This research introduces a real-time distributed co-simulation architecture designed for modern electrical systems. The framework seamlessly integrates MATLAB/Simulink for dynamic system modeling, the OPAL-RT OP4200 platform for rigorous Hardware-in-the-Loop (HIL) simulation, and Connext® as the high-performance communication layer. By executing Simulink models on OPAL-RT’s dedicated digital simulator hardware and leveraging RTI Connext to share data in real time via a publish-subscribe model, this architecture provides a highly robust methodology for evaluating complex electrical grids under strict distributed and real-time constraints.
The proposed co-simulation framework is structured into two primary nodes, entirely decoupled and interconnected by Connext:
Node 1 (Master & Control): Integrates the Master Subsystem (MS) responsible for power generation and the Control System (CS). It calculates and publishes the system's state variables.
Node 2 (Electrical Load): Models the consumption device (e.g., an R-L-C network). It is excited by the Thévenin equivalent voltage source (V_MS) received from Node 1 and dynamically returns the demand current (I_demand) back to the co-simulation loop.
To ensure deterministic, real-time interaction, communication is governed by a strict signal contract divided into two distinct flows: Flow 1 (Node 1 --> DDS) and Flow 2 (Node 2 --> Node 1). Both flows use DDS to communicate between them.
This architecture leverages Quality of Service (QoS) policies of Connext to optimize data delivery for HIL testing. The communication layer is explicitly configured to prioritize the most recent data sample and completely eliminate message queuing. This ensures that the simulation loops never suffer from stale data or artificial delays, maintaining strict temporal alignment between the physical hardware and the digital models.
In summary, this work successfully demonstrates the feasibility, scalability, and good performance of a distributed real-time co-simulation architecture built on Connext and OPAL-RT. By establishing rigorous experimental guidelines, this research quantifies the direct impacts of network latency and sampling granularity on both temporal and electrical metrics. Ultimately, it offers a highly replicable, industrial-grade foundation for future studies in smart grid integration and distributed power systems.
code
I see dds claim it is "real-time". To me this implies "deterministic". How is this accomplished via Ethernet, which inherently is not deterministic?
I know DDS can try to ensure data at a certain rate and notify you of a failure, but thats still a non-deterministic failure.
How does DDS use the term "Realtime"?
NOTICE: Any content you submit to the RTI Research Community Portal, including personal information, is not subject to the protections which may be afforded to information collected under other sections of RTI's Web site. You are entirely responsible for all content that you upload, post, e-mail, transmit or otherwise make available via RTI Community Portal. RTI does not control the content posted by visitors to RTI Community Portal and, does not guarantee the accuracy, integrity, or quality of such content. Under no circumstances will RTI be liable in any way for any content not authored by RTI, or any loss or damage of any kind incurred as a result of the use of any content posted, e-mailed, transmitted or otherwise made available via RTI Community Portal. Read the complete Terms prior to use.
Please see RTI's privacy policy and cookie policy if you have questions about any information collected during the sign-up process.
