Innovation et gestion de programmes synchronisée et collaborative pour les nouveaux programmes
Renault is a French multinational vehicle manufacturer that was established in 1899, and produces a range of cars and vans. In 2012, Renault employed more than 127,000 people, had a presence in 128 countries and realized global after-sales revenue of over 4 billion euros.
“Drive the Change” is the new brand tagline for Renault, one of the world’s largest automotive companies. It reflects Renault’s ambition to make sustainable mobility accessible to all, and is a daring vision that affects the company culture, organization and processes. Supported by a strong brand image, Renault seeks to foster international growth by optimizing all of its resources.
Renault is a global group that designs, manufactures and markets vehicles under three different brands: Renault, Dacia and Renault Samsung Motors. It operates in 115 countries so it may appear as a maze of administrative departments, research and development (R&D) centers and production sites. However, resource optimization is the guiding principle at all levels of the company; initiatives are pursued to dramatically streamline the vehicle development process.
Renault has been pioneering system simulation for more than 10 years so Simcenter Amesim™ software from Siemens Digital Industries Software has been widely deployed across the engineering departments. Attributes such as drivability, reliability and noise, vibration and harshness (NVH) are optimally assessed and balanced in early design stages. The engine design department in particular fully exploits the software to optimize engine performance and fuel economy. But whereas system simulation is successfully used for engine development, the models outlined by the controls department lack some maturity.
As a result, undertaking a modernization of tools and processes is vital to develop more innovative products that will enable the integration of complex technologies and comply with ever stricter standards.
The upcoming Euro 6 and Euro 7, the European emission standards that define acceptable exhaust emissions limits for new vehicles, have increased the workload of engine design and control design engineers. They now have to integrate complicated engine actuation and control systems that will help optimize engine performance with minimal emissions. The growing number of systems that are regulated by controls and the multiple interactions of those systems complicate the tasks as engineers need to take a higher number of degrees of freedom into account.
Complex technologies and stringent standards are not the only difficulties. It is a challenge to develop efficient teamwork with a group of engineers that speak different languages, use different codes and have different rules. How do you ensure smooth and seamless collaboration of engineering teams scattered over the world?
Renault is a global group that faces the challenge of facilitating cooperation. France centralizes all the engineering departments, Romania focuses on hardware- in-loop (HiL) test benches; India works on coding and validation; and Spain and Korea get the tuning tasks. There’s an amalgam of different cultures, disciplines and expertise that makes sharing knowledge a challenge, and makes sharing heavy or fragile prototypes nearly impossible.
Even misunderstandings threaten the teamwork of the French engine design and control development teams. Although they have a common goal, their tools and methodologies differ. Engine design engineers optimize engine parameters in steady-state operating conditions with the purpose of getting the highest performance figures. For example, their simulation codes have to accurately predict the gas dynamics in the air path system in order to get relevant information about the engine volumetric efficiency and in-cylinder gas composition.
The engineers in charge of control design, testing, validation and calibration also expect accurate and predictive models that are capable of being used for extrapolation and interpolation. At the same time, they want to be able to simplify complex models. In fact, control design engineers need flexible multiphysics models. Although engineers appreciate detailed models in the control definition phase, real-time validation requires simpler models, as detailed models would dramatically slow down computation times.
Renault engineers need a tool that can be used worldwide for efficient virtual prototyping and is standardized so engineers can easily share and exchange models. Those models should certainly be accurate, but also sufficiently flexible so that different teams with varying tasks can adapt them to their specific needs. Simcenter Amesim models can easily be altered in numerous versions, with an incorporated version management feature, or include a deeper level of component details. In turn, those detailed models can easily be adjusted to simpler models while remaining extremely accurate. This is why Renault standardizes on Simcenter Amesim as the multidomain system simulation tool for the entire control development process. In a sense, Simcenter Amesim is the common language that all engineers across regions and departments should speak to streamline development. Simcenter Amesim is already the trusted tool that supports the Renault engine team in every stage of engine development. Indeed, the use of the software is not limited to optimized engine design. It can also be deployed in the other phases of the development process, like testing and validation of the control strategies in software-in-the-loop (SiL) or HiL environments. Control engineers easily access the 1D high-fidelity plant models that are generated by the mechanical engineering teams in charge of the design of the engine and its subsystems. The interaction of models lets engineers balance their different, and sometimes conflicting, needs in terms of model detail levels.
The use of Simcenter Amesim enables Renault to seamlessly integrate plant models in the control environment. Renault engineers interface the control models to the plant models developed using Simcenter Amesim or import those into the Simulink® environment.
In the control design phase, it’s a click-and-go task in which the incorporated co-simulation feature means engineers simply connect the model created using Simcenter Amesim to the control model.
In the control validation and testing stage, the engineer will select the most suitable technical option, depending on the development status of the control. Inserting a full plant model created using Simcenter Amesim into the control environment, the engineer proceeds with a model-in-theloop (MiL) calculation only if a model of the complete engine control is available, or a SiL calculation is available when the C-code for the control has been generated, or goes for HiL, coupling the physical engine control unit with the computed model.
What sounds like a simple matter is an opportunity to save significant costs. It is all about trading physical engine test benches and physical engine prototypes for virtual engines on virtual test benches. Physical prototype testing on engine test benches is an extremely expensive step in the process. Cut this step out of prototype iterations, and it becomes a far less costly proposition. Renault engineers take full advantage of this process optimization. Techniques such as co-simulation, MiL, SiL and HiL are used across all of Renault Group’s brands for the development of all engine types as well as for gear boxes.
Standardization on Simcenter Amesim has resulted in significant benefits. Renault engineers can tap into a bank full of licenses, adapted libraries (such as IFP-Engine or IFP-Drive) and options, such as real-time and Simulink co-simulation.
“Standardizing on Simcenter Amesim opens a whole new world of possibilities,” says Vincent Talon, modelization engineer at Renault. “At first, we were the only two engineers to centralize modelization know-how. Nowadays, with global standardization, our Romanian colleagues implement engine models in the development loop or even perform minor improvements on them. Eventually, this team will help us validate models of existing engines or even create entirely new models.”
The use of Simcenter Amesim appears to provide benefits beyond technology, including acting as an efficient resource management tool.
“Engineering with Simcenter Amesim lets the Renault team members concentrate on their core activity, which is innovation,” says Talon. “The standardized use of Simcenter Amesim radically improves our level of efficiency. However, the main benefit lies in the huge cost savings that we realize with the tool. It not only drastically reduces test bench usage, but the upstream design of control laws has led to an overall quality improvement, enabling Renault to reduce warranty costs in the lifecycle of its automobiles. Being an off-the-shelf product, Simcenter Amesim also spares us the development and maintenance costs of property codes.”
Renault and Siemens Digital Industries Software look forward to working together in the future. Renault engineers believe that modelbased system engineering is a meaningful engineering approach. Further process improvements are on their way.
“Just consider the challenge of model identification,” says Talon. “With more and more engine variants under development, multiple engine models need to be validated. This explosion of models and variants poses the problem of identification or even re-identification of existing models. This project with Siemens Digital Industries Software has enabled us to reduce the time required for complete model identification by a factor of five, from 50 days to only 10. Our next objective is to reduce this process to two days.”
That’s an ambitious target that reflects the commitment of Renault engineers to driving the change of the automotive world.