PEPR TASE – DC-ARCHITECT

Research activity within PEPR TASE focused on AI-based diagnostics, prognostics and health monitoring of battery energy storage systems in AC/DC distribution grids.

Overview

This research activity is conducted within the framework of the French national research programme PEPR TASE through the DC-ARCHITECT project.

The work focuses on the development of artificial intelligence, diagnostic and prognostic methodologies for battery energy storage systems integrated into future AC/DC distribution networks and smart energy systems.

The project addresses one of the key challenges of future electrical infrastructures: ensuring the reliability, availability and resilience of storage elements operating under variable, distributed and highly dynamic conditions.

Funding

PEPR TASE – DC-ARCHITECT

Scientific Context

The increasing penetration of renewable energy sources requires new electrical architectures capable of integrating distributed generation, storage systems and flexible loads.

In this context, hybrid AC/DC distribution networks are emerging as a promising solution for improving efficiency, flexibility and resilience.

Battery energy storage systems play a central role in these architectures by:

  • Compensating renewable generation variability
  • Supporting grid stability
  • Enabling local energy management
  • Improving resilience during disturbances
  • Facilitating the integration of photovoltaic and other renewable energy sources

However, battery degradation, aging and operational uncertainty remain major limitations for their long-term deployment.

Main Objectives

The project pursues the following objectives:

  • Develop intelligent diagnostic methods for battery energy storage systems.
  • Estimate battery State of Health (SOH).
  • Predict degradation trajectories.
  • Support Remaining Useful Life (RUL) estimation.
  • Improve reliability of storage elements in AC/DC networks.
  • Integrate physics-informed and data-driven approaches.
  • Support resilient operation of future smart grids.

Methodology

The methodology combines battery modeling, data analysis, artificial intelligence and prognostics.

Hybrid diagnostic and prognostic framework for battery energy storage systems integrated into AC/DC distribution networks.

Axis 1 – Battery Monitoring

Monitoring of battery operating conditions through:

  • Voltage measurements
  • Current measurements
  • Temperature measurements
  • Operational profiles
  • Cycling conditions
  • Energy throughput indicators

Axis 2 – Health Indicator Extraction

Extraction of diagnostic indicators related to battery degradation.

Examples include:

  • Capacity-related indicators
  • Internal resistance indicators
  • Charge and discharge behavior
  • Thermal response
  • Dynamic electrical signatures

Axis 3 – Intelligent Diagnostics

Development of AI-based diagnostic methods for:

  • Abnormal behavior detection
  • Degradation pattern identification
  • Fault detection
  • Health-state classification
  • Operating regime characterization

Axis 4 – Prognostics and Decision Support

Development of prognostic tools to support:

  • State of Health estimation
  • Remaining Useful Life prediction
  • Maintenance planning
  • Storage management
  • Resilient operation of AC/DC networks

Scientific Contributions

The project contributes to:

  • AI-based diagnosis of battery storage systems
  • Hybrid modeling for energy storage monitoring
  • Prognostics and Health Management (PHM)
  • Intelligent energy management
  • Resilience of AC/DC distribution networks
  • Integration of battery diagnostics into smart-grid supervision

Expected Outcomes

The expected outcomes include:

  • Battery health monitoring methodologies
  • Diagnostic and prognostic algorithms
  • AI-based indicators for storage systems
  • Decision-support tools for grid operation
  • Contributions to PEPR TASE scientific objectives
  • Scientific publications and collaborative research outputs

Scientific Impact

PEPR TASE – DC-ARCHITECT contributes to the development of advanced technologies for future energy systems by improving the reliability and intelligence of storage elements in hybrid AC/DC electrical architectures.

The methodologies developed in this framework may also be transferred to:

  • Photovoltaic systems
  • Microgrids
  • Smart grids
  • Electric mobility
  • Second-life battery applications
  • Industrial energy systems

Keywords

PEPR TASE · DC-ARCHITECT · Battery Energy Storage Systems · Artificial Intelligence · Fault Diagnosis · Prognostics · State of Health · Remaining Useful Life · AC/DC Networks · Smart Grids · Energy Resilience · Predictive Maintenance

Project Status

Ongoing

References