- GAN Shengfeng
- DOI: 10.5281/zenodo.23586595
- GAS Journal of Engineering and Technology (GASJET)
To
address the severe impulse loads during tripping operations, the insufficient
ramp capability of diesel generators, and the poor adaptability of conventional
lithium-ion batteries to frequent deep charge–discharge cycles in islanded
drilling-well microgrids, a multi-time-scale optimal dispatch method using a
lithium titanate (LTO) battery as the core energy storage unit is proposed. The
three-component decomposition characteristics of drilling load (base, impulse,
and auxiliary components) are first analyzed, and the suitability of LTO
batteries for drilling impulse conditions is demonstrated in terms of high-rate
charge/discharge, long cycle life, wide temperature range, and safety. A
three-layer collaborative dispatch framework is then established, comprising
day-ahead economic dispatch, intra-day model predictive control (MPC), and
real-time PQ power control. The day-ahead layer optimizes diesel output and
storage schedules to minimize daily operating cost; the intra-day layer updates
PV forecast errors every 15 min; the real-time layer performs impulse power
smoothing within 100 ms and recovers downward-drilling braking energy.
Simulation based on a 50DB drilling rig well site in an oilfield shows that,
with a 1.725 MW/0.424 MWh LTO storage system, the diesel power fluctuation rate
is reduced by 79%, the diesel ramp rate is controlled within 50 kW/min, the
recovered braking energy reaches approximately 300 kWh per well (2.5% of daily
electricity consumption), and the daily operating cost of the three-layer
dispatch is reduced by about 19% compared with the two-layer scheme. The
results verify the engineering applicability of LTO storage for impulse load
smoothing and energy recovery in drilling microgrids, and can provide a
reference for the design of green power supply systems at well sites.
Keywords: drilling microgrid; lithium titanate battery; impulse load smoothing; multi-time-scale dispatch; braking energy recovery; model predictive control.
