Abstract
In recent years we have seen a rising use of electronic components in cars which a highly responsible for the increased value-added of our automobiles. In particular automatic transmissions tend to increasingly rely on electronic systems for intelligent control. The so-called electronic transmission control (ETC), which very decisively helps improve such customer-specific criteria as shifting comfort, consumption and emission as well as driveability, have in the meantime become standard for upper category vehicles.
Short development times, early verification of design and function, and efficient use of development expenditure make it necessary for the step from the real world to the virtual world to be taken at an early stage of the project. These complex conditions and the state-of-the-art technology can only be mastered and verified with the aid of new methods, tools and test procedures. The hardware-in-the-loop (HIL) simulation, which the paper/presentation concentrates on, is at the moment certainly one of the most favoured methods used to meet the conflicting targets, and also to successfully meet the challenges of an increasingly competitive situation.
The presentation describes the configuration and implementation of the hardware-in-the-loop real-time simulation at the Institute of Vehicle Engineering of the Technical University Braunschweig, which allows an electronic transmission control unit (TCU) to be operated as a real and physical hardware in its functional environment. Due consideration being given to major ETC functions, a modular model environment has been created for the control unit. The complete
TCU environment of the control unit the so-called DRV parameters driver, route and vehicle is replaced by computer simulation in such a way that a closed control loop is produced. This at the same time considers an exchange of electrical signals and variables with the peripheral control units by means of the CAN (controller area network) data bus.
The virtual real-time environment makes possible two kinds of simulation. The state-controlled simulation uses predefined test scenarios to cause a specific vehicle and driving condition in a reproducible manner. The real-time system responds to changes in values defined for specific points of time. Examples of how control variables can be integrated into the simulation are given, a comparison being made with measurements made in practice. This especially concerns the TCU controlled converter clutch. The control travels are replaced by a real-time simulation while the control processes still take place in the physical control unit.
The virtual real-time simulation environment thus created does not only allow individual ETC functions to be tested under absolutely reproducible conditions; it also provides for statistical long-term driving simulation within the DRV parameter space in order to investigate driving strategies and adaptive functions. Unlike the state-controlled and reproducible simulation, the so-called event-controlled driving simulation means therefore that the real-time system responds to the occurrence of specific events and not to defined changes in values taking place at specific points of time. The main idea is the integration of statistical driver-distributions to reflect a realistic driving style. These statistics which depend on type of driver, kind of driven route and driving situation have been measured and are now embedded into the simulation environment. The virtual driver model represents the categories aggressive, average and mild. For instance, just as in the case of real measurements the HIL test-bench enables the calculation of the shifting. frequency and the fuel consumption behaviour on various routes (city, mountain, country road and autobahn) with different driving styles in a long-term simulation without intervention on the part of the user.
The developed real-time environment can be used as a modern computer-aided tool not only for the test but also for the development and optimisation of ETC functions (including adaptive functions) into the field of consumption characteristics, driveability, operating strategy, enduring testing (diagnosis), collective population in a virtual manner. This means an important step in the progress of the control unit development process.