021, two scientific departments will group the teacher-researchers and researchers of UTTOP according to two major themes:

    Scientific Department Mechanics-Materials-Processes

    Scientific Department Systems

• Educational Background
  • Master’s or Engineering degree in a relevant field (e.g., Mechanical Engineering, Electric-Electronic Engineering, Computer Science, Applied Mathematics).
  • Solid academic record, with coursework or project experience in areas related to signal processing, data analysis, or machine learning.
• Technical Skills
  • Strong foundation in probabilistic models and machine learning techniques.
  • Proficiency in programming language (e.g., Python)
  • Knowledge of reliability engineering is a plus.
• Language and Communication
  • Good command of English (both written and spoken) is required.
  • Knowledge of French is a plus but not mandatory.
• Research Aptitude
  • Demonstrated ability to conduct independent research and tackle complex problems.
  • Experience with scientific publications or conference presentations is an advantage.
• Interpersonal Qualities
  • Self-driven, proactive, and capable of working collaboratively in a multidisciplinary team.
  • Strong analytical thinking, problem-solving skills, and attention to detail.