<div class="section abstract"><div class="htmlview paragraph">The computational efficiency of dynamic programming (DP) energy management
strategies is enhanced through the discretization of state variables in this
study. The upper and lower bounds of SOC (State of Charge) and the SOC variation
at each moment are calculated using the maximum and minimum power of the range
extender, which eliminates invalid state combinations and significantly reduces
the size of the feasible state set. To investigate the impact of different
sampling intervals on SOC during various phases, intervals at 1s, 2s, 4s, 5s,
and 10s are set for both charge retention and consumption phases. It is revealed
that in the consumption phase, different sampling intervals minimally affect
SOC, with trajectories closely matching. However, in the charge retention phase,
the impact of different sampling intervals on SOC is significant, resulting in
considerable differences in SOC trajectories. Additionally, in the
charging-discharging (CD) phase, fuel consumption significantly varies with
sampling intervals, decreasing as the interval increases. In contrast, during
the charge storage (CS) phase, minor differences in fuel consumption are
observed due to the larger power of the range extender. The DP computation time
in the CD phase is substantially less than in the CS phase, primarily because
the feasible domain in the CD phase is smaller. As sampling intervals decrease,
computation time increases exponentially, characteristic of the DP algorithm.
Sampling intervals are recommended to be increased in practical applications to
balance computational accuracy and efficiency. This research provides an
efficient computational approach for DP energy management strategies and
uncovers the impact patterns of sampling intervals on SOC stability and fuel
consumption, offering theoretical and practical guidance for the design of
energy management strategies.</div></div>