Gaolong Lv
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Gaolong Lv has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Shifeng Guo
Shenzhen Key Laboratory of Smart Sensing and Intelligent Systems, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
6 shared publications - Xinyu Wu
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences
4 shared publications - Huanqing Cao
Shenzhen Key Laboratory of Smart Sensing and Intelligent Systems, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
3 shared publications - Wei Feng
Chinese Academy of Sciences
3 shared publications - Qimin Zhu
Shenzhen Key Laboratory of Smart Sensing and Intelligent Systems, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
2 shared publications Wei Feng
Shenzhen Key Laboratory of Smart Sensing and Intelligent Systems, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China; University of Chinese Academy of Sciences, Beijing 100049, China; Guangdong Provincial Key Lab of Robotics and Intelligent System, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China
2 shared publications
Latest Publications
- Open Access
- 31 July 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (31)
Operando state monitoring of diversified lithium-ion batteries via laser-excited ultrasonic sensing with transformer networks
Emerging contact-based and immersion-based piezoelectric ultrasonic techniques encounter challenges in achieving accurate battery state monitoring under high-rate operations, where temperature fluctuations distort ultrasound and couplant contamination compromises practical applicability. Here, we propose air-coupled (couplant-free), noncontact laser-excited ultrasonic sensing (LEUS) with transformer networks for operando battery state estimation. The unique LEUS system pioneers a dual-laser design that uses a ring-shaped pulsed laser and a continuous laser to generate and detect high-quality ultrasound with a 10-fold increase in amplitude and a signal-to-noise ratio of 30 dB (16 dB higher than typical configurations), thereby enabling precise tracking of internal changes associated with state of charge (SoC) and state of health (SoH). By transforming ultrasonic signals into time-frequency scalograms, transformer networks autonomously extract discriminative features, eliminating manual feature engineering while achieving accurate prediction with mean errors below 5.7% for SoC and 2.1% for SoH. Through transfer learning, the base model generalizes rapidly to unseen chemistries, high-rate cycling, reducing training time and cost with minimal ultrasonic data. Extensive validation on over 100,000 ultrasonic signals from 40 commercial batteries, spanning two chemistries, three capacities, and 13 protocols, demonstrates the method’s robustness and reliability for operando battery state monitoring, paving the way for next-generation battery-management systems.
- 31 July 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (31)
Open AccessOperando state monitoring of diversified lithium-ion batteries via laser-excited ultrasonic sensing with transformer networks
Emerging contact-based and immersion-based piezoelectric ultrasonic techniques encounter challenges in achieving accurate battery state monitoring under high-rate operations, where temperature fluctuations distort ultrasound and couplant contamination compromises practical applicability. Here, we propose air-coupled (couplant-free), noncontact laser-excited ultrasonic sensing (LEUS) with transformer networks for operando battery state estimation. The unique LEUS system pioneers a dual-laser design that uses a ring-shaped pulsed laser and a continuous laser to generate and detect high-quality ultrasound with a 10-fold increase in amplitude and a signal-to-noise ratio of 30 dB (16 dB higher than typical configurations), thereby enabling precise tracking of internal changes associated with state of charge (SoC) and state of health (SoH). By transforming ultrasonic signals into time-frequency scalograms, transformer networks autonomously extract discriminative features, eliminating manual feature engineering while achieving accurate prediction with mean errors below 5.7% for SoC and 2.1% for SoH. Through transfer learning, the base model generalizes rapidly to unseen chemistries, high-rate cycling, reducing training time and cost with minimal ultrasonic data. Extensive validation on over 100,000 ultrasonic signals from 40 commercial batteries, spanning two chemistries, three capacities, and 13 protocols, demonstrates the method’s robustness and reliability for operando battery state monitoring, paving the way for next-generation battery-management systems.
- 28 May 2026
- Published by Elsevier in Ultrasonics
- Vol. 167, 108139
Plane-wave-like laser ultrasonic inspection of multilayer delaminations using beam-expanded excitation and migration-based reconstruction
- 28 May 2026
- Published by Elsevier in Ultrasonics
- Vol. 167, 108139
Plane-wave-like laser ultrasonic inspection of multilayer delaminations using beam-expanded excitation and migration-based reconstruction
- 14 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-13
Multifeature Fusion of Noncontact Laser-Generated Ultrasound and Electrochemical Data for In Situ Battery State Estimation
Accurate estimation of state of charge (SOC) and state of health (SOH) is critical for reliable battery operation monitoring. However, conventional voltage/current-based estimation methods face fundamental limitations due to their intrinsic voltage plateau and pronounced voltage hysteresis during fast operation (>3C). Non-invasive ultrasound sensing provides a promising solution for precise lithium-ion tracking, whereas conventional contact-based ultrasound techniques are limited by their reliance on coupling agents. This work therefore proposes a multi-feature fusion method that integrates non-contact laser generated ultrasound and electrochemical data, combined with machine learning algorithms, for in situ monitoring of battery SOC and SOH during high-rate cycles. A novel physics-informed, multi-kernel fusion hierarchical network for battery state estimation is proposed, which integrates electrochemical mechanism with Spearman correlation to construct a robust feature set from ultrasonic and electrochemical data. A dedicated multi-kernel architecture within a least squares support vector machine framework, where physically distinct features are processed through tailored kernels before fusion. The crested porcupine optimizer-enhanced hierarchical structure facilitates coherent, layered estimation of SOC and SOH, achieving RMSEs of 4.33% and 0.24%. To verify the robust generalization and stability of the proposed approach, extensive validation across diverse testing scenarios, including slow-charging (1C), cross-cell testing, and comparisons with other algorithms (random forest, long short-term memory, enhanced interpretable generalized additive model and extended Kalman filter) achieved superior accuracy in both SOC and SOH estimation. This multi-feature fusion strategy effectively circumvents the information limitations of single sensing modalities while establishing a novel paradigm for operando battery diagnostics. The in situ, non-contact nature and computational accuracy further suggest strong potential for industrial deployment in smart battery management systems.
- 14 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-13
Multifeature Fusion of Noncontact Laser-Generated Ultrasound and Electrochemical Data for In Situ Battery State Estimation
Accurate estimation of state of charge (SOC) and state of health (SOH) is critical for reliable battery operation monitoring. However, conventional voltage/current-based estimation methods face fundamental limitations due to their intrinsic voltage plateau and pronounced voltage hysteresis during fast operation (>3C). Non-invasive ultrasound sensing provides a promising solution for precise lithium-ion tracking, whereas conventional contact-based ultrasound techniques are limited by their reliance on coupling agents. This work therefore proposes a multi-feature fusion method that integrates non-contact laser generated ultrasound and electrochemical data, combined with machine learning algorithms, for in situ monitoring of battery SOC and SOH during high-rate cycles. A novel physics-informed, multi-kernel fusion hierarchical network for battery state estimation is proposed, which integrates electrochemical mechanism with Spearman correlation to construct a robust feature set from ultrasonic and electrochemical data. A dedicated multi-kernel architecture within a least squares support vector machine framework, where physically distinct features are processed through tailored kernels before fusion. The crested porcupine optimizer-enhanced hierarchical structure facilitates coherent, layered estimation of SOC and SOH, achieving RMSEs of 4.33% and 0.24%. To verify the robust generalization and stability of the proposed approach, extensive validation across diverse testing scenarios, including slow-charging (1C), cross-cell testing, and comparisons with other algorithms (random forest, long short-term memory, enhanced interpretable generalized additive model and extended Kalman filter) achieved superior accuracy in both SOC and SOH estimation. This multi-feature fusion strategy effectively circumvents the information limitations of single sensing modalities while establishing a novel paradigm for operando battery diagnostics. The in situ, non-contact nature and computational accuracy further suggest strong potential for industrial deployment in smart battery management systems.
- 30 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
In Situ Measurement of Elastic Constants and Thickness of Silicon–Underfill–Silicon Sandwiched Electronic Package Using Noncontact Laser Ultrasound Array
Quantifying Young’s modulus and Poisson’s ratio of underfill is critical in clarifying material failure mechanism for better optimizing their performance in reducing thermal stresses on solder joints in flip-chip devices. Conventional mechanical methods are generally developed for pure underfill block, and cannot directly measure thin underfill layers sandwiched in heterogeneous multilayered flip-chip devices during service. Non-destructive ultrasonic methods are more promising but are challenged by complex wave propagation induced by multilayered, anisotropic acoustic properties. This work presents a non-contact laser ultrasound (LU) method for in-situ and simultaneous measurement of elastic constants and thickness of thin underfill layer in silicon-underfill-silicon sandwich structure. Longitudinal and transverse waves in divergent propagation directions are acquired by using a line-scan point laser transmitter and fixed point receiver on opposite surfaces. Young’s modulus, Poisson’s ratio, and thickness of sandwiched underfill layer are inversely determined by iteratively matching experimental LU travel times with theoretical predictions considering interface reflection, refraction and mode conversion effects, with relative errors to ultrasonic measured reference values <3.722%. The in-situ LU method is advantageous in continuously monitoring mechanical property evolution during various cyclic tests, advancing reliability assessment and fabrication optimization of underfill materials in real-life electronic packaging applications.
- 30 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
In Situ Measurement of Elastic Constants and Thickness of Silicon–Underfill–Silicon Sandwiched Electronic Package Using Noncontact Laser Ultrasound Array
Quantifying Young’s modulus and Poisson’s ratio of underfill is critical in clarifying material failure mechanism for better optimizing their performance in reducing thermal stresses on solder joints in flip-chip devices. Conventional mechanical methods are generally developed for pure underfill block, and cannot directly measure thin underfill layers sandwiched in heterogeneous multilayered flip-chip devices during service. Non-destructive ultrasonic methods are more promising but are challenged by complex wave propagation induced by multilayered, anisotropic acoustic properties. This work presents a non-contact laser ultrasound (LU) method for in-situ and simultaneous measurement of elastic constants and thickness of thin underfill layer in silicon-underfill-silicon sandwich structure. Longitudinal and transverse waves in divergent propagation directions are acquired by using a line-scan point laser transmitter and fixed point receiver on opposite surfaces. Young’s modulus, Poisson’s ratio, and thickness of sandwiched underfill layer are inversely determined by iteratively matching experimental LU travel times with theoretical predictions considering interface reflection, refraction and mode conversion effects, with relative errors to ultrasonic measured reference values <3.722%. The in-situ LU method is advantageous in continuously monitoring mechanical property evolution during various cyclic tests, advancing reliability assessment and fabrication optimization of underfill materials in real-life electronic packaging applications.
- 9 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 68 (10) , 3216-3225
Automatic Quantification of Subsurface Defects by Analyzing Laser Ultrasonic Signals Using Convolutional Neural Networks and Wavelet Transform
The conventional machine learning algorithm for analyzing ultrasonic signals to detect structural defects necessarily identifies and extracts either time- or frequency-domain features manually, which has problems in reliability and effectiveness. This work proposes a novel approach by combining convolutional neural networks (CNNs) and wavelet transform to analyze the laser-generated ultrasonic signals for detecting the width of subsurface defects accurately. The novelty of this work is to convert the laser ultrasonic signals into the scalograms (images) via wavelet transform, which are subsequently utilized as the image input for the pretrained CNN to extract the defect features automatically to quantify the width of defects, avoiding the necessity and inaccuracy induced by artificial feature selection. The experimentally validated numerical model that simulates the interaction of laser-generated ultrasonic waves with subsurface defects is first established, which is further utilized to generate adequate laser ultrasonic signals for training the CNN model. A total number of 3104 data are obtained from simulation and experiments, with 2480 simulated signals for training the CNN model and the remaining 620 simulated data together with 4 experimental signals for verifying the performance of the proposed algorithm. This approach achieves the prediction accuracy of 98.5% on validation set, particularly with the prediction accuracy of 100% for the four experimental data. This work proves the feasibility and reliability of the proposed method for quantifying the width of subsurface defects and can be further expanded as a universal approach to various other defects detection, such as defect locations and shapes.
- 9 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 68 (10) , 3216-3225
Automatic Quantification of Subsurface Defects by Analyzing Laser Ultrasonic Signals Using Convolutional Neural Networks and Wavelet Transform
The conventional machine learning algorithm for analyzing ultrasonic signals to detect structural defects necessarily identifies and extracts either time- or frequency-domain features manually, which has problems in reliability and effectiveness. This work proposes a novel approach by combining convolutional neural networks (CNNs) and wavelet transform to analyze the laser-generated ultrasonic signals for detecting the width of subsurface defects accurately. The novelty of this work is to convert the laser ultrasonic signals into the scalograms (images) via wavelet transform, which are subsequently utilized as the image input for the pretrained CNN to extract the defect features automatically to quantify the width of defects, avoiding the necessity and inaccuracy induced by artificial feature selection. The experimentally validated numerical model that simulates the interaction of laser-generated ultrasonic waves with subsurface defects is first established, which is further utilized to generate adequate laser ultrasonic signals for training the CNN model. A total number of 3104 data are obtained from simulation and experiments, with 2480 simulated signals for training the CNN model and the remaining 620 simulated data together with 4 experimental signals for verifying the performance of the proposed algorithm. This approach achieves the prediction accuracy of 98.5% on validation set, particularly with the prediction accuracy of 100% for the four experimental data. This work proves the feasibility and reliability of the proposed method for quantifying the width of subsurface defects and can be further expanded as a universal approach to various other defects detection, such as defect locations and shapes.
- Open Access
- 1 January 2021
- Published by Wiley in Complexity
- Vol. 2021 (1)
Porosity Characterization of Thermal Barrier Coatings by Ultrasound with Genetic Algorithm Backpropagation Neural Network
Porosity is considered as one of the most important indicators for the characterization of the comprehensive performance of thermal barrier coatings (TBCs). In this study, the ultrasonic technique and the artificial neural network optimized with the genetic algorithm (GA_BPNN) are combined to develop an intelligent method for automatic detection and accurate prediction of TBCs’s porosity. A series of physical models of plasma‐sprayed ZrO2 coating are established with a thickness of 288 μm and porosity varying from 5.71% to 26.59%, and the ultrasonic reflection coefficient amplitude spectrum (URCAS) is constructed based on the time‐domain numerical simulation signal. The characteristic features (f1, f2, Amax, ΔA) of the URCAS, which are highly dependent on porosity, are extracted as input data to train the GA_BPNN model for predicting the unknown porosity. The average error of the prediction results is 1.45%, which suggests that the proposed method can achieve accurate detection and quantitative characterization of the porosity of TBCs with complex pore morphology.
- 1 January 2021
- Published by Wiley in Complexity
- Vol. 2021 (1)
Open AccessPorosity Characterization of Thermal Barrier Coatings by Ultrasound with Genetic Algorithm Backpropagation Neural Network
Porosity is considered as one of the most important indicators for the characterization of the comprehensive performance of thermal barrier coatings (TBCs). In this study, the ultrasonic technique and the artificial neural network optimized with the genetic algorithm (GA_BPNN) are combined to develop an intelligent method for automatic detection and accurate prediction of TBCs’s porosity. A series of physical models of plasma‐sprayed ZrO2 coating are established with a thickness of 288 μm and porosity varying from 5.71% to 26.59%, and the ultrasonic reflection coefficient amplitude spectrum (URCAS) is constructed based on the time‐domain numerical simulation signal. The characteristic features (f1, f2, Amax, ΔA) of the URCAS, which are highly dependent on porosity, are extracted as input data to train the GA_BPNN model for predicting the unknown porosity. The average error of the prediction results is 1.45%, which suggests that the proposed method can achieve accurate detection and quantitative characterization of the porosity of TBCs with complex pore morphology.