An operating system for the optimization of technical systems using the example of transmission calibration

Abstract

Technical systems must be continuously improved so that they can remain competitive in the market. The duration and the costs of the development process are important factors of success for a company. Therefore practical methods and processes have to be provided to create an efficient optimization process. Within this paper an operation system for a time efficient, test based optimization of technical systems will be presented. Therefore methods from simulations and tests are being combined. Based on the X-in-the-Loop approach (XiL) it is possible to examine the Unit under Test (UuT) on an overall vehicle level in interaction with the environment and the driver. This way, even complex interactions between the systems can be emulated concerning examination- and optimization- goals. This also allows validating and optimizing modern, strongly interconnected drive systems. Especially hybrid powertrains present major challenges to the test because the additional state variables like SOC as well as engine and operating temperatures have to be considered and adjusted if necessary. This operation system provides the methods to reduce the time of conventional test runs as well as methods for model based optimization. Conventional approaches use non-physical models like polynomial-approaches or artificial neural networks for the optimization. In contrast, the presented approach allows the use of models partly consisting of physical parts. Thereby, the model quality and the expenditure of time for testing can be reduced significantly. Even conventional test runs with various measuring points, like monitoring of shifting quality, can be shortened significantly by this operation system. In the example of the gear shifting evaluation of a dual clutch transmission the expenditure of time can be reduced by 80%. Therefore, several measuring points are being defined throughout the different shifting modes (different accelerator pedal positions, rpms, gears, driving situations and driving maneuvers) and put into an ideal order. This means, the system calculates the order for a minimum time of conditioning between the measuring points (reaching the desired initial condition). © Springer-Verlag 2013.