South Korean deep tech startup xDots has launched the xSee quantum sensing system, designed to capture high-precision power consumption data in data centers and industrial facilities. This system serves as the hardware measurement foundation for the xEnergy optimization platform, utilizing diamond nitrogen-vacancy quantum sensors to measure current, magnetic fields, and energy flow, with a nominal measurement accuracy of ±0.01%.
The sensing core of xSee is the diamond NV center. The NV center consists of a nitrogen atom and an adjacent vacancy in the diamond lattice, and its electron spin state is affected by external magnetic fields, current changes, temperature, and stress perturbations. The device monitors changes in the spin state through optical excitation and readout, then inversely deduces these changes into local magnetic fields and current fluctuations. Traditional power monitoring equipment typically focuses on energy consumption aggregation at the level of distribution cabinets, circuits, meters, or server rooms, with a coarse sampling granularity that tends to average out minor fluctuations in server power supplies, cooling equipment, motors, variable frequency drives, UPS, PDU, and local circuits. xSee aims to push measurement points closer to equipment and process units, using quantum sensors to capture subtle current changes and magnetic field disturbances, and then feeding these high-resolution power consumption data into xEnergy for modeling.
The system operates at room temperature and does not require cryogenic cooling conditions. For data centers and industrial facilities, this determines whether it can transition from a laboratory instrument to deployable on-site measurement hardware.
After the xEnergy platform receives the power consumption data collected by xSee, it enters the AI agent analysis engine and monitoring dashboard. The data processing pipeline includes current and magnetic field sampling, time-series synchronization, noise filtering, device-level energy consumption curve reconstruction, anomaly fluctuation identification, and generation of energy-saving control recommendations. Server clusters, cooling units, fans, pumps, power conversion equipment, and rack-level power distribution units in data centers may generate energy consumption fluctuations of varying frequencies and amplitudes; motors, compressors, process equipment, and conveyor systems in industrial facilities also experience start-stop cycles, load changes, and inefficient operating ranges. xSee provides the underlying measurement precision, while xEnergy is responsible for converting these minor fluctuations into identifiable device states, load patterns, and energy consumption anomalies.
When diamond NV sensing is used for power consumption measurement, the focus is not on "quantum computing" but on the sensitivity of quantum states to physical perturbations. External currents generate magnetic fields, which alter the spin energy levels of the NV center; by reading changes in the spin resonance signal, the system can obtain more granular current information. Compared to merely observing the total power curve, this method is better suited for capturing short-term spikes, low-amplitude oscillations, local losses, current characteristic changes caused by equipment aging, and the dynamic coupling relationship between cooling systems and IT loads.
xSee, as the hardware layer, is responsible for continuous measurement and high-precision sampling; xEnergy, as the platform layer, handles data analysis, visualization, and optimization decisions; the xMon dashboard is used to present device-level, system-level, or process-level power consumption status. When combined, the technical chain begins with quantum sensor readings, proceeds through edge collection, AI agent analysis, energy consumption model construction, and culminates in control strategy adjustments for data centers or industrial facilities.
Subsequent technical validation will focus on several parameters: drift control of ±0.01% accuracy during long-term operation, anti-interference capability of NV sensors in complex electromagnetic environments, consistency of multi-point synchronous sampling, installation methods in high-density power distribution scenarios in data centers, and whether xEnergy can stably map microscopic power consumption signals to specific energy-saving actions for server loads, cooling equipment, power supply and distribution links, and industrial process equipment.