Liu, Z., Ivanco, A., and Filipi, Z., "Quantification of Drive Cycle's Rapid Speed Fluctuations Using Fourier Analysis," SAE Int. J. Alt. Power. 4(1):170-177, 2015, doi:10.4271/2015-01-1213.
This paper presents a new way to evaluate vehicle speed profile aggressiveness, quantify it from the perspective of the rapid speed fluctuations, and assess its impact on vehicle fuel economy. The speed fluctuation can be divided into two portions: the large-scale low frequency speed trace which follows the ongoing traffic and road characteristics, and the small-scale rapid speed fluctuations normally related to the driver's experience, style and ability to anticipate future events. The latter represent to some extent the driver aggressiveness and it is well known to affect the vehicle energy consumption and component duty cycles. Therefore, the rapid speed fluctuations are the focus of this paper. Driving data collected with the GPS devices are widely adopted for study of real-world fuel economy, or the impact on electrified vehicle range and component duty cycles. However, the accompanying signal noise poses a challenge, and needs to be separated from realistic rapid speed fluctuations. Filtering is commonly used, but aggressive smoothing technique can lead to loss of useful driving information. In contrast, mild smoothing technique can lead to under-filtering and inclusion of redundant information. The main contribution of this paper is a proposed metric, denoted as “Ripple Aggressiveness”, to quantify the rapid speed fluctuations over a drive cycle based on the Fourier analysis. This metric allows assessment of the filtering level, and detection of over-filtering, or under-filtering. The data used to develop and demonstrated the new technique are from the 2001∼2002 Southern California Household Travel Survey, 2010∼2012 California Household Travel Survey, 2001∼2005 Michigan Road Departure Crash Warning System Field Operational Tests, and EPA dynamometer drive schedules. The newly developed metric is correlated with fuel economy using the vehicle simulation, and the results show a positive correlation between Ripple Aggressiveness and vehicle fuel consumption. Hence the metric can also be used as a surrogate to quantify the driver aggressiveness.