Mr. Beggiora, what makes the Safran ENGINeUS™ 100 electric motor so special? 

The ENGINeUSTM 100 is the world's first electric motor to be certified in accordance with EASA Special Condition E-19 for electric and hybrid propulsion systems. The engine is designed for use in all-electric aircraft carrying two to four passengers. We are already working with several clients on this. In addition, the project lays the groundwork for an expansion to higher-power applications of up to 180 kW, targeting both hybrid-electric or full battery platforms for aircraft with up to 19 seats using a full electric propeller like ENGINeUSTM 100. This is a major Safran priority for decarbonizing aviation.

What exactly does EASA Special Condition E-19 entail?

Special Condition E-19 provides a modern certification framework that enables new propulsion architectures for electric and hybrid aircraft propulsion systems. It was developed because traditional regulations, which apply to turbines and internal combustion engines, do not cover new electric and hybrid propulsion systems. The SC E-19 certification is intended to enable innovative technologies while ensuring a level of safety comparable to that of traditional turbine/combustion-engine aircraft.   

What requirements did the engine have to meet to be successfully certified? 

Essentially, the goal was to test and validate the embedded software and the functions of the ENGINeUSTM 100 Engine Control System with regard to the stringent requirements for certification and operational safety. It was necessary to demonstrate that the Engine Control System is capable of adequately detecting and correcting any faults in order to prevent dangerous or even catastrophic consequences for the engine or the aircraft. Since this is a highly advanced motor control application which features two control channels (COM) and an independent monitoring channel (MON) with high-frequency control and communication requirements, closed-loop testing using HIL is essential to validate the software and demonstrate safe integration, in particular, the system’s ability to detect and respond to introduced faults.

What challenges did the strict certification guidelines pose for your work, and how did you overcome them?

One of the biggest challenges was developing a HIL test bench capable of stimulating and capturing all control system interfaces and executing the ENGINeUSTM 100 model at high frequency in real time. In addition, we needed a way to inject internal ENGINeUSTM 100 errors through simulation. At the same time, the test execution should be automatable in order to operate with maximum efficiency. Our team also had to master the dSPACE tool chain, develop customer-specific automation solutions, and involve dSPACE as needed for specialized development projects. To achieve all of this, we conducted an internal benchmark of all existing test benches at Safran Electrical & Power for electric motor control. Based on dSPACE solutions, our engineers developed all automation tools in-house. Support from dSPACE was required only occasionally for specific functionalities. 
 

In general, what approach did you take in your work? 

We rely on an end-to-end development and testing process that combines model-based design (MBD), model-in-the-loop (MIL) and hardware-in-the-loop (HIL) simulations, and physical test runs. The HIL test benches are fully integrated into our MBD process. The test cases are first executed in the MIL simulation and then replicated on the HIL test bench under identical conditions (Figure 1). This is how we ensure that the actual control hardware behaves as expected in the same scenarios and that the simulation results are reliably validated. 

Can you describe the range of test cases?

The test cases covered software verification, validation of the engine control system, and error analysis and troubleshooting. They were developed jointly by the software and systems development teams, in accordance with the ED-12C / DO-178C guidelines and the SC E-19 Special Condition for Electric/Hybrid Propulsion Systems.  

What was the deciding factor in your decision to use a dSPACE SCALEXIO HIL simulator?

Among the most important criteria were the high real-time performance for high-frequency control loops – made possible by FPGA technology – broad support for hardware interfaces such as CAN, RS485, SPI, GPIO, and I2C, and the ease of integrating our models into the FPGA environment. In this regard, in particular, dSPACE clearly stands out from other providers. This allowed us to model the operating environment of the electronic control unit, including the inverter, electrical systems, aircraft control systems, and relevant external components. In addition, we can simulate failure scenarios that we would prefer not to test on actual hardware in order to avoid destructive testing. With this solution, we were able to perform real-time simulations at the microsecond level or below and accurately model complex components such as electric machines and inverters.  
 

It sounds as though the dSPACE tools played a key role in your work?

Absolutely. The dSPACE tools enabled and significantly accelerated the verification and validation of our engine control system. Without the SCALEXIO-based test bench, it would have been nearly impossible to meet our ambitious timeline and successfully complete the certification process for the electric motor by the end of 2024. The HIL test bench was operated for six months, six days a week, at our main design office in Créteil. It took only a short time for two or three of our employees to become familiar with the dSPACE tools, allowing the entire team of about 20 engineers to quickly adopt the automated test workflows.

What makes the dSPACE tools stand out?

Several things come to mind right away, first and foremost the enormous time savings. By automating the test procedures using Python scripts and dSPACE automation functions, we were able to drastically reduce the duration of the test campaigns, from 2-3 months using manual methods to just one week. We have expanded the number of HIL test benches from two in 2024 to seven in just 2-3 years, all of which support our electric motor drive applications. In total, we have conducted several thousand hours of testing on these test benches. The dSPACE tools have significantly accelerated software verification and enabled robust validation of the control system, particularly in error-injection scenarios. This allowed us to cut our test verification schedule in half and drastically reduce, or even eliminate, the integration errors during physical test integration in order to achieve right first time.

What are your next steps in this project?

Based on our very positive experience to date, we plan to further expand our model-based design process and integrate HIL testing into all electric engine, motor, or generator control applications to propose robust, safe, and competitive solutions for the next generation of electric aircraft.

 

Thank you for talking with us.