Remote Sensing and Nonlinear Auto-regressive Neural Network (NARNET) Based Surface Water Chemical Quality Study: A Spatio-Temporal Hybrid Novel Technique (STHNT)

Ramaraj, M.; Sivakumar, R.

doi:10.1007/s00128-022-03646-9

Cited by 4 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Particularly, the NAR network, a type of dynamic neural network, is suitable for the prediction of time series. Previous studies have shown its advantages in terms of small computation times and high prediction accuracy [36]. Therefore, the NAR neural network was used in this study to establish a slope prediction model based on the slope time series.…”

Section: Road Slope Prediction Based On the Nar Neural Networkmentioning

confidence: 99%

Real-Time Predictive Energy-Saving Control for Electric Vehicle Based on Road Slope Prediction

Wu,

Lin,

et al. 2023

International Journal of Energy Research

View full text Add to dashboard Cite

Predictive energy-saving control (PEC) is aimed at reducing energy consumption by designing the vehicle speed while considering future road and traffic information. In particular, the slope of the road ahead is necessary and critical for PEC. This paper proposes a road slope prediction method for production vehicles that uses the nonlinear autoregressive (NAR) neural network model based on road slope sensors. To adaptively balance the energy savings and trip time, this paper proposed a real-time variable weight PEC method for a four-wheel-drive (4WD) intelligent electric vehicle. The weight coefficients are automatically changed according to the characteristics of the road slope, where the vehicle energy-saving rate on the steep downhill road can be maximized. The results of real-time simulation on the dSPACE platform indicated that the road slope predictive model can be run in real time and adapted to changes in road slope and speed. The root mean square error (RMSE) of the predictive results is 0.3063. On a steep downhill road, the energy-saving rate of the proposed PEC method can reach 30.87% at a small expense of time of 3.75%. On uphill and flat roads, energy can be saved by 6.35% at a time cost of 3.0%. Compared with the PEC with constant weight factors, the two control objectives of energy savings and traveling time can be better balanced on various types of roads.

show abstract

Section: Road Slope Prediction Based On the Nar Neural Networkmentioning

confidence: 99%