プロフィール詳細

プロジェクトを作成

★★★★★

☆☆☆☆☆

★★★★★

☆☆☆☆☆

Dr. Panagiotis K.に依頼

Netherlands

PhD in AI for Predictive Maintenance | PHM/SHM, sensor analytics, deep learning for industrial & aerospace systems

プロフィール概要

Machine learning engineer with a PhD from TU Delft, focused on designing, training, and deploying model architectures. I have built and shipped LSTM/CNN models, attention-based and diffusion models, deep reinforcement learning systems, plus interpretable and Bayesian deep learning, across time-series, image, and tabular data. My peer-reviewed publications cover the above. I have also led production deployments as a founder of the startup Prognora, with working knowledge of transformer architectures and modern LLM tooling. Available for hands-on ML/DL engineering, model R&D, and research-to-production work.

専門分野

• ☆ Electrical Engineering
• ☆ Signal Processing
• ☆ Data Analysis
• ☆ Predictive Modeling & Analytics
• ☆ Machine Learning
• ☆ Deep Learning
• ☆ Python
• ☆ Software Development
• ☆ Web Application Development
• ☆ DevOps

サービス

• Writing
• Research
• Consulting
• Data & AI
• Product Development

Writing Technical Writing, Business & Legal Writing

Research Market Research, User Research, Feasibility Study, Technology Scouting, Scientific and Technical Research, Systematic Literature Review

Consulting Business Strategy Consulting, Scientific and Technical Consulting

Data & AI Predictive Modeling, Statistical Analysis, Image Processing, Image Analysis, Algorithm Design-ML, Data Visualization, Big Data Analytics, Data Cleaning, Data Processing, Data Insights

Product Development Product Evaluation, Product Validation, Concept Development, Prototyping, Reverse Engineering

職務経験

Co-Founder / CEO

Prognora

6月 2025 - 現在

Delft University of Technology

- 4月 2026

PhD Researcher

Delft University of Technology

5月 2021 - 5月 2025

PhD researcher

Delft University of Technology

5月 2021 - 5月 2025

学歴

Masters (Science)

University of Patras

9月 2018 - 6月 2020

Dipl. Ing. (Mechanical Engineering & Aeronautics)

University of Patras

1月 2015 - 1月 2020

認定資格

• Microsoft Certified: Azure AI Fundamentals

  Microsoft

  https://learn.microsoft.com/en-us/users/panagiotiskomninos-1988/transcript/7x5nzb80n240jwe?wt.mc_id=certnurture_eml14_email_wwl

  1月 2026 - 現在

• Accelerating End-to-End Data Science Workflows

  NVIDIA

  https://learn.nvidia.com/certificates?Yl6zLhHCTECPvMNciLmMkA

  12月 2024 - 現在

• Reinforcement Learning Specialization

  coursera

  https://www.coursera.org/account/accomplishments/specialization/certificate/6SDU578UKH76

  6月 2022 - 現在

• Bayesian Methods for Machine Learning

  coursera

  https://www.coursera.org/account/accomplishments/verify/RKTGWLJV6AQP?utm_source=link&utm_medium=certificate&utm_content=cert_image&utm_campaign=sharing_cta&utm_product=course

  3月 2022 - 現在

出版物

JOURNAL ARTICLE

Panagiotis Komninos, Thanos Kontogiannis, Nick Eleftheroglou, Dimitrios Zarouchas (2026). A robust generalized deep monotonic feature extraction model for label-free prediction of degenerative phenomena . Data-Centric Engineering.

P. Komninos, G. Galanopoulos, T. Kontogiannis, N. Eleftheroglou, D. Zarouchas (2025). A Bayesian inference-based framework for modeling imperfect post-repair behavior of remaining useful life under uncertainty . Expert Systems with Applications.

P. Komninos, D. Zarouchas (2025). A framework for post-prognosis decision-making utilizing deep reinforcement learning considering imperfect maintenance decisions and Value of Information . Array.

Panagiotis Komninos, Xin Yang, Sergio Cantero-Chinchilla, Morteza Moradi, Chen Fang, Yunlai Liao, Pradeep Kundu, Dimitrios Zarouchas, Dimitrios Chronopoulos (2025). Damage imaging in structural health monitoring with fine-tuned conditional diffusion model . Mechanical Systems and Signal Processing.

P. Komninos, G. Galanopoulos, T. Kontogiannis, N. Eleftheroglou, D. Zarouchas(2025). A Bayesian inference-based framework for modeling imperfect post-repair behavior of remaining useful life under uncertainty . Expert Systems with Applications. 288. Elsevier

Panagiotis Komninos, Thanos Kontogiannis, Dimitrios Zarouchas, Nick Eleftheroglou(2025). A robust generalized deep monotonic feature extraction model for label-free prediction of degenerative phenomena . Data-Centric Engineering. 7. Cambridge University Press

P. Komninos, D. Zarouchas(2025). A framework for post-prognosis decision-making utilizing deep reinforcement learning considering imperfect maintenance decisions and Value of Information . Array. 27. Elsevier

Constructing explainable health indicators for aircraft engines by developing an interpretable neural network with discretized weights @article{472b2745ccd940f69e22eade59c060eb, title = "Constructing explainable health indicators for aircraft engines by developing an interpretable neural network with discretized weights", abstract = "Abstract: Remaining useful life predictions depend on the quality of health indicators (HIs) generated from condition monitoring sensors, evaluated by predefined prognostic metrics such as monotonicity, prognosability, and trendability. Constructing these HIs requires effective models capable of automatically selecting and fusing features from pertinent measurements, given the inherent noise in sensory data. While deep learning approaches have the potential to automatically extract features without the need for significant specialist knowledge, these features lack a clear (physical) interpretation. Furthermore, the evaluation metrics for HIs are nondifferentiable, limiting the application of supervised networks. This research aims to develop an intrinsically interpretable ANN, targeting qualified HIs with significantly lower complexity. A semi-supervised paradigm is employed, simulating labels inspired by the physics of progressive damage. This approach implicitly incorporates nondifferentiable criteria into the learning process. The architecture comprises additive and newly modified multiplicative layers that combine features to better represent the system{\textquoteright}s characteristics. The developed multiplicative neurons are not restricted to pairwise actions, and they can also handle both division and multiplication. To extract a compact HI equation, making the model mathematically interpretable, the number of parameters is further reduced by discretizing the weights via a ternary set. This weight discretization simplifies the extracted equation while gently controlling the number of weights that should be overlooked. The developed methodology is specifically tailored to construct interpretable HIs for commercial turbofan engines, showcasing that the generated HIs are of high quality and interpretable.", keywords = "Artificial neural network, Feature fusion, Interpretable health indicator, Multiplicative neuron, Prognostics and health management, Ternary weights", author = "Morteza Moradi and Panagiotis Komninos and Dimitrios Zarouchas", note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. ", year = "2025", doi = "10.1007/s10489-024-05981-2", language = "English", volume = "55", journal = "Applied Intelligence", issn = "0924-669X", publisher = "Springer", number = "2", } . Applied Intelligence.

P. Komninos, A.E.C. Verraest, N. Eleftheroglou, D. Zarouchas (2024). Intelligent fatigue damage tracking and prognostics of composite structures utilizing raw images via interpretable deep learning . Composites Part B: Engineering.

Panagiotis Komninos, Dharun Vadugappatty Srinivasan, Morteza Moradi, Dimitrios Zarouchas, Anastasios P. Vassilopoulos (2024). A generalized machine learning framework to estimate fatigue life across materials with minimal data . Materials & Design.

A generalized machine learning framework to estimate fatigue life across materials with minimal data @article{78ef63b339ed4fd991345d91f24c42f3, title = "A generalized machine learning framework to estimate fatigue life across materials with minimal data", abstract = "In this research, a generalized machine learning (ML) framework is proposed to estimate the fatigue life of epoxy polymers and additively manufactured AlSi10Mg alloy materials, leveraging their failure surface void characteristics. An extreme gradient boosting algorithm-based ML framework encompassing Synthetic Minority Over-sampling TEchnique (SMOTE), categorical data encoding, and external loop cross-validation is developed to evaluate the fatigue life across materials. The influence of different training strategies based on materials, input features, encoding method, and data standardization on the model performance is explored. Additionally, the importance of anti-data-leakage and anti-overfitting measures over the ML model performance is addressed. The result shows that the data-leakage-free, external loop cross-validated model can estimate the fatigue life of selective epoxy polymers and metal alloys with an average R2 of 0.71 ± 0.06 using a mere 12 to 27 experimental data points per material category. Whereas the model trained with data-leakage and overfitting results in high R2 of 0.9.", keywords = "Composites, Fatigue, Machine learning, Metal alloys, Minimal data, Void", author = "Srinivasan, \{Dharun Vadugappatty\} and Morteza Moradi and Panagiotis Komninos and Dimitrios Zarouchas and Vassilopoulos, \{Anastasios P.\}", year = "2024", doi = "10.1016/j.matdes.2024.113355", language = "English", volume = "246", journal = "Materials and Design", issn = "0264-1275", publisher = "Elsevier", } . Materials and Design.

P. Komninos, A.E.C. Verraest, N. Eleftheroglou, D. Zarouchas(2024). Intelligent fatigue damage tracking and prognostics of composite structures utilizing raw images via interpretable deep learning . Composites Part B: Engineering. 287. Elsevier

Panagiotis Komninos, Christos Nastos, Dimitrios Zarouchas (2023). Non-destructive strength prediction of composite laminates utilizing deep learning and the stochastic finite element methods . Composite Structures.

Panagiotis Komninos, Christos Nastos, Dimitrios Zarouchas(2023). Non-destructive strength prediction of composite laminates utilizing deep learning and the stochastic finite element methods . Composite Structures. 311. Elsevier

DISSERTATION THESIS

P. Komninos(2025). An end-to-end framework for Prognostics and Health Management . TU Deflt

CONFERENCE PAPER

Panagiotis Komninos, Morteza Moradi, Rinze Benedictus, Dimitrios Zarouchas(2022). Interpretable Neural Network with Limited Weights for Constructing Simple and Explainable HI using SHM Data . Annual Conference of the PHM Society. 14. (1). {PHM} Society

M. Moradi, P. Komninos, R. Benedictus, D. Zarouchas(2022). Interpretable neural network with limited weights for constructing simple and explainable HI using SHM data . Annual Conference of the PHM Society. 14. (1). PHM Society