Search Results

Drive cycles have been an integral part of emission tests and virtual simulations for decades. A drive cycle is a representation of running behavior of a typical vehicle, involving the drive pattern, road characteristics and traffic characteristics. Drive cycles are typically used to assess vehicle performance parameters, perform system sizing and perform accelerated testing on a test bed or a virtual test environment, hence reducing the expenses on road tests. This study is an attempt to design a relatively robust process to generate a real world drive cycle. It is based on a Six Sigma design approach which utilizes data acquired from real world road trials. It explicitly describes the process of generating a drive cycle which closely represents the real world road drive scenario. The study also focuses on validation of the process by simulation and statistical analysis.

With continuous pressure for reducing vehicle development time and cost, without compromising on system reliability, it is imperative to move from Road to Lab to Model (RLM) based development. Every OEM is currently using virtual environment to complete functional checks of systems during development. The method of developing control functions and calibration in virtual environment brings repeatability and reproducibility which typically is challenging in real world testing. This process is cost effective and optimizes the time for development and brings high level of system maturity before testing it in the vehicle. This paper focuses on defining a front-loading approach for setting up of virtual test environment. Development of virtual test environment and its validation with respect to real vehicle data will be discussed, with focus on vehicle plant model and driver model.

OEMs these days are focusing on front loading the activities to Virtual Test Environment (VTE) based development owing to high development cost and complexity in achieving repeatability during testing phase of vehicle development,. This process not only helps in reducing the cost and time but also helps in increasing the maturity and confidence level of the developed system before actual prototype is built. In the past, extensive research has happened for increasing the fidelity of VTE by improving plant model efficacy which involves powertrain and other vehicle systems. On the other hand, improving the precision of driver model which is one of the most complex nonlinear systems of virtual environment still remains a challenge. It is apparent that various drivers show different behavior in real world for a given drive profile. While modelling a driver for a VTE, the real world driver attributes are seldom considered.

Within the last decade, due to increasing fuel prices, unstable political situation in major oil producing nations and global warming, there is an increased demand for fuel efficient and environment friendly vehicles. In this context, research is being concentrated in the field of advanced, greener powertrain configurations ranging from hybrids to EVs to fuel cells to HCCI engines. The efficacy for any of the above stated powertrain technology, lies in the optimum component specification. Component specification, operational reliability, & life prediction are highly dependent on the traffic condition, driving nature and vary from country to country. For developing countries, like India, where the traffic & drive pattern are dense & slow moving, there is a dire need for generating load cycle based on Real World Usage Profile (RWUP). The paper will propose a systematic approach to create load cycles in order to derive component specifications for the powertrain based on RWUP.