Mingxiang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mingxiang Wang has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Pengfei Jia
Guangxi University
7 shared publications - Jianjun Cao
School of Electrical Engineering
5 shared publications Jiefeng Liu
5 shared publicationsYiyi Zhang
5 shared publicationsDachang Chen
4 shared publicationsMin Xu
College of Food Science and Bioengineering
4 shared publications
Latest Publications
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dielectrics and Electrical Insulation
- Vol. PP (10709878) , 1
A ML-Driven Study on Feature Descriptors and Mechanisms for Adsorption of Air Discharge Decomposition Components
Insulation materials in air switchgear may degrade due to partial discharge (PD). Characteristic gases generated by PD (such as CO, NO, and NO₂) can be detected by gas sensors. However, existing gas-sensitive material screening relies on trial-and-error methods or costly density functional theory (DFT) calculations. Machine learning (ML)-driven DFT calculations have emerged as a new paradigm in gas-sensitive material research. This approach significantly accelerates the rational design of high-performance gas-sensitive materials by efficiently predicting adsorption behavior and revealing underlying microscopic mechanisms. This study screened 14 key features from 200 sets of adsorption data across 50 literature sources (2018–2024), then compared and optimized six ML models (XGBoost, GBR, etc.) to evaluate the predictive performance of D and Ead. Subsequently, SHAP and PDPs analyses were conducted to quantify feature contributions and identify core factors. Research findings indicate that XGBoost achieved prediction R² values of 91.55% for adsorption distance and 89.79% for adsorption energy, significantly outperforming other models (GBR ranked second with R² values of 89.79% and 83.56%, respectively). In terms of generalization ability, XGBoost outperforms GBR in extreme value usage and feature balancing utilization, which is overly dependent on substrate material features. Charge transfer (Qₜ) and dopant atom density (ρd) are the core factors affecting the adsorption behavior, and the SHAP analysis shows that Qₜ has the highest contribution to the model output. Substrate and dopant characteristics contribute equally in the XGBoost model (∼45% each), while the GBR model is overly dependent on substrate characteristics (>60%).
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dielectrics and Electrical Insulation
- Vol. PP (10709878) , 1
A ML-Driven Study on Feature Descriptors and Mechanisms for Adsorption of Air Discharge Decomposition Components
Insulation materials in air switchgear may degrade due to partial discharge (PD). Characteristic gases generated by PD (such as CO, NO, and NO₂) can be detected by gas sensors. However, existing gas-sensitive material screening relies on trial-and-error methods or costly density functional theory (DFT) calculations. Machine learning (ML)-driven DFT calculations have emerged as a new paradigm in gas-sensitive material research. This approach significantly accelerates the rational design of high-performance gas-sensitive materials by efficiently predicting adsorption behavior and revealing underlying microscopic mechanisms. This study screened 14 key features from 200 sets of adsorption data across 50 literature sources (2018–2024), then compared and optimized six ML models (XGBoost, GBR, etc.) to evaluate the predictive performance of D and Ead. Subsequently, SHAP and PDPs analyses were conducted to quantify feature contributions and identify core factors. Research findings indicate that XGBoost achieved prediction R² values of 91.55% for adsorption distance and 89.79% for adsorption energy, significantly outperforming other models (GBR ranked second with R² values of 89.79% and 83.56%, respectively). In terms of generalization ability, XGBoost outperforms GBR in extreme value usage and feature balancing utilization, which is overly dependent on substrate material features. Charge transfer (Qₜ) and dopant atom density (ρd) are the core factors affecting the adsorption behavior, and the SHAP analysis shows that Qₜ has the highest contribution to the model output. Substrate and dopant characteristics contribute equally in the XGBoost model (∼45% each), while the GBR model is overly dependent on substrate characteristics (>60%).
- 17 June 2025
- Published by American Chemical Society (ACS) in Langmuir
- Vol. 41 (25) , 16610-16624
Gas-Sensing Mechanisms and Detection Optimization of ZnO/TiO2–MoTe2 Composites for CF3SO2F Decomposition Products
This study focuses on the detection requirements for breakdown products (HF, SO2, SO2F2, and CF4) produced by the partial discharge of the innovative, environmentally friendly insulating gas CF3SO2F in high-voltage electrical apparatus. A doping modification technique utilizing a metal oxide (ZnO/TiO2) based on two-dimensional MoTe2 is proposed. The method for enhancing gas sensing is being systematically investigated through multiscale theoretical computations. Molecular dynamics simulations are initially utilized to ascertain the structural stability of the doped systems, confirming that ZnO-MoTe2 and TiO2-MoTe2 maintain robust structural integrity in a thermodynamic equilibrium condition. Subsequently, density functional theory is utilized to compare and analyze the adsorption behaviors of the intrinsic MoTe2 and its doped systems toward the four characteristic decomposition products. The findings indicate that doping ZnO markedly improves MoTe2's adsorption capacity for SO2. Adsorption configuration analysis reveals that doping strengthens the interactions between the material surface and SO2 molecules. Electronic structure estimates suggest significant charge transfer and band structure modifications throughout the adsorption process. The density of electronic states in the ZnO-MoTe2 combination exhibits significant variation, indicating that the chemical adsorption of SO2 is predominant. Additionally, the TiO2-doped system shows a selective adsorption tendency for acidic gases such as HF. The comparison of total electron density and differential charge density distributions demonstrates that the charge redistribution at the interface, generated by doping, is the crucial factor enhancing gas adsorption performance. This work reveals, at the atomic scale, the mechanism by which metal oxide doping modulates the gas-sensing properties of MoTe2, providing a theoretical foundation for developing highly selective gas sensors for detecting CF3SO2F decomposition products.
- 17 June 2025
- Published by American Chemical Society (ACS) in Langmuir
- Vol. 41 (25) , 16610-16624
Gas-Sensing Mechanisms and Detection Optimization of ZnO/TiO2–MoTe2 Composites for CF3SO2F Decomposition Products
This study focuses on the detection requirements for breakdown products (HF, SO2, SO2F2, and CF4) produced by the partial discharge of the innovative, environmentally friendly insulating gas CF3SO2F in high-voltage electrical apparatus. A doping modification technique utilizing a metal oxide (ZnO/TiO2) based on two-dimensional MoTe2 is proposed. The method for enhancing gas sensing is being systematically investigated through multiscale theoretical computations. Molecular dynamics simulations are initially utilized to ascertain the structural stability of the doped systems, confirming that ZnO-MoTe2 and TiO2-MoTe2 maintain robust structural integrity in a thermodynamic equilibrium condition. Subsequently, density functional theory is utilized to compare and analyze the adsorption behaviors of the intrinsic MoTe2 and its doped systems toward the four characteristic decomposition products. The findings indicate that doping ZnO markedly improves MoTe2's adsorption capacity for SO2. Adsorption configuration analysis reveals that doping strengthens the interactions between the material surface and SO2 molecules. Electronic structure estimates suggest significant charge transfer and band structure modifications throughout the adsorption process. The density of electronic states in the ZnO-MoTe2 combination exhibits significant variation, indicating that the chemical adsorption of SO2 is predominant. Additionally, the TiO2-doped system shows a selective adsorption tendency for acidic gases such as HF. The comparison of total electron density and differential charge density distributions demonstrates that the charge redistribution at the interface, generated by doping, is the crucial factor enhancing gas adsorption performance. This work reveals, at the atomic scale, the mechanism by which metal oxide doping modulates the gas-sensing properties of MoTe2, providing a theoretical foundation for developing highly selective gas sensors for detecting CF3SO2F decomposition products.
- 30 April 2024
- Published by Elsevier in Chemosphere
- Vol. 358, 142198
Quantum-level investigation of air decomposed pollutants gas sensor (Pd-modified g-C3N4) influenced by micro-water content.
- 30 April 2024
- Published by Elsevier in Chemosphere
- Vol. 358, 142198
Quantum-level investigation of air decomposed pollutants gas sensor (Pd-modified g-C3N4) influenced by micro-water content.
- 10 January 2024
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 16 (3) , 3674-3684
Highly Sensitive Gas Sensor for Detection of Air Decomposition Pollutant (CO, NOx): Popular Metal Oxide (ZnO, TiO2)-Doped MoS2 Surface
When partial discharges occur in air-insulated equipment, the air decomposes to produce a variety of contamination products, resulting in a reduction in the insulation performance of the insulated equipment. By monitoring the concentration of typical decomposition products (CO, NO, and NO2) within the insulated equipment, potential insulation faults can be diagnosed. MoS2 has shown promising applications as a gas-sensitive semiconductor material, and doping metal oxides can improve the gas-sensitive properties of the material. Therefore, in this work, MoS2 has been doped using the popular metal oxides (ZnO, TiO2) of the day, and its gas-sensitive properties to the typical decomposition products of air have been analyzed and compared using density functional theory (DFT) calculations. The stability of the doped system was investigated using molecular dynamics methods. The related adsorption mechanism was analyzed by adsorption configuration, energy band structure, density of states (DOS) analysis, total electron density (TED) analysis, and differential charge density (DCD) analysis. Finally, the practical application of related sensing performance is evaluated. The results show that the doping of metal oxide nanoparticles greatly improves the conductivity, gas sensitivity, and adsorption selectivity of MoS2 monolayer to air decomposition products. The sensing response of ZnO-MoS2 for CO at room temperature (25 °C) reaches 161.86 with a good recovery time (0.046 s). TiO2-MoS2 sensing response to NO2 reaches 3.5 × 106 at 25 °C with a good recovery time (0.108 s). This study theoretically solves the industrial challenge of recycling sensing materials and provides theoretical value for the application of resistive chemical sensors in air-insulated equipment.
- 10 January 2024
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 16 (3) , 3674-3684
Highly Sensitive Gas Sensor for Detection of Air Decomposition Pollutant (CO, NOx): Popular Metal Oxide (ZnO, TiO2)-Doped MoS2 Surface
When partial discharges occur in air-insulated equipment, the air decomposes to produce a variety of contamination products, resulting in a reduction in the insulation performance of the insulated equipment. By monitoring the concentration of typical decomposition products (CO, NO, and NO2) within the insulated equipment, potential insulation faults can be diagnosed. MoS2 has shown promising applications as a gas-sensitive semiconductor material, and doping metal oxides can improve the gas-sensitive properties of the material. Therefore, in this work, MoS2 has been doped using the popular metal oxides (ZnO, TiO2) of the day, and its gas-sensitive properties to the typical decomposition products of air have been analyzed and compared using density functional theory (DFT) calculations. The stability of the doped system was investigated using molecular dynamics methods. The related adsorption mechanism was analyzed by adsorption configuration, energy band structure, density of states (DOS) analysis, total electron density (TED) analysis, and differential charge density (DCD) analysis. Finally, the practical application of related sensing performance is evaluated. The results show that the doping of metal oxide nanoparticles greatly improves the conductivity, gas sensitivity, and adsorption selectivity of MoS2 monolayer to air decomposition products. The sensing response of ZnO-MoS2 for CO at room temperature (25 °C) reaches 161.86 with a good recovery time (0.046 s). TiO2-MoS2 sensing response to NO2 reaches 3.5 × 106 at 25 °C with a good recovery time (0.108 s). This study theoretically solves the industrial challenge of recycling sensing materials and provides theoretical value for the application of resistive chemical sensors in air-insulated equipment.
- 11 November 2023
- Published by Elsevier in Surfaces and Interfaces
- Vol. 44, 103619
Crn (n=1–4) clustered (8, 0) single-walled CNT: Comparison of gas-sensitive properties to air discharge pollutants (CO, NOx)
- 11 November 2023
- Published by Elsevier in Surfaces and Interfaces
- Vol. 44, 103619
Crn (n=1–4) clustered (8, 0) single-walled CNT: Comparison of gas-sensitive properties to air discharge pollutants (CO, NOx)
- 18 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (19) , 22188-22195
Gas Sensitive Analysis of Composite MoS2(001) Surface to Air Discharge Products: A DFT Study
Air discharge characteristic gas detection is considered as one of the most reliable methods in fault diagnosis of air switchgear. Based on first-principles of density functional theory, the adsorption and sensing properties of transition metal (Cr) modified MoS 2 (001) monolayer on three air discharge characteristic gases (Co, NO, and NO 2 ) are analyzed. Firstly, the Cr doped MoS 2 (001) monolayer model is obtained by modeling and analyzing different doping sites, and the subsequent DFT calculation is done. Based on this model, many kinds of adsorption structures were constructed and optimized. The parameters of adsorption structure, adsorption energy and adsorption distance were obtained. Furthermore, the differential charge density, state density, response performance and adsorption performance of the system were analyzed. The results show that Cr-MoS 2 (001) monolayer has good adsorption performance to three gases. CO gas molecules can be rapidly desorbed from Cr-MoS 2 (001) monolayer surface at room temperature. Cr-MoS 2 (001) monolayer has the potential as a low-power gas sensor material for detecting CO characteristic gases and a solid adsorbent material for cleaning NO and NO 2 gases. This work provides theoretical guidance for the development of MoS 2 -based sensors for detecting characteristic gases in air discharge.
- 18 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (19) , 22188-22195
Gas Sensitive Analysis of Composite MoS2(001) Surface to Air Discharge Products: A DFT Study
Air discharge characteristic gas detection is considered as one of the most reliable methods in fault diagnosis of air switchgear. Based on first-principles of density functional theory, the adsorption and sensing properties of transition metal (Cr) modified MoS 2 (001) monolayer on three air discharge characteristic gases (Co, NO, and NO 2 ) are analyzed. Firstly, the Cr doped MoS 2 (001) monolayer model is obtained by modeling and analyzing different doping sites, and the subsequent DFT calculation is done. Based on this model, many kinds of adsorption structures were constructed and optimized. The parameters of adsorption structure, adsorption energy and adsorption distance were obtained. Furthermore, the differential charge density, state density, response performance and adsorption performance of the system were analyzed. The results show that Cr-MoS 2 (001) monolayer has good adsorption performance to three gases. CO gas molecules can be rapidly desorbed from Cr-MoS 2 (001) monolayer surface at room temperature. Cr-MoS 2 (001) monolayer has the potential as a low-power gas sensor material for detecting CO characteristic gases and a solid adsorbent material for cleaning NO and NO 2 gases. This work provides theoretical guidance for the development of MoS 2 -based sensors for detecting characteristic gases in air discharge.