A Digital Substation in-the- Loop (DSiL) at INESC TEC
INESC TEC
Academia, Energy
07 / 17 / 2026

The Customer
The Institute for Systems and Computer Engineering, Technology and Science (INESC TEC) is a leading R&D institution in Portugal with international reach, focused on scientific research and technological development, technology transfer, advanced consulting and training, and pre-incubation of new technology-based companies. Its Centre for Power and Energy Systems (CPES) works in power and energy systems, including the large-scale integration of Distributed Energy Resources.
The Challenges
Realistic Interoperability: Validate multi-vendor interoperability under realistic network timing and load conditions.
- Deterministic Performance Assessment: Assess system performance with hardware-timestamped evidence.
- Advanced Protection & Control: Develop and validate advanced protection and control logic.
- Pre-Deployment Validation: Thoroughly validate and de-risk new applications before deployment.
The OPAL-RT Solution
In response to these challenges, INESC TEC is developing the Digital Substation in-the-Loop (DSiL) platform using OPAL-RT’s solution as the technological provider.
- Real-Time Digital Substation Platform: The DSiL platform uses an OP5707XG real-time simulator running EMT models with a 125 μs time step, supported by RT-LAB and HYPERSIM.
- IEC 61850 & Multi-Vendor Integration: Physical and virtual IEDs are synchronized through IEEE 1588 PTP, with IEC 61850 GOOSE, Sampled Values, and MMS communications.
- Repeatable HIL Validation: The platform combines physical hardware-in-the-loop, analog front-end amplifiers, automated campaigns, and structured test data for repeatable PAC validation.
The Results
- Large-Scale Validation: The Adaptive Protection Scheme (APS) was evaluated across 1,210 operating scenarios and more than 60,000 fault scenarios.
- Improved Protection Performance: The APS reduced the failure rate from more than 35% to 0.67% and achieved 99.9% fault clearance across the evaluated scenarios.
- High Classification Accuracy: The machine-learning classifier achieved 99.8% accuracy across 1,210 operational scenarios.
- Fast Real-Time Response: HIL validation demonstrated an average decision time of approximately 0.6 ms, with the overall classification and setting activation process completed in under 1 ms.



