Rapid prototyping of an e-driveline at MOLcy

Get our
casebook

By Frederick Duyvejonck, Automation Engineer at Molcy and Vincent Theunynck, Partner at Vintecc

Our competences

  • Matlab/Simulink Modeling
  • Rapid Prototyping
  • Automatic Code Generation

Choosing the right development approach accelerates development enormously and allows for more complex and more performant machine control.

Vincent Theunynck

2min read

MOLcy, located in Staden, BE, distinguishes itself as a company that does not shy away from technical challenges: more than 400 motivated members are cooperating together for the development of a high quality range of products. The company produces a range of trailers, port equipment, rail equipment, waste systems and special trucks. There is also more research for ecological solutions. The development of a full electrical terminal tractor fits this strategic scope perfectly.

The Problem 

An electrified drivetrain has some fundamentally different characteristics compared to a Diesel powered drivetrain.

  • With maximum torque available at zero speed, it is a challenge to gently release all this power from standstill.
  • Likewise, whereas a diesel powered drivetrain uses mechanical friction for breaking, the electrified version should use regeneration as much as possible to optimise efficiency.   

The challenge of this situation is to control the drivetrain very naturally with a one pedal drive (OPD), very similar to a diesel operated vehicle, though very different in nature. The goal for the control algorithm is to give a very ‘natural’ feedback to the operator, as if he was driving a diesel since this is what most operators are used to. The challenge is to get the right mix of acceleration, idling and regeneration at just the right vehicle conditions and operator input. 

In short, such a development requires a lot of testing, tuning and adaptation. Molcy also needed a solution very rapidly. This is a perfect case for a rapid prototyping workflow supported by Vintecc. 

The Solution 

Molcy presented an OPD approach which proved to be a very good starting point. 

First, a Matlab/Simulink model was created. The model was extended with a CAN harness to control the vehicle drivetrain using Vehicle Network Control Toolbox. In this way, parameters could be easily changed and new iterations could be tested. Particular challenges such as (1) avoiding torque jitter at standstill, (2) precise and gentle control of the drivetrain while picking up and dropping of a trailer and (3) switching between forward and backward driving in a smooth way, were perfect cases to develop with this rapid prototyping workflow.  All these situations could be evaluated and tested in a very short time frame.

Figure 2: Control of vehicle from Matlab/Simulink using Vehicle Network Control Toolbox 

Next, the control logic developed as Matlab/Simulink models were configured for code generation with PLC coder. The automatic code generation delivers a function block for integration in the onboard vehicle PLC (IFM electronic). The code was optimised for maximum reusability is generated in seconds. 

Finally, the outcome was tested against the original Matlab/Simulink model using a bypassing strategy.  

The Result 

The new OPD driveline control from idea to fully working implementation took about 2.5 weeks. In this short time, 80% of the effort could be attributed to functional development, while only 20% was spent onto the code implementation and testing. The OPD control was very well perceived by management and customers.  

The Advice 

Choosing the right development approach accelerates development enormously and allows for more complex and more performant machine control.

Vintecc masters these development methods and can help accelerate your innovations and significantly shorten time-to-market!

Choosing the right development approach accelerates development enormously and allows for more complex and more performant machine control.

Vincent Theunynck

2min read

Our competences

  • Matlab/Simulink Modeling
  • Rapid Prototyping
  • Automatic Code Generation

Other cases you might find interesting

Our solutions

Software programming / Sensor technologies / Automatic code generation

Process simulation / Virtual commissioning / Automated testing

Data collection & exploration / Reporting & alerting / IoT & cloud

Machine learning / Data driven insights / Vision systems