1. Presented by Marcel Helmer and Patrick Meuth
Integration and evaluation of MBT in a platform based software development process
Presented by Marcel Helmer and Patrick Meuth
© All rights reserved

2. Advanced Automated Testing
Agenda
Presenters Introduction
Goals
Challenges:
Usage in different platforms
Usage in different vehicle types
Spreading MBT over different V-Model testing stations
User Conference on
Advanced Automated Testing
© All rights reserved

3. Advanced Automated Testing
Presenters and Authors
Marcel Helmer (TKI Automotive GmbH)
Function Development Engineer
Patrick Meuth (TKI Automotive GmbH)
Test Development Engineer (Model Based Testing) on HiL Testing
Jorge Pascal (TKI Automotive GmbH)
Technical Lead (HiL & Test Automation)
User Conference on
Advanced Automated Testing
© All rights reserved

4. Advanced Automated Testing
Introducing V-Model for model-based Software Development
Software Requirement
Software System Test
Software Architecture
Software Integration Test
Specification
SW-Testing
Software Modeling
Software Unit Test
Software Implementation
User Conference on
Advanced Automated Testing

5. Advanced Automated Testing
Goal
Battery- and
Plugin Hybrid electric vehicles (BEV, PHEV)
Complex thermal manage-ment systems
Goal
To find synergies in order to use the same MBT usage Model
User Conference on
Advanced Automated Testing
Complex cooling systems
With different Components and circles.
We want to test these diffierent cooling circles with one Model
Thats why we have to find synergies between these Systems
 To put it in one Model

6. Hybrid Vehicles Challenge: Multiple Variants
Environment
Restbus- and Simulation values
Approach of Valves/Pumps
Evaluation
Env. Temperature
Device Temperatures
Env. Humidity
Device States
Configurations
Challenge solved by reading parameter values dynamically from ECU Data sets!!!
Read parameters (e.g. threshold value) via calibration software
Parameterize the test globally with these values (e.g. cooling demand)
Implement values (e.g. temperatures, test evaluation) based on the read parameter
Process
User Conference on
Advanced Automated Testing
Here are some requirements listed, that show you the important parts of the model.
Depends on them, our Thermo Management ECU response with the right answer.
The first of this three parts contains the environment requirements. The environment Temperature and humidity allows us, to force the ECU into the right condition.
The values of these Parameters, depends on the particular variant.
The Second part is the Restbus and Simulation values, who need to be set. There is for example the Device Temperature of the DCDC or Electrical machine.
Device States who show the availabilitys of different devices.
The Evaluation contains and has differences in the approach of Valves and pumps and also in the configurations.
How do we face this Challenge??  automation of Dataset reading ~ About 30 Dataset in summary
I will give you an Example: If we want to give a cooling requirement on a device like the Batterie, we get the limit for the demand out of the Dataset, that we can read with our calibration Software CANape, who allows us to read the ECU internal signals.
After this we set the temperature over the limit, instead of setting a fixed value.
_______
When we take a look at the Failure analysis, we see, that there were problems with the Restbus and Simulationsvalues. We had the change of a scope for a particular signal, and the ECU was not adjusted on the new scope, so we had a Error Memorie and also a Function barrier as consequence.
Addiational failures were detected on the approach of the valves and pumps. These contains overcharge of the pump speed and faulty valve relays.
IST-Wert = Actual value
Soll-Wert= desired value
Lastenheft= specification book, product concept catalogue
Gleichstrom= direct current
Wechselstrom= alternating current

7. Advanced Automated Testing
Challenge: Plugin Hybrid  Electrical Vehicle
Additional functions should be tested
 Battery Heating
 Thermo Management Controller Fault Reaction
Extend the Model Structure with new states and transitions
User Conference on
Advanced Automated Testing
Additional functions to cool down and heat up the system. For example there is the Batterie heating.
For this we dont need to change something on theactual modell, we just have to extend the actual modell with new transitions and States.
An other aspect is, that we have to take a look at the failure values of the devices. We set a lot of Values in our precondition, like the power of different devices etc. We have to take a look that there are no limits exceeded.
Active battery cooling: with the chiller, who is the heat exchanger, if the heat cant get away at the condenser.
Heatpump: just for cabin to heat or reheat (dehumidify)

8. Challenge: Plugin Hybrid  Electrical Vehicle
Additional CAN-Bus, LIN-Bus and ECU internal Signals
 More Electronical devices
Different Systemconfigurations
Precondition>Coolingdemand
Postcondition>Cooling>Driving
Challenge solved by extending the model to cover all possible variants (Super Set Model)!!!
User Conference on
Advanced Automated Testing
New specification needed: Additional Cooling devices: 2-3 Electrical machines, new Systemconfigurations, the Valves are no longer High side out with binary bypass valves, but LIN valves with a control range from 0-100%.
After this we go inside the Model and take a look to our cooling Demand Electrical machine. First we just had one Device where we set the cooling demand.
Now we have more than one. So we make an additional Transition and set on the one EM1 up to cooling demand and on the other a normal operating temperature of 20 °C.
In contrast to the other transition.
On this page you can see, how the Structure of the model is constructed: In the Precondition state we got an other state inside who set the cooling demands. So we have propper structures through the model and model structures who allow us, to make changes on just the affected parts.
Additionally we have the new valves. We changed the variant of the switching valve to a mish water valve who is evaluated.

9. Advanced Automated Testing
Spreading MBT over different V-Model testing stations
Software Requirement
Software System Test
Software Architecture
Software Integration Test
Specification
SW-Testing
Software Modeling
Software Unit Test
Software Implementation
User Conference on
Advanced Automated Testing

10. Difference in Testing Level
Testing Bench
Actual Plant Model,
Bus Simulation
Software System Test
(SW including Control Unit)
HIL (Bus- and Enviroment Simulation)
Bus Simulation , Actuators and Simulated Plant Model
Mini HIL (Bus-Simulation)
Bus Simulation,
Actuators assembled
Software Integration Test (C-Code on PC)
SiL
Application SW without basic SW and ECU
Software Unit Test
(Testing Simulink Model)
MiL
Model Test on Modul Level
User Conference on
Advanced Automated Testing

11. Spreading MBT over all HIL testing levels
Automation
Simulation Platform
Usage Model
EXAM
Control Desk
Test Bench with actual plant model
Conditioning of test bench
Usage Model
EXAM
Control Desk
HIL (Bus- and simulated plant model)
Usage Model
EXAM
CANoe
Mini HIL (Bus-Simulation)
User Conference on
Advanced Automated Testing

12. Spreading MBT over MiL & SiL testing levels
Automation
Simulation Platform
Compatible
Usage Model
EXAM
TPT
Software Integration Test
SiL
Interface
Usage Model
EXAM
Control Desk
HiL (Bus- and Environment Simulation)
New Model
Mapping
Usage Model
EXAM
Matlab
MiL on Modul Level
User Conference on
Advanced Automated Testing

13. Reusability of the model and variant managment
Summary
Variant management
Vehicle development
- Dynamic data set reading
- Super set Model
Automated MBT Process covers …
- Adapting existing model
 3 times faster than building up a new model
Platform development
- Integration of MBT in the new SW development process
Reusability of the model and variant managment
User Conference on
Advanced Automated Testing

14. Advanced Automated Testing
Thank you for your attention! Q&A Time! Time for Questions and hopefully also for some Answers…
Special Thanks to our Colleagues from Assystems, Audi (EK-4, EE-I3), dSPACE and Micronova
User Conference on
Advanced Automated Testing