Search Results

Today, the battery development process for automotive applications is relatively decoupled from the vehicle integration and system validation phase. Battery pack design targets are often disregarded at very early development phases even though they are thoroughly linked to the vehicle-level requirements such as performance, lifetime and cost. Here, AVL proposes a methodology guided by virtual testing techniques to frontload vehicle-level validation tasks in the earlier phase of battery pack testing. This paper focuses on the benefits of the methodology for both battery suppliers and automotive OEMs. Applications will be explained, based on a modular virtual testing toolchain, which involves the simulation platform and models as well as the generation of model parameters and test cases.

BS-VI or equivalent development calls for tremendous efforts in concept investigation and calibration for engine out, after treatment, diagnostic checks, off-cycle emissions, field performance, component safety etc. Achieving calibration quality for all these tasks is very challenging considering development time and cost with conventional physical testing approach. Present article focuses on assessment of testing and calibration using virtual approach. To prove and validate this approach, a six-cylinder heavy duty diesel engine is selected and configured in HiL environment. The engine plant model is built offline and validated with base engine data at steady state and transient operations and RT model is integrated with ECU hardware. Data for plant model corrections is generated with short measurement campaign. Refined real time plant model is prepared for evaluating different calibration strategies on virtual test bed environment.

Modern day diesel engines use systems like Exhaust Gas Recirculation (EGR), Variable Geometric Turbo Charger (VGT), inlet throttle for air regulation, multiple injection strategies, high pressure rail systems for fuel regulation to optimize the combustion for meeting the strict emission and fuel consumption demands. Torque based ECU structures which are commonly used for diesel engines require a large amount of calibration work. Conventional manual methods for emission and fuel consumption optimization (Full factorial or Line search method) results in increased test bed usage and it is almost impossible to use these methods as the number of parameters to optimize are very high. The conventional DoE tests have been limited by the necessity of calibration engineer’s expertise and manual prescreening of test points to be within thermal & mechanical limits of engine systems. This subsequently leads to excessive screening of variables; which is time consuming.