Dongguang Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Dongguang Zhang has a total of 55 co-authors. Here are the top co-authors by number of shared publications:
- Jiayi Yang
College of Artificial Intelligence and Computer Science, Xi’an University of Science and Technology, Xi’an, China
9 shared publications Yali Wu
Taiyuan University of Technology
9 shared publications- Yali Wu
College of Mechanical and Vehicle Engineering, Taiyuan University of Technology, Taiyuan 030024, China
6 shared publications Xiaoyan Xiong
4 shared publicationsYuheng Kuang
Taiyuan University of Technology
3 shared publicationsChengqi Liu
College of Robotics Science and Engineering
2 shared publications
Latest Publications
- 15 September 2026
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 8 (19) , 8575-8591
Liquid Metal Elastomer Resonant Sensors for Frequency-Domain Distributed Strain Sensing
Traditional sensor arrays face the trade-off dilemma between distribution density and circuit complexity, while strain sensors suffer from weak circuit stability under flexible deformation. Here, we propose a liquid metal elastomer resonant (LMER) sensor for frequency-domain distributed strain sensing, which employs liquid metal elastomers as the sensing elements and an RLC series circuit as the core component. The sensor enhances frequency selectivity and resonance-response stability by precisely regulating inductance and capacitance parameters, enabling stable and distinguishable resonant-frequency shifts under flexible deformation and thereby achieving efficient conversion from mechanical strain to frequency-domain signals. Experimental results demonstrate that the three LMER sensing units exhibit highly linear resonant-frequency-shift responses (R2 > 0.99) within the core compressive strain range of 0–30%, along with low rate dependence (<0.5%) and excellent long-term cyclic reliability, with frequency-response attenuation below 1% after 10,000 cycles. Furthermore, a distributed sensing system is constructed based on the parallel integration of LMER sensors with distinct resonant-frequency windows, enabling spatially distributed strain responses to be discriminated in the frequency domain through a shared signal line. This work establishes a frequency-domain distributed strain-sensing strategy based on LMER sensors, offering a compact and stable route for distributed flexible sensing in applications such as human–computer interaction, soft robotics, and wearable devices.
- 15 September 2026
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 8 (19) , 8575-8591
Liquid Metal Elastomer Resonant Sensors for Frequency-Domain Distributed Strain Sensing
Traditional sensor arrays face the trade-off dilemma between distribution density and circuit complexity, while strain sensors suffer from weak circuit stability under flexible deformation. Here, we propose a liquid metal elastomer resonant (LMER) sensor for frequency-domain distributed strain sensing, which employs liquid metal elastomers as the sensing elements and an RLC series circuit as the core component. The sensor enhances frequency selectivity and resonance-response stability by precisely regulating inductance and capacitance parameters, enabling stable and distinguishable resonant-frequency shifts under flexible deformation and thereby achieving efficient conversion from mechanical strain to frequency-domain signals. Experimental results demonstrate that the three LMER sensing units exhibit highly linear resonant-frequency-shift responses (R2 > 0.99) within the core compressive strain range of 0–30%, along with low rate dependence (<0.5%) and excellent long-term cyclic reliability, with frequency-response attenuation below 1% after 10,000 cycles. Furthermore, a distributed sensing system is constructed based on the parallel integration of LMER sensors with distinct resonant-frequency windows, enabling spatially distributed strain responses to be discriminated in the frequency domain through a shared signal line. This work establishes a frequency-domain distributed strain-sensing strategy based on LMER sensors, offering a compact and stable route for distributed flexible sensing in applications such as human–computer interaction, soft robotics, and wearable devices.
- 8 February 2026
- Published by Elsevier in Measurement
- Vol. 269
High-Sensitivity flexible strain sensor with the stress line extrusion effect
- 8 February 2026
- Published by Elsevier in Measurement
- Vol. 269
High-Sensitivity flexible strain sensor with the stress line extrusion effect
- 9 January 2026
- Published by American Chemical Society (ACS) in Langmuir
- Vol. 42 (3) , 2883-2895
ML-Driven Surface Structure Optimization of Aluminum Alloy Superhydrophobic Surfaces for Enhanced Anti-Icing Performance
Surface icing poses a serious threat to the safe operation of aerospace, transportation, and power transmission systems, highlighting the necessity of studying surface anti-icing performance. Superhydrophobic surfaces have been widely used in anti-icing applications; however, the relationship between their surface characteristics and anti-icing performance is very complex. Traditional experimental methods are costly and time-consuming, and optimizing anti-icing performance through surface characteristic tuning faces challenges. This study proposes an ML-driven optimization method for surface anti-icing performance. We use the gray level co-occurrence matrix (GLCM) to characterize multiscale surface features and predict the anti-icing performance of aluminum-based superhydrophobic surfaces. The key features for controlling the anti-icing performance were determined through various feature importance analysis methods. A mathematical model for optimizing anti-icing performance was constructed by combining these key features with classical nucleation theory. This model quantifies the synergistic regulatory effects of key features on anti-icing performance, elucidates their impact on freezing delay time, and provides a theoretical basis for the rational design of superhydrophobic anti-icing surfaces.
- 9 January 2026
- Published by American Chemical Society (ACS) in Langmuir
- Vol. 42 (3) , 2883-2895
ML-Driven Surface Structure Optimization of Aluminum Alloy Superhydrophobic Surfaces for Enhanced Anti-Icing Performance
Surface icing poses a serious threat to the safe operation of aerospace, transportation, and power transmission systems, highlighting the necessity of studying surface anti-icing performance. Superhydrophobic surfaces have been widely used in anti-icing applications; however, the relationship between their surface characteristics and anti-icing performance is very complex. Traditional experimental methods are costly and time-consuming, and optimizing anti-icing performance through surface characteristic tuning faces challenges. This study proposes an ML-driven optimization method for surface anti-icing performance. We use the gray level co-occurrence matrix (GLCM) to characterize multiscale surface features and predict the anti-icing performance of aluminum-based superhydrophobic surfaces. The key features for controlling the anti-icing performance were determined through various feature importance analysis methods. A mathematical model for optimizing anti-icing performance was constructed by combining these key features with classical nucleation theory. This model quantifies the synergistic regulatory effects of key features on anti-icing performance, elucidates their impact on freezing delay time, and provides a theoretical basis for the rational design of superhydrophobic anti-icing surfaces.
- 1 August 2025
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (8) , 086106
A lightweight gearbox fault diagnosis method based on multi-sensor information multi-dimensional fusion under small sample conditions
Given challenges such as single-sensor data’s susceptibility to environmental noise, limited fault identification accuracy, scarce and hard-to-collect fault data, and the large parameter sizes of existing fault diagnosis models, this study developed a small-sample lightweight multi-sensor information multi-dimensional fusion method. This model includes two parallel fusion modules, aiming to capture fault characteristics of the gearbox’s operating state in multiple dimensions. Firstly, sensor signals from different locations obtained through synchronous sampling are simultaneously input into the network. Two core fusion strategies are designed to extract multi-dimensional fault features: the common feature fusion (CFF) module and the associated feature fusion (AFF) module. The CFF module extracts common features of multiple sensors by minimizing the Frobenius norm between different sensor feature matrices. The AFF module extracts associated features among multi-sensor signals by introducing a channel-adaptive weighting mechanism within the convolutional block. Subsequently, the global features formed by fusing the common features and associated features are input into a classification module to obtain prediction labels. Finally, the loss function is constructed by combining the classification loss, the penalty term, and the Frobenius norm of the multi-sensor common feature matrix, and the model’s training is continuously optimized through the backpropagation algorithm. The proposed method has been rigorously evaluated through a series of comparative experiments on our own dataset and two public datasets, demonstrating its effectiveness, superior performance, and lightweight advantages. Moreover, with only 10 samples, an accuracy rate of up to 98% can be achieved for gearbox fault identification.
- 1 August 2025
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (8) , 086106
A lightweight gearbox fault diagnosis method based on multi-sensor information multi-dimensional fusion under small sample conditions
Given challenges such as single-sensor data’s susceptibility to environmental noise, limited fault identification accuracy, scarce and hard-to-collect fault data, and the large parameter sizes of existing fault diagnosis models, this study developed a small-sample lightweight multi-sensor information multi-dimensional fusion method. This model includes two parallel fusion modules, aiming to capture fault characteristics of the gearbox’s operating state in multiple dimensions. Firstly, sensor signals from different locations obtained through synchronous sampling are simultaneously input into the network. Two core fusion strategies are designed to extract multi-dimensional fault features: the common feature fusion (CFF) module and the associated feature fusion (AFF) module. The CFF module extracts common features of multiple sensors by minimizing the Frobenius norm between different sensor feature matrices. The AFF module extracts associated features among multi-sensor signals by introducing a channel-adaptive weighting mechanism within the convolutional block. Subsequently, the global features formed by fusing the common features and associated features are input into a classification module to obtain prediction labels. Finally, the loss function is constructed by combining the classification loss, the penalty term, and the Frobenius norm of the multi-sensor common feature matrix, and the model’s training is continuously optimized through the backpropagation algorithm. The proposed method has been rigorously evaluated through a series of comparative experiments on our own dataset and two public datasets, demonstrating its effectiveness, superior performance, and lightweight advantages. Moreover, with only 10 samples, an accuracy rate of up to 98% can be achieved for gearbox fault identification.
- 26 March 2025
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 7 (7) , 2981-2991
Liquid Metal Elastomer Foam-Based Soft Sensors with Decoupled Three-Axis Strain-Sensing Transmission
- 26 March 2025
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 7 (7) , 2981-2991
Liquid Metal Elastomer Foam-Based Soft Sensors with Decoupled Three-Axis Strain-Sensing Transmission
- 11 January 2025
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 17 (3) , 5527-5537
Liquid Metal Elastomer-Based U-Shaped Electrode Flexible Strain Sensors with Enhanced Stability and Sensitivity
Soft and stretchable strain sensors are crucial for applications in human-machine interfaces, flexible robotics, and electronic skin. Among these, capacitive strain sensors are widely used and studied; however, they face challenges due to material and structural constraints, such as low baseline capacitance and susceptibility to external interference, which result in low signal-to-noise ratios and poor stability. To address these issues, we propose a U-shaped electrode flexible strain sensor based on liquid metal elastomer (LME). By incorporating liquid metal into the elastomer, we create a high-dielectric constant LME dielectric layer. Additionally, we innovatively introduce an external U-shaped flexible textile electrode in conjunction with an internal planar flexible textile electrode to form the flexible strain sensor, effectively resolving the trade-off between stability and integration in capacitive sensors. We also investigated the dielectric and mechanical properties of the LME at varying liquid metal volume fraction (ϕLM). A comparative simulation was conducted between parallel plate structure and U-shaped package structure designs, leading to the fabrication of the U-shaped electrode flexible strain sensor based on LME. Furthermore, we demonstrated its potential applications in monitoring human motion, soft gripper electronic skin, and robotic arm movements.
- 11 January 2025
- Published by American Chemical Society (ACS) in ACS Applied Materials & Interfaces
- Vol. 17 (3) , 5527-5537
Liquid Metal Elastomer-Based U-Shaped Electrode Flexible Strain Sensors with Enhanced Stability and Sensitivity
Soft and stretchable strain sensors are crucial for applications in human-machine interfaces, flexible robotics, and electronic skin. Among these, capacitive strain sensors are widely used and studied; however, they face challenges due to material and structural constraints, such as low baseline capacitance and susceptibility to external interference, which result in low signal-to-noise ratios and poor stability. To address these issues, we propose a U-shaped electrode flexible strain sensor based on liquid metal elastomer (LME). By incorporating liquid metal into the elastomer, we create a high-dielectric constant LME dielectric layer. Additionally, we innovatively introduce an external U-shaped flexible textile electrode in conjunction with an internal planar flexible textile electrode to form the flexible strain sensor, effectively resolving the trade-off between stability and integration in capacitive sensors. We also investigated the dielectric and mechanical properties of the LME at varying liquid metal volume fraction (ϕLM). A comparative simulation was conducted between parallel plate structure and U-shaped package structure designs, leading to the fabrication of the U-shaped electrode flexible strain sensor based on LME. Furthermore, we demonstrated its potential applications in monitoring human motion, soft gripper electronic skin, and robotic arm movements.
- Open Access
- 29 April 2024
- Published by MDPI AG in Micromachines
- Vol. 15 (5) , 592
Flexible Thermoelectric Device Based on Protrusion-Structured Liquid Metal Elastomer for Gravity Heat Pipe
Monitoring the temperature of the coal gangue mountains is fundamental to preventing their spontaneous combustion. However, the existing temperature monitoring systems fail to achieve stable, pollution-free temperature monitoring without affecting vegetation growth in these mountains. To address this issue, this work proposes a flexible thermoelectric device (FTD) based on a protrusion-structured liquid metal elastomer (LME). Utilizing a high-thermal-conductivity LME, the FTD adheres closely to the surface of the gravity heat pipe (GHP), ensuring compatibility between FTD and the curved surface of the GHP. Simultaneously, employing a low-thermal-conductivity elastomer helps concentrate heat onto FTD, thereby enhancing thermoelectric power generation efficiency. Additionally, the impact of the shape, size, and height of the protrusion structure at the cold end of the GHP on its efficiency was also investigated. The practical application of FTD on GHP was demonstrated.
- 29 April 2024
- Published by MDPI AG in Micromachines
- Vol. 15 (5) , 592
Open AccessFlexible Thermoelectric Device Based on Protrusion-Structured Liquid Metal Elastomer for Gravity Heat Pipe
Monitoring the temperature of the coal gangue mountains is fundamental to preventing their spontaneous combustion. However, the existing temperature monitoring systems fail to achieve stable, pollution-free temperature monitoring without affecting vegetation growth in these mountains. To address this issue, this work proposes a flexible thermoelectric device (FTD) based on a protrusion-structured liquid metal elastomer (LME). Utilizing a high-thermal-conductivity LME, the FTD adheres closely to the surface of the gravity heat pipe (GHP), ensuring compatibility between FTD and the curved surface of the GHP. Simultaneously, employing a low-thermal-conductivity elastomer helps concentrate heat onto FTD, thereby enhancing thermoelectric power generation efficiency. Additionally, the impact of the shape, size, and height of the protrusion structure at the cold end of the GHP on its efficiency was also investigated. The practical application of FTD on GHP was demonstrated.
- 24 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (9) , 9176-9182
Convex Microarrays-Based Liquid Metal Soft Piezoresistive Stress Sensor With High Sensitivity and Large Measurement Range
Liquid metal (LM) soft piezoresistive sensors-that is, devices that change resistance in response to mechanical forces have low hysteresis and high stability, which exhibit great promise for human health monitoring. Applying stress reduces the cross-sectional area of the LM conductive path in the sensor, increasing the resistance and enabling the measurement of stress. However, increasing the sensitivity of the sensor results in a reduced measurement range and an increase in the cost. To solve this problem, this paper presents a convex microarrays (CMs)-based LM soft piezoresistive stress sensor with high sensitivity and large measurement range. The CMs are located inside the microchannel, which improves the sensitivity of the sensor without reducing the measurement range. The CMs also enable the sensor to monitor bending. The mechanism of the CMs is verified by theoretical analysis, finite element simulation and experiments. The relationship between the height of the CMs and the sensitivity is investigated. The sensor can be used in the field of human health monitoring.
- 24 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (9) , 9176-9182
Convex Microarrays-Based Liquid Metal Soft Piezoresistive Stress Sensor With High Sensitivity and Large Measurement Range
Liquid metal (LM) soft piezoresistive sensors-that is, devices that change resistance in response to mechanical forces have low hysteresis and high stability, which exhibit great promise for human health monitoring. Applying stress reduces the cross-sectional area of the LM conductive path in the sensor, increasing the resistance and enabling the measurement of stress. However, increasing the sensitivity of the sensor results in a reduced measurement range and an increase in the cost. To solve this problem, this paper presents a convex microarrays (CMs)-based LM soft piezoresistive stress sensor with high sensitivity and large measurement range. The CMs are located inside the microchannel, which improves the sensitivity of the sensor without reducing the measurement range. The CMs also enable the sensor to monitor bending. The mechanism of the CMs is verified by theoretical analysis, finite element simulation and experiments. The relationship between the height of the CMs and the sensitivity is investigated. The sensor can be used in the field of human health monitoring.
- Open Access
- 23 March 2023
- Published by MDPI AG in Micromachines
- Vol. 14 (4) , 716
Heterogeneous Multi-Material Flexible Piezoresistive Sensor with High Sensitivity and Wide Measurement Range
Flexible piezoresistive sensors (FPSs) have the advantages of compact structure, convenient signal acquisition and fast dynamic response; they are widely used in motion detection, wearable electronic devices and electronic skins. FPSs accomplish the measurement of stresses through piezoresistive material (PM). However, FPSs based on a single PM cannot achieve high sensitivity and wide measurement range simultaneously. To solve this problem, a heterogeneous multi-material flexible piezoresistive sensor (HMFPS) with high sensitivity and a wide measurement range is proposed. The HMFPS consists of a graphene foam (GF), a PDMS layer and an interdigital electrode. Among them, the GF serves as a sensing layer, providing high sensitivity, and the PDMS serves as a supporting layer, providing a large measurement range. The influence and principle of the heterogeneous multi-material (HM) on the piezoresistivity were investigated by comparing the three HMFPS with different sizes. The HM proved to be an effective way to produce flexible sensors with high sensitivity and a wide measurement range. The HMFPS-10 has a sensitivity of 0.695 kPa−1, a measurement range of 0–14,122 kPa, fast response/recovery (83 ms and 166 ms) and excellent stability (2000 cycles). In addition, the potential application of the HMFPS-10 in human motion monitoring was demonstrated.
- 23 March 2023
- Published by MDPI AG in Micromachines
- Vol. 14 (4) , 716
Open AccessHeterogeneous Multi-Material Flexible Piezoresistive Sensor with High Sensitivity and Wide Measurement Range
Flexible piezoresistive sensors (FPSs) have the advantages of compact structure, convenient signal acquisition and fast dynamic response; they are widely used in motion detection, wearable electronic devices and electronic skins. FPSs accomplish the measurement of stresses through piezoresistive material (PM). However, FPSs based on a single PM cannot achieve high sensitivity and wide measurement range simultaneously. To solve this problem, a heterogeneous multi-material flexible piezoresistive sensor (HMFPS) with high sensitivity and a wide measurement range is proposed. The HMFPS consists of a graphene foam (GF), a PDMS layer and an interdigital electrode. Among them, the GF serves as a sensing layer, providing high sensitivity, and the PDMS serves as a supporting layer, providing a large measurement range. The influence and principle of the heterogeneous multi-material (HM) on the piezoresistivity were investigated by comparing the three HMFPS with different sizes. The HM proved to be an effective way to produce flexible sensors with high sensitivity and a wide measurement range. The HMFPS-10 has a sensitivity of 0.695 kPa−1, a measurement range of 0–14,122 kPa, fast response/recovery (83 ms and 166 ms) and excellent stability (2000 cycles). In addition, the potential application of the HMFPS-10 in human motion monitoring was demonstrated.
- 8 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (6) , 5841-5848
Biomimetic Liquid Metal Elastomer Foam With Stress Sensing
Liquid metal elastomer has a high permittivity due to the equipotentiality of liquid metal (LM) particles. Introducing a porous structure prepares LM elastomer foam (LMEF) with piezopermittivity, which has a high–stress sensitivity as a stress sensor. However, the random porous structure of LMEF limits the piezopermittivity and the improvement of sensitivity. To solve this problem, a biomimetic LMEF (BLMEF) inspired by the lotus seeds is introduced that has high porosity (>70%), and large permittivity variation (≈5.6). Mechanical and dielectric properties are investigated through finite element analysis and experiments. A flexible capacitive stress sensor is fabricated based on BLMEF. The BLMEF–based stress sensor has high stress sensitivity (2.59 kPa -1 ) and high stability. The sensor can measure human motions, demonstrating potential applications in the field of human motion monitoring. In addition, a smart ergonomic cushion is demonstrated.
- 8 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (6) , 5841-5848
Biomimetic Liquid Metal Elastomer Foam With Stress Sensing
Liquid metal elastomer has a high permittivity due to the equipotentiality of liquid metal (LM) particles. Introducing a porous structure prepares LM elastomer foam (LMEF) with piezopermittivity, which has a high–stress sensitivity as a stress sensor. However, the random porous structure of LMEF limits the piezopermittivity and the improvement of sensitivity. To solve this problem, a biomimetic LMEF (BLMEF) inspired by the lotus seeds is introduced that has high porosity (>70%), and large permittivity variation (≈5.6). Mechanical and dielectric properties are investigated through finite element analysis and experiments. A flexible capacitive stress sensor is fabricated based on BLMEF. The BLMEF–based stress sensor has high stress sensitivity (2.59 kPa -1 ) and high stability. The sensor can measure human motions, demonstrating potential applications in the field of human motion monitoring. In addition, a smart ergonomic cushion is demonstrated.