Zhihui Xu
Zhihui Xu has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
Beiyuan Guo
Beijing Jiaotong University
2 shared publicationsAobo Wang
Beijing Jiaotong University
1 shared publicationDe-Song Su
China Nuclear Power Engineering Co., Ltd
1 shared publication- Furui Zhao
School of Mechanical Engineering, University of Science and Technology Beijing, Beijing 100083, China
1 shared publication - Gang Liu
School of Electric Power Engineering, South China University of Technology, Guangzhou 510640, China
1 shared publication - Hengrui Guo
School of Mechanical Engineering, University of Science and Technology Beijing, Beijing 100083, China
1 shared publication
Latest Publications
- Open Access
- 19 September 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (18) , 9290
Speech State Analysis Based on Self-Supervised Representation Shift Under Complex Speech Interference
Safety-critical control environments, such as nuclear power plant main control rooms, involve multi-operator collaboration, continuous information exchange, and stringent reliability requirements, making operator-state monitoring important for task performance and system safety. Speech is naturally produced, non-intrusive, and readily available for continuous acquisition, but multi-operator settings introduce substantial speech interference, while high-arousal or abnormal-state speech is difficult to collect at scale. This study proposes a speech-state analysis method based on self-supervised representation shift. Low- and high-arousal speech conditions are constructed from a public emotional speech dataset using the valence–arousal framework. Emotion2Vec extracts high-level embeddings, an individual low-arousal baseline is established, and cosine distance quantifies representation shift. The method is compared with a convolutional neural network (CNN) and a CNN with long short-term memory (CNN+LSTM) under clean speech, white-noise interference, and multi-speaker speech interference; supervised-model dependence on training set size is also examined. Low- and high-arousal speech remain partially separable in representation space, while the proposed method shows less performance degradation than the supervised baselines under interference, particularly multi-speaker interference. These findings support individual baseline representation shift as a feasible approach for non-intrusive state monitoring in safety-critical multi-operator environments.
- 19 September 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (18) , 9290
Open AccessSpeech State Analysis Based on Self-Supervised Representation Shift Under Complex Speech Interference
Safety-critical control environments, such as nuclear power plant main control rooms, involve multi-operator collaboration, continuous information exchange, and stringent reliability requirements, making operator-state monitoring important for task performance and system safety. Speech is naturally produced, non-intrusive, and readily available for continuous acquisition, but multi-operator settings introduce substantial speech interference, while high-arousal or abnormal-state speech is difficult to collect at scale. This study proposes a speech-state analysis method based on self-supervised representation shift. Low- and high-arousal speech conditions are constructed from a public emotional speech dataset using the valence–arousal framework. Emotion2Vec extracts high-level embeddings, an individual low-arousal baseline is established, and cosine distance quantifies representation shift. The method is compared with a convolutional neural network (CNN) and a CNN with long short-term memory (CNN+LSTM) under clean speech, white-noise interference, and multi-speaker speech interference; supervised-model dependence on training set size is also examined. Low- and high-arousal speech remain partially separable in representation space, while the proposed method shows less performance degradation than the supervised baselines under interference, particularly multi-speaker interference. These findings support individual baseline representation shift as a feasible approach for non-intrusive state monitoring in safety-critical multi-operator environments.
- 23 October 2025
- Published by Springer Nature
- p. 38-44
Impact of Procedural Task Types on Operational Control Performance of Operators in Nuclear Power Plant Control Rooms
- 23 October 2025
- Published by Springer Nature
- p. 38-44
Impact of Procedural Task Types on Operational Control Performance of Operators in Nuclear Power Plant Control Rooms
- 23 October 2025
- Published by Springer Nature
- p. 45-51
A Cognitive Stage-Oriented Evaluation of Performance Shaping Factors in Intelligent Supervisory Control Task
- 23 October 2025
- Published by Springer Nature
- p. 45-51
A Cognitive Stage-Oriented Evaluation of Performance Shaping Factors in Intelligent Supervisory Control Task
- Open Access
Influence of Combined Electrothermal Aging on Dielectric and Thermal Properties of HVAC XLPE Cable
To explore the influence of electrothermal aging on the properties of HVAC XLPE cable during its normal operation, two circuits of 110 kV XLPE cable were subjected to a combined electrothermal accelerated aging test for 180 days. The test voltage was set at 95 kV, and the test temperature was set at 90 °C. The constant temperature aging mode was adopted for one circuit, and the periodic thermal cycle aging mode was adopted for the other circuit. The dielectric property of the cable was diagnosed with dielectric spectroscopy measurement, and the thermal property was diagnosed with thermal parameters (thermal capacity and thermal resistance) measurement. Combined with Fourier infrared spectroscopy (FTIR) and differential scanning calorimetry (DSC) measurements, morphological changes in the insulation under different electrothermal aging modes were analyzed to construct the correlation with the dielectric and thermal properties. The results show that the short-term electrothermal constant temperature aging mode effectively enhances the dielectric and thermal properties of the cable by modifying spherulite morphology and migrating polar compounds; the short-term electrothermal cycle aging mode also enhances the thermal properties of the cable due to the drop in small polar compounds. However, different distributions of polar compounds and crystal interfaces in the insulation lead to extreme distortion in the electric field distribution.
Open AccessInfluence of Combined Electrothermal Aging on Dielectric and Thermal Properties of HVAC XLPE Cable
To explore the influence of electrothermal aging on the properties of HVAC XLPE cable during its normal operation, two circuits of 110 kV XLPE cable were subjected to a combined electrothermal accelerated aging test for 180 days. The test voltage was set at 95 kV, and the test temperature was set at 90 °C. The constant temperature aging mode was adopted for one circuit, and the periodic thermal cycle aging mode was adopted for the other circuit. The dielectric property of the cable was diagnosed with dielectric spectroscopy measurement, and the thermal property was diagnosed with thermal parameters (thermal capacity and thermal resistance) measurement. Combined with Fourier infrared spectroscopy (FTIR) and differential scanning calorimetry (DSC) measurements, morphological changes in the insulation under different electrothermal aging modes were analyzed to construct the correlation with the dielectric and thermal properties. The results show that the short-term electrothermal constant temperature aging mode effectively enhances the dielectric and thermal properties of the cable by modifying spherulite morphology and migrating polar compounds; the short-term electrothermal cycle aging mode also enhances the thermal properties of the cable due to the drop in small polar compounds. However, different distributions of polar compounds and crystal interfaces in the insulation lead to extreme distortion in the electric field distribution.
- 19 April 2022
- Published by Springer Nature
- p. 650-665
Study for Human Reliability Analysis on Emergency Feedwater Injection Under Severe Accident in Nuclear Power Plant
- 19 April 2022
- Published by Springer Nature
- p. 650-665
Study for Human Reliability Analysis on Emergency Feedwater Injection Under Severe Accident in Nuclear Power Plant