AVL List, a company building powertrain systems (combustion engines, hybrid systems, electric drives), has developed a combustion controller that reduces test bed time for engine calibration by approximately 50 per cent, compared to implementation using manual programming. The real-time combustion controller was developed using MathWorks Model-Based Design tools.
Engine calibration involves controlling an engine to its optimum combustion position, which has traditionally been a slow process of trial and error with an engine running on the test bed. To address this challenge, AVL developed a combustion controller that drastically reduces test bed time for engine calibration teams. AVL built a system-level model in Simulink to verify the controller in simulation over its full range of operation and test the reaction to limit violations to protect the engine under test.
Real-Time Workshop and the AVL ARTE.Lab tool helped the team quickly generate code from the controller model and execute it on the test bed. In addition to reducing development time, AVL integrated a new test bed with the controller in two days and cut testing time by 80 per cent on a recent pilot project.
'Using MathWorks tools was critical to the success of our combustion controller, which will, in turn, enable our test bed customers to reduce testing time,' said Dr Klaus Rothbart, product manager at AVL. 'Integrating MathWorks tools gives us an advantage because a majority of our customers also use Matlab and Simulink, and can customise the combustion controller without additional ramp-up time.'
'With rising engine complexity, calibration remains a challenging and time-consuming task for automotive engineers,' said Jon Friedman, automotive industry marketing manager at MathWorks. 'We are encouraged to see AVL's success in test bed development as well as the new controller that helps calibration engineers gain testing efficiency using Model-Based Design and MathWorks tools.'
