Improved grey system models for predicting traffic parameters
نویسندگان
چکیده
In transportation applications such as real-time route guidance, ramp metering, congestion pricing and special events traffic management, accurate short-term flow prediction is needed. For this purpose, paper proposes several novel online Grey system models (GM): GM(1,1|cos(?t)), GM(1,1|sin(?t),cos(?t)), GM(1,1|e-at,sin(?t),cos(?t)). To evaluate the performance of proposed models, they are compared against a set benchmark models: GM(1,1) model, Verhulst with without Fourier error corrections, linear time series nonlinear model. The evaluation performed using loop detector probe vehicle data from California, Virginia, Oregon. Among corrected model outperformed series, non-linear models. turn, three GM(1,1|e-at,sin(?t),cos(?t)), in by at least 65%, 16% 11%, terms Root Mean Squared Error, 82%, 58% 42%, Absolute Percentage respectively. It observed that more adaptive to location (e.g., perform well for all roadway types) parameters speed, travel time, occupancy, volume), do not require many points training (4 observations found be sufficient).
منابع مشابه
Development and Comparison of Methods Based on Grey and Fractal Models for Predicting Natural Gas Prices
Today there is consensus regarding the importance of predicting the price of energy carriers for economic development, including natural gas as a preferred energy source on environmental grounds. In this paper we analyze the nonlinear behavior of natural gas prices over time by use of the Grey Method (GM), Developed Grey Method (DGM), Fractal Discrete Grey Method (FDGM) and a compound Method II...
متن کاملPredicting Model of Traffic Volume Based on Grey-Markov
Grey-markov forecasting model of traffic volume was founded by applying the model of GM (1,1) and Markov random process theory. The model utilizes the advantages of Grey-markov GM (1,1) forecasting model and Markov random process in order to discover the developing and varying tendency of the forecasting data sequences of traffic volume. The analysis of an example indicates that the grey-markov...
متن کاملAutomatic Monitoring & Detection System (AMDS) for Grey Traffic
International incoming grey traffic is currently one of the most important issues for both the Regulator and legal telecommunication operators in Pakistan. The major reason for higher grey traffic in Pakistan is because of international settlement rates which are higher than the cost of domestic termination rate. The grey traffic operators gain financial advantage by offering low international ...
متن کاملImproved nearest-neighbor parameters for predicting DNA duplex stability.
Thermodynamic data were determined from UV absorbance vs temperature profiles of 23 oligonucleotides. These data were combined with data from the literature for 21 sequences to derive improved parameters for the 10 Watson-Crick nearest neighbors. The observed trend in nearest-neighbor stabilities at 37 degrees C is GC > CG > GG > GA approximately GT approximately CA > CT > AA > AT > TA (where o...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Expert Systems With Applications
سال: 2021
ISSN: ['1873-6793', '0957-4174']
DOI: https://doi.org/10.1016/j.eswa.2021.114972