Grid-Oriented Probabilistic Load Forecasting for AI Data Centers

Qingyi Yao; Ding Lin; Tianqiao Zhao; Meng Yue; Jianhui Wang

Home/Archive/arXiv

arXiv · 2611.08929

Grid-Oriented Probabilistic Load Forecasting for AI Data Centers

Qingyi Yao·Ding Lin·Tianqiao Zhao·Meng Yue·Jianhui Wang

View original source ↗Read source document ↗Export metadata ↓

Abstract

Probabilistic load forecasts for AI data centers need to represent both uncertainty under changing workloads and temporal dependence to support grid operations. This paper develops an open, grid-oriented benchmark using measured GPU power from a 448-GPU AI cluster. Forecasts are evaluated through four stages: probabilistic prediction, online uncertainty calibration, temporal scenario generation, and horizon-wide reserve sizing. Following a change in cluster utilization, a quantile LightGBM forecast achieves only 63.4% coverage for a nominal 90% interval, while online regime-conditioned calibration increases coverage to 87.3%. Even with well-calibrated marginal forecasts, independent sampling produces unrealistic ramp distributions and excessive reserve procurement. Reconstructing temporal dependence from historical forecast errors reduces the discrepancy in ramp statistics and lowers realized reserve cost by 7.4-10.6% across three forecasting tasks. These findings show that marginal forecast quality alone is insufficient for operational decisions and motivate joint evaluation of calibration, temporal structure, and decision outcomes.

Explore related subjects

eess.SY cs.LG cs.SY

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Qingyi Yao, Ding Lin, Tianqiao Zhao, Meng Yue, Jianhui Wang. 2026-11-06. Grid-Oriented Probabilistic Load Forecasting for AI Data Centers. https://arxiv.org/abs/2611.08929

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Safety Blind Spot in Remote Driving: Considerations for Risk Assessment of Connection Loss Fallback Strategies

As part of the overall goal of driverless road vehicles, remote driving is a major emerging field of research of its own. Current remote driving concepts for public road traffic often establish a fallback strategy of immediate braking to a standstill in the event of a connection loss. This may seem like the most logical option when human control of the vehicle is lost. However, our simulation results from hundreds of scenarios based on naturalistic traffic scenes indicate high collision rates for any immediate substantial deceleration to a standstill in urban settings. We show that such a fallback strategy can result in a SOTIF-relevant hazard, raising the question whether this design decision is acceptable. Therefore, from a safety perspective, we call this problem a safety blind spot, as safety analyses in this regard seem to be very rare. In this article, we first present a simulation on a naturalistic dataset that shows a high probability of collision in the described case. Second, we discuss the severity of the resulting potential rear-end collisions and provide an even more severe example by including a large commercial vehicle in the potential collision.

eess.SY↗

arXiv2026

An Enhanced Submodule for Modular Multilevel Converters With DC Fault-Clearing Capability

Modular multilevel converter (MMC) has been successfully applied in various power electronic systems owing to its high efficiency, scalability, and superior output performance. Although the half-bridge submodule (HBSM) is widely used in MMCs for its structural simplicity, it is incapable of handling direct-current (DC) short-circuit faults. The diode-clamp submodule (DCSM) addresses this limitation by providing DC fault ride-through capability. However, because its two identical capacitors are connected in series, the equivalent capacitance is halved. To overcome this drawback, an enhanced SM is proposed in this paper. For the same energy storage capacity, the proposed SM reduces the total required capacitance by 75% compared with the DCSM. In addition, the proposed SM requires one fewer diode than the DCSM, thereby lowering the overall MMC capital cost. The topology and the operating modes of the proposed SM are described in detail, and its functionality is experimentally validated. The results demonstrate that the proposed SM can suppress DC fault currents and restore normal operation without additional external protection devices.

eess.SY↗

arXiv2026

Voltage Sensitivity-Based Fair Allocation of Operating Envelopes in Distribution Networks

Operating envelopes (OEs) define the admissible active and reactive power injections of distributed energy resources while respecting distribution network limits. In unbalanced low-voltage feeders, OEs based only on voltage magnitude and thermal limits may overestimate available flexibility because they neglect power quality requirements. This paper develops a three-phase unbalanced AC optimal power flow framework to compute coupled active and reactive power OEs while enforcing limits on the voltage unbalance factor (VUF). It also proposes a fairness mechanism based on voltage sensitivities to allocate available flexibility among multiple photovoltaic (PV) units. The proposed mechanism is compared with lexicographic max--min fairness as a benchmark for equitable resource allocation. Case studies on unbalanced test feeders evaluate how VUF constraints change the feasible operating region of PV units. The results show how VUF constraints affect feasible flexibility and demonstrate that the proposed voltage sensitivity-based mechanism is nearly six times faster than lexicographic max-min fairness.

eess.SY↗