Partial discharge originates from microscopic insulation imperfections in high-voltage apparatus and is widely considered a critical marker of incipient deterioration. Conventional partial discharge detection methods are typically constrained by limited bandwidth and often rely on predefined feature extraction, which impedes reliable recognition of broadband transient signals. In this work, we employ a Rydberg atomic sensor to directly capture time-domain responses of partial discharge emissions and construct distinctive spectral fingerprints for different types. A 1D ResNet deep learning model is then applied to recognize these fingerprints from time-domain signals without manual feature engineering. Under increased source-antenna distances, where spectral features are significantly attenuated, the model attains a recognition accuracy of approximately 94\% across four partial discharge categories, demonstrating robustness to attenuation and noise. We further validate the approach in a simulated early-warning scenario, where partial discharge signals mixed with noise are analyzed and the model successfully generates predictive alarms. These results underscore the potential of integrating Rydberg-based broadband sensing with data-driven analysis for non-invasive, high-sensitivity diagnostics of electrical insulation systems.
Partial discharge originates from microscopic insulation imperfections in high-voltage apparatus and is widely considered a critical marker of incipient deterioration. Conventional partial discharge detection methods are typically constrained by limited bandwidth and often rely on predefined feature extraction, which impedes reliable recognition of broadband transient signals. In this work, we employ a Rydberg atomic sensor to directly capture time-domain responses of partial discharge emissions and construct distinctive spectral fingerprints for different types. A 1D ResNet deep learning model is then applied to recognize these fingerprints from time-domain signals without manual feature engineering. Under increased source-antenna distances, where spectral features are significantly attenuated, the model attains a recognition accuracy of approximately 94% across four partial discharge categories, demonstrating robustness to attenuation and noise. We further validate the approach in a simulated early-warning scenario, where partial discharge signals mixed with noise are analyzed and the model successfully generates predictive alarms. These results underscore the potential of integrating Rydberg-based broadband sensing with data-driven analysis for non-invasive, high-sensitivity diagnostics of electrical insulation systems.
Abstract Partial discharge (PD) detection has garnered considerable attention due to its critical role in condition assessment and fault prediction of power system equipment. However, efficient and accurate acquisition of PD signals remains a significant technical challenge. Among mainstream technologies, the ultra-high frequency (UHF) detection method plays a key role in capturing PD signals. This work introduces a Rydberg atom-based electric field sensing approach for PD detection, employing a three-photon excitation scheme in cesium vapor. Four representative PD types are investigated: void, floating, particle, and corona, revealing the ability of the nonmetallic atomic sensor to capture their distinct time-domain characteristics and periodic behaviors at the power frequency. In addition, an antenna–waveguide–vapor cell coupling configuration is proposed. This electro-optical to opto-electronic conversion mechanism allows the sensor to achieve long-range detection while preserving favorable signal-to-noise ratios. Although the sensor’s limited instantaneous bandwidth leads to pulse broadening in high-speed signals, its theoretical high sensitivity and wide operating bandwidth highlight its potential as a complement to traditional metal-antenna solutions. These findings provide a robust foundation for implementing quantum-enhanced PD monitoring in practical power engineering scenarios.
Rydberg-atom sensors have emerged as a promising quantum technology for broadband radio-frequency detection owing to their intrinsic calibration, wide frequency coverage, and high sensitivity. However, their potential for intelligent identification of real-world wireless signals remains largely unexplored. Here, we demonstrate a quantum-enabled framework for mobile signal classification based on Rydberg electromagnetically induced transparency spectroscopy and deep learning. Three types of wireless emissions, including cellular voice call, Wi-Fi, and 5G signals from three commercial smartphone brands, are measured to construct a spectral-fingerprint dataset. A convolutional neural network trained on this dataset achieves an initial baseline classification accuracy of 93.44\%. To compensate for spectral drifts induced by temporal and environmental variations, a progressive fine-tuning strategy is employed to continuously adapt the pretrained model, enabling robust classification across diverse deployment scenarios. Furthermore, continuous real-time monitoring is demonstrated under dynamically varying wireless conditions, including cellular voice call, Wi-Fi, and their coexistence. The proposed system achieves an overall frame-level recognition accuracy of 87.67\% over a 60-s measurement. These results establish a pathway toward intelligent quantum radio-frequency sensing and highlight the potential of Rydberg-atom platforms for real-time wireless signal awareness and spectrum monitoring.
Accurate detection of internal partial discharge (PD) in gas-insulated switchgear (GIS), a core component of modern power systems, is crucial for assessing insulation condition, ensuring equipment reliability, and maintaining power grid stability. However, the practical application of traditional ultra-high frequency (UHF) electromagnetic sensors in GIS condition-based maintenance faces significant limitations, including severe signal attenuation (20～40 dB/m) caused by the metal enclosure, the risk of SF6 leakage due to compromised airtightness with window-based installation, and a limited dynamic range (60～80 dB) insufficient for covering the full range of discharge magnitudes. To overcome these technical bottlenecks, this paper innovatively proposes a quantum-based UHF electric field sensing method utilizing the Rydberg atomic electromagnetically induced transparency (EIT) effect. The core advantages of this method lie in: leveraging the extremely high electric polarizability of highly excited cesium atoms (n=72D5/2) to achieve intrinsically safe measurement through an optical readout mechanism; establishing a stable EIT window using a dual-frequency laser system (852 nm probe light and 509 nm pump light) combined with precise frequency-locking technology; and designing a non-metallic atomic vapor cell antenna to realize the first near-field sensing of transient pulse electric fields in the 0.5～3 GHz band excited by particle discharge within GIS. This research demonstrates the capability of Rydberg atomic sensing technology to effectively detect discharges from a typical GIS defect (metal particles) for the first time. By employing quantum precision measurement, it provides a novel optical fiber sensing solution characterized by high sensitivity, broadband capability, and intrinsic safety for the online insulation monitoring of power equipment. This approach holds significant promise for advancing condition-based maintenance towards intelligent and precise evolution, offering key supporting technologies for the development of smart grids.
In recent years, the Rydberg-atom sensor has attracted significant attention due to its high sensitivity, wide dynamic range, and miniaturization capabilities. However, noise from laser power fluctuations, frequency drift, and environment is introduced alongside the intrinsic quantum noise of Rydberg atoms, significantly degrading system performance for microwave sensing. This study tries to reduce the noise impact by combining the Rydberg-atom sensor with the convolutional neural network (CNN)-transformer model. The CNN module captures local temporal features of the signals, while the transformer module extracts global correlations within these features. We apply it to the following classification tasks of atomic microwave sensing: multifrequency microwave recognition, broadband spectrum monitoring, and automatic modulation recognition. We demonstrate that the CNN-transformer model achieves high accuracy across the three tasks and can effectively reduce the noise impact, which holds significant potential for applications in radar detection and wireless communication.
Rydberg-atom electrometry based on electromagnetically induced transparency (EIT) offers an optical approach for electric-field readout, but reliable readout of weak transient responses remains challenging under low signal-to-noise ratio (SNR) conditions. Here, we demonstrate machine-learning-enhanced readout of partial-discharge-like (PD-like) transient responses in a Cs 73 D 5/2 Rydberg-EIT vapor-cell system. Matched filtering (MF) and normalized cross-correlation (NCC) are used for candidate generation, followed by principal component analysis (PCA)/logistic-regression-based waveform-morphology discrimination and a cross-cycle temporal-consistency refinement. Across experimentally acquired low-SNR cycles, the PCA/logistic morphology selector achieved a strict localization accuracy of 63.25%, which increased to 86.32% after incorporating cross-cycle temporal consistency. The results indicate that weak Rydberg-EIT transient responses can retain discriminative local waveform morphology even when conventional template-peak ranking becomes unreliable. Overall, these results demonstrate that morphology-based machine-learning discrimination, complemented by temporal-consistency refinement, can improve the reliability of weak Rydberg-EIT transient readout under low-SNR conditions.
Recognition of multifrequency microwave (MW) electric fields is challenging because of the complex interference of multifrequency fields in practical applications. Rydberg atom-based measurements for multifrequency MW electric fields is promising in MW radar and MW communications. However, Rydberg atoms are sensitive not only to the MW signal but also to noise from atomic collisions and the environment, meaning that solution of the governing Lindblad master equation of light-atom interactions is complicated by the inclusion of noise and high-order terms. Here, we solve these problems by combining Rydberg atoms with deep learning model, demonstrating that this model uses the sensitivity of the Rydberg atoms while also reducing the impact of noise without solving the master equation. As a proof-of-principle demonstration, the deep learning enhanced Rydberg receiver allows direct decoding of the frequency-division multiplexed signal. This type of sensing technology is expected to benefit Rydberg-based MW fields sensing and communication.