Auto OEMHow Co-Simulation and Design Exploration Streamline Hybrid Control Development

Long story short

Optimising the energy management system of a hybrid vehicle used to involve a great deal of trial and error. However, we have replaced this with a smart, automated approach that combines co-simulation and design exploration tools, and we have reduced fuel consumption by 13%. Even better, we eliminated the need for weeks of manual tuning, making the control development process faster, more reliable and far more scalable.

🔧 “You don’t need to guess when you can simulate and explore intelligently.”
Mathieu Dutré, Managing Director at CTRL engineering

The challenge: Too many parameters, no clear path

Tuning a simple control loop, such as a PID controller, is relatively straightforward. Over time, engineers have developed plenty of guidelines and rules of thumb for this purpose. However, advanced, rule-based control strategies, such as the supervisory energy management system of a hybrid hydraulic vehicle, present a more challenging task.

These control algorithms typically involve a large number of adjustable parameters and target maps. Each of these influences the system’s behaviour, and they often interact in non-obvious ways. The result? A vast design space in which small changes can have a significant impact and where there is no clear path to the optimal configuration. Not only is manually fine-tuning such a controller time-consuming, it’s practically impossible!

Our Approach: Co-simulation meets automated design exploration

At CTRL Engineering, we tackled this challenge by combining the power of simulation with automated design exploration.

First, we created a detailed model of the hybrid powertrain using Simcenter Amesim software. This virtual plant simulates the interaction between the mechanical, hydraulic and electrical systems. We implemented the supervisory control logic in Simulink, where we could manage all the tuning parameters. We connected these two environments in a closed-loop co-simulation setup, which allowed us to evaluate vehicle performance, especially fuel consumption, under realistic driving conditions using a full New European Driving Cycle (NEDC).

Then came the real breakthrough: we integrated HEEDS, an advanced design exploration tool. Rather than manually adjusting parameters and hoping for the best, HEEDS intelligently navigates the control parameter space. It automatically runs co-simulations, measures the resulting performance and updates its strategy based on previous outcomes. The best-performing controller configuration is identified through multiple iterations, with no need for manual intervention.

Results: 13% less fuel, zero manual tuning

By automating the tuning process, we achieved a 13% reduction in fuel consumption, bringing the vehicle’s performance down to 3.21 l/100 km on the NEDC cycle.

Just as importantly, we significantly reduced the time required for control development. What would have taken weeks of trial-and-error tweaking was now done in a fraction of the time and with greater confidence in the final result.

Why it matters to you? Smarter workflows unlock hidden performance

This case confirms the value of combining model-based engineering with automated design exploration. Complex control strategies don’t have to mean complex development workflows. By leveraging the right tools, we uncovered correlations between parameters and system behavior that would otherwise remain hidden. This leads to better system performance at a faster rate.

At CTRL Engineering, this is how we approach system optimization: not by guessing, but by smartly simulating, analyzing, and iterating.