Lydion Energy Research
Research
“Can future dynamics be learned and predicted without knowing when, where, or what disturbance occurred?”
Themes
- 01
Online neural dynamics forecasting
Neural ordinary differential equations trained on post-fault measurements while an event unfolds, to forecast frequency and other states and trigger corrective control in time.
- 02
Contingency screening and cascading failures
Reinforcement learning that ranks N-k contingencies under a fixed computation budget, and probabilistic estimates of blackout risk.
- 03
Machine learning with digital twins
How high-fidelity simulation models and fast learned surrogates can be combined for monitoring and control of dynamic security.
- 04
Low-inertia grids
Monitoring technology, stability challenges and data-driven methods for systems dominated by power electronics, with the IEEE PES working group on Big Data & Analytics for Transmission Systems.
Journal articles 3
2026
Towards low-inertia power systems: an overview of monitoring technology, challenges and emerging solutions
International Journal of Electrical Power & Energy Systems, vol. 179, 112021
Review written by the IEEE working group on Big Data & Analytics for Transmission Systems.
2025
Online Neural Dynamics Forecasting for power system security
International Journal of Electrical Power & Energy Systems, vol. 167, 110566
Abstract
The increase in variable renewable energy sources has brought about significant changes in power system dynamics, mainly due to the widespread adoption of power electronics and nonlinear controllers. The resulting complex system dynamics and the unpredictable nature of disturbances pose substantial challenges for real-time dynamic security assessment (DSA). Machine learning (ML) methods offer advantages in terms of computational speed compared to numerical methods and simulators. Offline-trained ML models, however, are limited by their training domain; e.g., they cannot easily consider various grid topologies and data changes. Neural Ordinary Differential Equations (NODEs) leverage the integration of neural networks and ODE solvers to enable continuous-time dynamic trajectory predictions from time series data, resolving the limitation on topological and data changes. This paper introduces the Online Neural Dynamics Forecaster (ONDF) workflow, designed to monitor and assess system security in real-time using multiple NODEs trained solely with local post-fault measurements. Through several case studies, we compare the regression and DSA classification capabilities of ONDF with various ML models. Our findings demonstrate that ONDF provides a novel and scalable approach for system operators to make informed decisions and apply corrective control actions based on predicted dynamics.
2025
Power system frequency monitoring and emergency control with neural ordinary differential equations
Sustainable Energy, Grids and Networks, vol. 43, 101815
Abstract
Increasing renewable energy supply and distributed generating sources in the power grid lead to lower inertia levels. Lower inertia combined with higher uncertainty in operation can cause drastic frequency fluctuations when a disturbance occurs. System operators must know whether the transmission system is secure against a disturbance. Real-time models attempt to predict the frequency in near-real time however require pretraining on a large variety of possible disturbances. Training models in real-time would not require pre-training as they are directly trained on the occurring disturbance. However, training in real-time is not feasible until now as fast-occurring system dynamics require shorter prediction (and training) times for security and operation which standard machine learning models are not capable of. For the first time, this work proposes a fast training strategy that learns Neural Ordinary Differential Equations (NODE) in near real-time directly on the occurring disturbance, simultaneously addressing the inaccuracy issue of model-based dynamic studies. NODE learns the dynamics or derivatives of an ODE system that standard ODE solvers can solve. NODE provides a continuous function for predicting future dynamics in a decentralized way, hence faster frequency stability assessment for longer time spans. We propose a collocation-based sampling using the collocation gradients. Case studies on the IEEE 39-bus system show the approach is feasible for near real-time operation, accurately predicts future system states, and enables operators to apply predesigned corrective control actions, potentially making the system secure for future disturbances.
Under review 1
2026
Sequential Contingency Finder for Power System Security Tasks
Electric Power Systems Research, under review
Abstract
The increasing penetration of variable renewable energy sources has significantly altered power system dynamics, compelling operators to analyze a wide range of credible contingencies to ensure security. The high variability of operating conditions continuously reshapes the critical contingency set, making exhaustive screening computationally intractable under a limited computation budget, especially for multiple contingencies. This paper proposes a reinforcement learning framework with an adaptive reward formulation for contingency prioritization as a sequential decision-making problem. The framework learns adaptive selection policies directly from power flow features and is applicable to both dynamic and static security formulations. Evaluations on the IEEE 39-bus system demonstrate the effective identification of critical contingencies across both security formulations, confirming the viability of RL-based screening as a practical tool for system operators to reduce the search space under multi-contingency scenarios.
Book chapter 1
2023
Machine learning and digital twins: monitoring and control for dynamic security in power systems
Monitoring and Control of Electrical Power Systems Using Machine Learning Techniques, Chapter 4, Elsevier, pp. 79-106
Abstract
The reader of the chapter will be able to connect techniques from machine learning (ML) and digital twins (DTs) to gain insights for monitoring and control of (dynamic) security for electrical power systems. DTs are validated and verified high-fidelity (hf) models providing high simulation accuracy. DTs can be used for simulation of the supervised process of system operation and are therefore able to provide synthetic studied data, where measurement data are scarce. However, for some real-time applications in monitoring and control, such high-fidelity simulation models are not appropriate due to the corresponding computational barrier. There, ML aims to create an application-specific, low-fidelity (lf) approximation of the digital twin. Such trained lf models are used in real-time applications where computational time is scarce and lf information is sufficient. The conceptual intersection of hf and lf models has been little explored and becomes increasingly complex. This chapter aims to provide a conceptual overview of how such hf and lf models can be combined.
Technical report 1
2026
Data-driven methods for secure low-inertia grids
IEEE PES Technical Report TR-140, IEEE Power & Energy SocietyChapter lead
Conference papers 2
2022
Prediction of possible power system blackout risk with machine learning algorithms
CIGRE Session, Paris
2018
A new index for voltage quality assessment at distribution systems
53rd International Universities Power Engineering Conference (UPEC), pp. 1-6
Talks, teaching, supervision
- 2026Panel speaker at IEEE SGSMA 2026, Santiago de Chile. Presented IEEE PES TR-140 for the working group, and “Neural Ordinary Differential Equations for Forecasting Power System Dynamics”.
- TU DelftSupervised MSc theses on probabilistic risk estimation, including “Probabilistic Framework for Assessing Cascading Failures in Power Systems” (A. Rajagopal, 2025), and on AI-based emergency frequency control.
- 2024Organising committee, DynPOWER 2024, international workshop on dynamic stability challenges of future power grids, Delft.
- TU DelftCo-developed the MSc course EE4C12 Machine Learning for Electrical Engineering and served as one of its head teaching assistants.