Wenyu Jiang
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
Wenyu Jiang has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
- Aayush Madan
School of Civil and Environmental Engineering, Nanyang Technological University, Singapore 639798, Singapore
2 shared publications - Kejie Huang
Zhejiang University
2 shared publications - Kian Ann Ng
Engineering Cluster, Singapore Institute of Technology, Singapore 828608, Singapore
2 shared publications - Zeyu Wang
College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China
2 shared publications Andres Garcia Coleto
1 shared publicationArka Majumdar
1 shared publication
Latest Publications
- Open Access
- 1 May 2026
- Published by MDPI AG in Micromachines
- Vol. 17 (5) , 566
Wall Thinning Monitoring in Boiler U-Bends: A Review and Future Prospects with Fiber Optic Sensing
Tube boilers are extensively employed in oil and gas refineries, as well as in petroleum, energy, and power generation industries, where they serve critical functions in local steam-generation units and combined-cycle gas turbine (CCGT) plants. However, these boilers are prone to defects arising from waterside corrosion (e.g., thinning of U-bend tubes), fireside corrosion, and material degradation caused by stress or creeping. Among these issues, wall thinning of tube bends is particularly severe, as it results in localized metal loss, reduced structural integrity, and an elevated risk of tube rupture or failure under high-temperature and high-pressure operating conditions. Such failures can significantly compromise boiler safety and efficiency, potentially leading to forced outages, costly unplanned repairs, or catastrophic damage if not detected in time. The current condition-monitoring policy for U-bends relies on scheduled preventive maintenance and unscheduled corrective interventions. In practice, this involves randomly checking approximately 10–20% of the tubes through spot scanning, partial scanning, or full scanning, with repairs typically carried out only after an undetected failure occurs. Such maintenance strategies generally require plant shutdowns, making the process time-consuming, labor-intensive, and ultimately not cost-effective. This paper reviews existing solutions, technologies, and research addressing the problem, and introduces femtosecond laser micromachined fiber optic sensors as a transformative approach for real-time monitoring of wall thickness reduction in U-bend boiler tubes, thereby opening pathways for further research.
- 1 May 2026
- Published by MDPI AG in Micromachines
- Vol. 17 (5) , 566
Open AccessWall Thinning Monitoring in Boiler U-Bends: A Review and Future Prospects with Fiber Optic Sensing
Tube boilers are extensively employed in oil and gas refineries, as well as in petroleum, energy, and power generation industries, where they serve critical functions in local steam-generation units and combined-cycle gas turbine (CCGT) plants. However, these boilers are prone to defects arising from waterside corrosion (e.g., thinning of U-bend tubes), fireside corrosion, and material degradation caused by stress or creeping. Among these issues, wall thinning of tube bends is particularly severe, as it results in localized metal loss, reduced structural integrity, and an elevated risk of tube rupture or failure under high-temperature and high-pressure operating conditions. Such failures can significantly compromise boiler safety and efficiency, potentially leading to forced outages, costly unplanned repairs, or catastrophic damage if not detected in time. The current condition-monitoring policy for U-bends relies on scheduled preventive maintenance and unscheduled corrective interventions. In practice, this involves randomly checking approximately 10–20% of the tubes through spot scanning, partial scanning, or full scanning, with repairs typically carried out only after an undetected failure occurs. Such maintenance strategies generally require plant shutdowns, making the process time-consuming, labor-intensive, and ultimately not cost-effective. This paper reviews existing solutions, technologies, and research addressing the problem, and introduces femtosecond laser micromachined fiber optic sensors as a transformative approach for real-time monitoring of wall thickness reduction in U-bend boiler tubes, thereby opening pathways for further research.
- Open Access
- 1 March 2026
- Published by AIP Publishing in APL Engineering Physics
- Vol. 1 (1)
Enhancing noise robustness in SiN microring photonic reservoirs enabled by 2D material nonlinearities
Photonic reservoir computing offers a promising route toward ultrafast, low-energy temporal signal processing, but its performance is intrinsically limited by the nonlinear characteristics and noise of its nodes. Conventional silicon microring reservoirs suffer from high optical loss and thermo-optic fluctuations, which raise the normalized mean square error (NMSE) and reduce the signal-to-noise ratio (SNR). While silicon nitride (SiN) mitigates optical losses, it has the weakness of lacking inherent nonlinearity and bistability. Here, we incorporate a monolayer of WSe2 onto a SiN microring resonator (MRR) to enhance the third-order susceptibility, resulting in a sharper resonance response and a deeper bistable potential. This hybrid integration improves dynamic stability against noise, suppresses unwanted resonance drift, and enhances noise robustness. The WSe2–SiN MRR reservoirs achieve a 4 dB increase in SNR and a 30% reduction in NMSE. By integrating with 2D material layers, training accuracy reaches ∼94% in 30–40 iterations, which is over five times faster than the bare SiN counterpart. We evaluated time-series prediction using the NARMA-10 and NARMA-20 benchmarks. Both simulations and experiments confirm robust hysteresis, longer memory retention, and stable operation exceeding 5000 cycles, highlighting a practical pathway toward noise-tolerant, energy-efficient photonic reservoirs for high-speed edge computing and signal prediction.
- 1 March 2026
- Published by AIP Publishing in APL Engineering Physics
- Vol. 1 (1)
Open AccessEnhancing noise robustness in SiN microring photonic reservoirs enabled by 2D material nonlinearities
Photonic reservoir computing offers a promising route toward ultrafast, low-energy temporal signal processing, but its performance is intrinsically limited by the nonlinear characteristics and noise of its nodes. Conventional silicon microring reservoirs suffer from high optical loss and thermo-optic fluctuations, which raise the normalized mean square error (NMSE) and reduce the signal-to-noise ratio (SNR). While silicon nitride (SiN) mitigates optical losses, it has the weakness of lacking inherent nonlinearity and bistability. Here, we incorporate a monolayer of WSe2 onto a SiN microring resonator (MRR) to enhance the third-order susceptibility, resulting in a sharper resonance response and a deeper bistable potential. This hybrid integration improves dynamic stability against noise, suppresses unwanted resonance drift, and enhances noise robustness. The WSe2–SiN MRR reservoirs achieve a 4 dB increase in SNR and a 30% reduction in NMSE. By integrating with 2D material layers, training accuracy reaches ∼94% in 30–40 iterations, which is over five times faster than the bare SiN counterpart. We evaluated time-series prediction using the NARMA-10 and NARMA-20 benchmarks. Both simulations and experiments confirm robust hysteresis, longer memory retention, and stable operation exceeding 5000 cycles, highlighting a practical pathway toward noise-tolerant, energy-efficient photonic reservoirs for high-speed edge computing and signal prediction.
- Open Access
The emerging applications of silicon photonics
- 17 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech and Language Processing
- Vol. 33, 4797-4807
Parallel Delayed Memory Units for Enhanced Temporal Modeling in Biomedical and Bioacoustic Signal Analysis
Advanced deep learning architectures, particularly recurrent neural networks (RNNs), have been widely applied in audio, bioacoustic, and biomedical signal analysis, especially in data-scarce environments. While gated RNNs remain effective, they can be relatively over-parameterised and less training-efficient in some regimes [1], [2], while linear RNNs tend to fall short in capturing the complexity inherent in bio-signals. To address these challenges, we propose the Parallel Delayed Memory Unit (PDMU), a delay-gated state-space module for short-term temporal credit assignment targeting audio and bioacoustic signals, which enhances short-term temporal state interactions and memory efficiency via a gated delay-line mechanism. Unlike previous Delayed Memory Units (DMU) that embed temporal dynamics into the delay-line architecture, the PDMU further compresses temporal information into vector representations using Legendre Memory Units (LMU). This design serves as a form of causal attention, allowing the model to dynamically adjust its reliance on past states and improve real-time learning performance. Notably, in low-information scenarios, the gating mechanism behaves similarly to skip connections by bypassing state decay and preserving early representations, thereby facilitating long-term memory retention. The PDMU is modular, supporting parallel training and sequential inference, and can be easily integrated into existing linear RNN frameworks. Furthermore, we introduce bidirectional, efficient, and spiking variants of the architecture, each offering additional gains in performance or energy efficiency. Experimental results on diverse audio and biomedical benchmarks demonstrate that the PDMU significantly enhances both memory capacity and overall model performance.
- 17 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech and Language Processing
- Vol. 33, 4797-4807
Parallel Delayed Memory Units for Enhanced Temporal Modeling in Biomedical and Bioacoustic Signal Analysis
Advanced deep learning architectures, particularly recurrent neural networks (RNNs), have been widely applied in audio, bioacoustic, and biomedical signal analysis, especially in data-scarce environments. While gated RNNs remain effective, they can be relatively over-parameterised and less training-efficient in some regimes [1], [2], while linear RNNs tend to fall short in capturing the complexity inherent in bio-signals. To address these challenges, we propose the Parallel Delayed Memory Unit (PDMU), a delay-gated state-space module for short-term temporal credit assignment targeting audio and bioacoustic signals, which enhances short-term temporal state interactions and memory efficiency via a gated delay-line mechanism. Unlike previous Delayed Memory Units (DMU) that embed temporal dynamics into the delay-line architecture, the PDMU further compresses temporal information into vector representations using Legendre Memory Units (LMU). This design serves as a form of causal attention, allowing the model to dynamically adjust its reliance on past states and improve real-time learning performance. Notably, in low-information scenarios, the gating mechanism behaves similarly to skip connections by bypassing state decay and preserving early representations, thereby facilitating long-term memory retention. The PDMU is modular, supporting parallel training and sequential inference, and can be easily integrated into existing linear RNN frameworks. Furthermore, we introduce bidirectional, efficient, and spiking variants of the architecture, each offering additional gains in performance or energy efficiency. Experimental results on diverse audio and biomedical benchmarks demonstrate that the PDMU significantly enhances both memory capacity and overall model performance.
- 23 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Magazine
- Vol. 16 (3) , 97-98
The IEEE Solid-State Circuits Society Singapore Chapter Organized a Series of Technical Lectures in the Fourth Week of May 2024, Featuring Three Distinguished Speakers [Chapters]
Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.
- 23 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Magazine
- Vol. 16 (3) , 97-98
The IEEE Solid-State Circuits Society Singapore Chapter Organized a Series of Technical Lectures in the Fourth Week of May 2024, Featuring Three Distinguished Speakers [Chapters]
Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (13) , 21012-21024
A Fourier-Transform-Based Framework With Asymptotic Attention for Mobile Thermal InfraRed Object Detection
Thermal InfraRed (TIR) technology has emerged as a significant tool in autonomous driving systems. Unlike natural images, TIR images are distinguished by their enriched thermal and illumination information while lacking chromatic contrast. Traditional object detection on natural images normally use deep neural networks based on convolutional layers or attention modules. However, TIR-based object detection necessitates high computational efficiency to eliminate the extraction of redundant chromatic features. Furthermore, the robust space-frequency perception and expansive receptive field are critical due to the distinct brightness and contour features of TIR images. In this paper, we propose a novel network, namely Lightweight-Fourier-Transform Detector ( LFTDet ), meticulously designed to strike a balance between computational efficiency and accuracy in TIR object detection. Specifically, our innovative Fourier Transform-Efficient Layer Aggregation Network (FT-ELAN) backbone takes advantage of Fourier Transform (FT) in synergy with deep neural networks. In addition, we propose the detection neck called Asymptotic Attention-based Feature Pyramid Network (AA-FPN), which integrates the SimA mechanism in the asymptotic structure to facilitate the FT-based operation. Extensive experiments conducted on FLIR and LLVIP datasets demonstrate that LFTDet surpasses all baselines while maintaining an extremely low computational cost. Code is available at https://github.com/zeyuwang-zju/LFTDet.
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (13) , 21012-21024
A Fourier-Transform-Based Framework With Asymptotic Attention for Mobile Thermal InfraRed Object Detection
Thermal InfraRed (TIR) technology has emerged as a significant tool in autonomous driving systems. Unlike natural images, TIR images are distinguished by their enriched thermal and illumination information while lacking chromatic contrast. Traditional object detection on natural images normally use deep neural networks based on convolutional layers or attention modules. However, TIR-based object detection necessitates high computational efficiency to eliminate the extraction of redundant chromatic features. Furthermore, the robust space-frequency perception and expansive receptive field are critical due to the distinct brightness and contour features of TIR images. In this paper, we propose a novel network, namely Lightweight-Fourier-Transform Detector ( LFTDet ), meticulously designed to strike a balance between computational efficiency and accuracy in TIR object detection. Specifically, our innovative Fourier Transform-Efficient Layer Aggregation Network (FT-ELAN) backbone takes advantage of Fourier Transform (FT) in synergy with deep neural networks. In addition, we propose the detection neck called Asymptotic Attention-based Feature Pyramid Network (AA-FPN), which integrates the SimA mechanism in the asymptotic structure to facilitate the FT-based operation. Extensive experiments conducted on FLIR and LLVIP datasets demonstrate that LFTDet surpasses all baselines while maintaining an extremely low computational cost. Code is available at https://github.com/zeyuwang-zju/LFTDet.
- 19 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Magazine
- Vol. 16 (1) , 60-61
SSCS Singapore Chapter Update [Chapters]
Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.
- 19 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Solid-State Circuits Magazine
- Vol. 16 (1) , 60-61
SSCS Singapore Chapter Update [Chapters]
Provides society information that may include news, reviews or technical notes that should be of interest to practitioners and researchers.
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2663-2672
TIRDet: Mono-Modality Thermal InfraRed Object Detection Based on Prior Thermal-To-Visible Translation
Cross-modality images that combine visible-infrared spectra can provide complementary information for object detection. In particular, they are well-suited for autonomous vehicle applications in dark environments with limited illumination. However, it is time-consuming to acquire a large number of pixel-aligned visible-thermal image pairs, and real-time alignment is challenging in practical driving systems. Furthermore, the quality of visible-spectrum images can be adversely affected by complex environmental conditions. In this paper, we propose a novel neural network called TIRDet, which only utilizes Thermal InfraRed (TIR) images for mono-modality object detection. To compensate for the lacked visible-band information, we adopt a prior Thermal-To-Visible (T2V) translation model to obtain the translated visible images and the latent T2V codes. In addition, we introduce a novel attention-based Cross-Modality Aggregation (CMA) module, which can augment the modality-translation awareness of TIRDet by preserving the T2V semantic information. Extensive experiments on FLIR and LLVIP datasets demonstrate that our TIRDet significantly outperforms all mono-modality detection methods based on thermal images, and it even surpasses most State-Of-The-Art (SOTA) multispectral methods using visible-thermal image pairs. Code is available at https://github.com/zeyuwang-zju/TIRDet
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2663-2672
Open AccessTIRDet: Mono-Modality Thermal InfraRed Object Detection Based on Prior Thermal-To-Visible Translation
Cross-modality images that combine visible-infrared spectra can provide complementary information for object detection. In particular, they are well-suited for autonomous vehicle applications in dark environments with limited illumination. However, it is time-consuming to acquire a large number of pixel-aligned visible-thermal image pairs, and real-time alignment is challenging in practical driving systems. Furthermore, the quality of visible-spectrum images can be adversely affected by complex environmental conditions. In this paper, we propose a novel neural network called TIRDet, which only utilizes Thermal InfraRed (TIR) images for mono-modality object detection. To compensate for the lacked visible-band information, we adopt a prior Thermal-To-Visible (T2V) translation model to obtain the translated visible images and the latent T2V codes. In addition, we introduce a novel attention-based Cross-Modality Aggregation (CMA) module, which can augment the modality-translation awareness of TIRDet by preserving the T2V semantic information. Extensive experiments on FLIR and LLVIP datasets demonstrate that our TIRDet significantly outperforms all mono-modality detection methods based on thermal images, and it even surpasses most State-Of-The-Art (SOTA) multispectral methods using visible-thermal image pairs. Code is available at https://github.com/zeyuwang-zju/TIRDet
- Open Access
- 15 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 8 (2) , 111-118
An Energy Efficient Time-Multiplexing Computing-in-Memory Architecture for Edge Intelligence
The growing data volume and complexity of Deep Neural Networks (DNN) require new architectures to surpass the limitation of the von-Neumann bottleneck, with Computing-in-memory (CIM) as a promising direction for implementing energy-efficient neural networks. However, CIM’s peripheral sensing circuits are usually power and area hungry components. We propose a time-multiplexing Computing-in-Memory architecture (TM-CIM) based on memristive analog computing to share the peripheral circuits and process one column at a time. The memristor array is arranged in a column-wise manner that avoids wasting power/energy on unselected columns. In addition, DAC (digital-to-analog converter) power and energy efficiency, which turns out to be an even greater overhead than ADC (analog-to-digital converter), can be fine-tuned in TM-CIM for significant improvement. For a 256*256 crossbar array with a typical setting, TM-CIM saves 18.4× in energy with 0.136 pJ/MAC efficiency, and 19.9× area for 1T1R case and 15.9× for 2T2R case. Performance estimation on VGG-16 indicates that TM-CIM can save over 16× area. A trade-off between the chip area, peak power, and latency is also presented, with a proposed scheme to further reduce the latency on VGG-16, without significantly increasing chip area and peak power.
- 15 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 8 (2) , 111-118
Open AccessAn Energy Efficient Time-Multiplexing Computing-in-Memory Architecture for Edge Intelligence
The growing data volume and complexity of Deep Neural Networks (DNN) require new architectures to surpass the limitation of the von-Neumann bottleneck, with Computing-in-memory (CIM) as a promising direction for implementing energy-efficient neural networks. However, CIM’s peripheral sensing circuits are usually power and area hungry components. We propose a time-multiplexing Computing-in-Memory architecture (TM-CIM) based on memristive analog computing to share the peripheral circuits and process one column at a time. The memristor array is arranged in a column-wise manner that avoids wasting power/energy on unselected columns. In addition, DAC (digital-to-analog converter) power and energy efficiency, which turns out to be an even greater overhead than ADC (analog-to-digital converter), can be fine-tuned in TM-CIM for significant improvement. For a 256*256 crossbar array with a typical setting, TM-CIM saves 18.4× in energy with 0.136 pJ/MAC efficiency, and 19.9× area for 1T1R case and 15.9× for 2T2R case. Performance estimation on VGG-16 indicates that TM-CIM can save over 16× area. A trade-off between the chip area, peak power, and latency is also presented, with a proposed scheme to further reduce the latency on VGG-16, without significantly increasing chip area and peak power.
- 16 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 69 (1) , 309-321
In Situ Aging-Aware Error Monitoring Scheme for IMPLY-Based Memristive Computing-in-Memory Systems
Stateful logic through memristor is a promising technology to build Computing-in-Memory (CIM) systems. However, aging-induced degradation of memristors’ threshold voltage imposes a major challenge to the reliability and guardbands estimation of memristive CIM systems, especially the Material Implication (IMPLY) logic based CIM systems. In this paper, a novel in-situ aging-aware error monitoring scheme for memristor-based IMPLY logic is proposed. The proposed in-situ error monitoring scheme can achieve faster error detection speed and higher detection accuracy than the straightforward program-verify monitoring scheme. Simulation results under Monte-Carlo simulation show that the proposed monitoring scheme can effectively detect the major operation failures existing in IMPLY logic operations with a detection accuracy up to 99.95%. Moreover, a case study of error monitoring design of 4-bit IMPLY-based adder is carried out. The analysis result exhibits that the proposed in-situ monitoring scheme can achieve 75.2% improvement on the detection speed against the program-verify scheme. Further analysis on a convolution filter in VGG-11 based Binarized Neural Network shows that 74% improvement on the detection speed can also be achieved by using the proposed monitoring scheme, which suggests that the proposed in-situ error monitoring scheme is an efficient solution to improve the reliability of IMPLY-based memristive CIM systems.
- 16 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 69 (1) , 309-321
In Situ Aging-Aware Error Monitoring Scheme for IMPLY-Based Memristive Computing-in-Memory Systems
Stateful logic through memristor is a promising technology to build Computing-in-Memory (CIM) systems. However, aging-induced degradation of memristors’ threshold voltage imposes a major challenge to the reliability and guardbands estimation of memristive CIM systems, especially the Material Implication (IMPLY) logic based CIM systems. In this paper, a novel in-situ aging-aware error monitoring scheme for memristor-based IMPLY logic is proposed. The proposed in-situ error monitoring scheme can achieve faster error detection speed and higher detection accuracy than the straightforward program-verify monitoring scheme. Simulation results under Monte-Carlo simulation show that the proposed monitoring scheme can effectively detect the major operation failures existing in IMPLY logic operations with a detection accuracy up to 99.95%. Moreover, a case study of error monitoring design of 4-bit IMPLY-based adder is carried out. The analysis result exhibits that the proposed in-situ monitoring scheme can achieve 75.2% improvement on the detection speed against the program-verify scheme. Further analysis on a convolution filter in VGG-11 based Binarized Neural Network shows that 74% improvement on the detection speed can also be achieved by using the proposed monitoring scheme, which suggests that the proposed in-situ error monitoring scheme is an efficient solution to improve the reliability of IMPLY-based memristive CIM systems.