Table 2 Optimal configuration of the NAR and NARX models for forecasting charging time and energy for one-day ahead on the validation dataset over the 10 different simulation trials.

From: Electric vehicles charging station allocation based on load profile forecasting and Dijkstra’s algorithm for optimal path planning

Forecasting model

Forecasting variable

\({{N}_{h}}_{opt}\)

\({d}_{opt}\)

\(RMSE\)

\(R\)[%]

Computation time [minutes]

NAR

Charging Time (\(T\))

1

1

28.06 Hours

63.17

1.90

Charging Energy (\(E\))

1

1

54.30 kWh

65.46

2.24

NARX

Charging Time (\(T\))

10

1

17.08 Hours

88.38

3.92

Charging Energy (\(E\))

8

1

30.27 kWh

90.26

3.64