Xin Sun
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
Research Subjects
Xin Sun has a total of 123 co-authors. Here are the top co-authors by number of shared publications:
Hongwei Li
14 shared publicationsRan Hao
School of Environmental and Municipal Engineering, Xi’an University of Architecture and Technology, Xi’an, China
9 shared publicationsPengfei Li
7 shared publicationsAli Turab
6 shared publicationsLei Ye
6 shared publicationsBinchen Mao
5 shared publications
Latest Publications
- 21 September 2026
- Published by Elsevier in Journal of Cleaner Production
- Vol. 577
Extracellular ROS fingerprinting guides mixed-gas ARTP optimization for lipid-oriented breeding of Chlorella sorokiniana
- 21 September 2026
- Published by Elsevier in Journal of Cleaner Production
- Vol. 577
Extracellular ROS fingerprinting guides mixed-gas ARTP optimization for lipid-oriented breeding of Chlorella sorokiniana
- Open Access
- 21 September 2026
- Published by Elsevier in SSRN Electronic Journal
Explainable, Imbalance-Aware Algal-Density Prediction and Model-Derived Threshold Comparison across Lakes with Contrasting Trophic States
Eutrophication is increasing the risk of algal blooms in lakes, yet monitoring records commonly contain few high-density events, limiting the ability of conventional regression models to represent bloom extremes. This study developed an imbalance-aware classification–regression framework using 38,402 paired water-quality and meteorological observations collected from November 2020 to November 2025 in four lakes on the Yunnan–Guizhou Plateau: oligotrophic Fuxian Lake, mesotrophic Erhai Lake, and eutrophic Dianchi and Xingyun lakes. Random forest, XGBoost, and LightGBM were evaluated as standalone and hybrid models. The hybrid framework distinguished low- and high-density states at 1.0×107 cells L-1 and used regime-specific regressors to predict continuous algal density. SMOTE was restricted to classifier training, and performance was evaluated using the original chronologically held-out observations. SHapley Additive exPlanations (SHAP) were used to characterize model-attributed environmental relationships, while the normalized threshold offset (NTO) standardized selected SHAP zero-crossing positions against lake-specific study-period distributions. The largest improvement occurred in Fuxian Lake, where the SMOTE-assisted random-forest framework increased test R2 from 0.27 to 0.75. SMOTE increased classifier recall from 37.5% to 81.2% and F1 score from 52.2% to 74.3%. The corresponding R2 values remained 0.79, 0.94, and 0.70 for Erhai, Dianchi, and Xingyun lakes, respectively. SHAP identified water temperature as the leading predictor in Fuxian Lake and wind speed in Erhai Lake, whereas conductivity and ammonium nitrogen ranked highly in Dianchi Lake and turbidity and conductivity in Xingyun Lake. NTO placed model-derived transitions on a common standardized scale: nutrient-related transitions in the eutrophic lakes generally occurred closer to their historical means, whereas several transitions in Fuxian and Erhai lakes occurred farther into the upper tails of their distributions. The framework improves high-density prediction under severe class imbalance and provides a standardized basis for comparing model-attributed environmental responses among lakes.
- 21 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessExplainable, Imbalance-Aware Algal-Density Prediction and Model-Derived Threshold Comparison across Lakes with Contrasting Trophic States
Eutrophication is increasing the risk of algal blooms in lakes, yet monitoring records commonly contain few high-density events, limiting the ability of conventional regression models to represent bloom extremes. This study developed an imbalance-aware classification–regression framework using 38,402 paired water-quality and meteorological observations collected from November 2020 to November 2025 in four lakes on the Yunnan–Guizhou Plateau: oligotrophic Fuxian Lake, mesotrophic Erhai Lake, and eutrophic Dianchi and Xingyun lakes. Random forest, XGBoost, and LightGBM were evaluated as standalone and hybrid models. The hybrid framework distinguished low- and high-density states at 1.0×107 cells L-1 and used regime-specific regressors to predict continuous algal density. SMOTE was restricted to classifier training, and performance was evaluated using the original chronologically held-out observations. SHapley Additive exPlanations (SHAP) were used to characterize model-attributed environmental relationships, while the normalized threshold offset (NTO) standardized selected SHAP zero-crossing positions against lake-specific study-period distributions. The largest improvement occurred in Fuxian Lake, where the SMOTE-assisted random-forest framework increased test R2 from 0.27 to 0.75. SMOTE increased classifier recall from 37.5% to 81.2% and F1 score from 52.2% to 74.3%. The corresponding R2 values remained 0.79, 0.94, and 0.70 for Erhai, Dianchi, and Xingyun lakes, respectively. SHAP identified water temperature as the leading predictor in Fuxian Lake and wind speed in Erhai Lake, whereas conductivity and ammonium nitrogen ranked highly in Dianchi Lake and turbidity and conductivity in Xingyun Lake. NTO placed model-derived transitions on a common standardized scale: nutrient-related transitions in the eutrophic lakes generally occurred closer to their historical means, whereas several transitions in Fuxian and Erhai lakes occurred farther into the upper tails of their distributions. The framework improves high-density prediction under severe class imbalance and provides a standardized basis for comparing model-attributed environmental responses among lakes.
- 14 August 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Bifurcation and Chaos
- Vol. 36 (15)
Sustained Chaotic Oscillations of a Liquid Crystal Elastomer Fiber-Based Collision System
Sustained chaotic motion systems based on active materials have potential applications in energy harvesting, artificial hearts, medical instruments and other fields. This paper proposes a novel light-fueled sustained chaotic oscillation system consisting of a Liquid Crystal Elastomer (LCE) fiber, a mass ball and a rigid substrate. Under periodic illumination, the fiber in the system expands and contracts continuously, as the mass ball collides with the rigid substrate. Ultimately, the system achieves sustained oscillation motion through the fiber contraction work to compensate for the energy dissipation. With the help of the dynamic constitutive model of LCE, the nonlinear dynamic theoretical model of the system is established. The numerical results demonstrate that the fiber can oscillate in two modes, i.e. periodic oscillation and chaotic oscillation, under periodic illumination. The corresponding mechanisms of sustained periodic oscillation and chaotic oscillation of the fiber system are revealed. Moreover, the effect of system parameters on the sustained oscillation behavior is studied in depth, and the effect of parameter variations on the oscillation mode is illustrated by bifurcation diagrams. The research results of this work can deepen the understanding of chaotic motion and provide ideas for chaotic system design and chaotic machine development.
- 14 August 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Bifurcation and Chaos
- Vol. 36 (15)
Sustained Chaotic Oscillations of a Liquid Crystal Elastomer Fiber-Based Collision System
Sustained chaotic motion systems based on active materials have potential applications in energy harvesting, artificial hearts, medical instruments and other fields. This paper proposes a novel light-fueled sustained chaotic oscillation system consisting of a Liquid Crystal Elastomer (LCE) fiber, a mass ball and a rigid substrate. Under periodic illumination, the fiber in the system expands and contracts continuously, as the mass ball collides with the rigid substrate. Ultimately, the system achieves sustained oscillation motion through the fiber contraction work to compensate for the energy dissipation. With the help of the dynamic constitutive model of LCE, the nonlinear dynamic theoretical model of the system is established. The numerical results demonstrate that the fiber can oscillate in two modes, i.e. periodic oscillation and chaotic oscillation, under periodic illumination. The corresponding mechanisms of sustained periodic oscillation and chaotic oscillation of the fiber system are revealed. Moreover, the effect of system parameters on the sustained oscillation behavior is studied in depth, and the effect of parameter variations on the oscillation mode is illustrated by bifurcation diagrams. The research results of this work can deepen the understanding of chaotic motion and provide ideas for chaotic system design and chaotic machine development.
- 10 July 2026
- Published by Springer Nature in ENGINEERING Environment
- Vol. 20 (11) , 164
Climate value of energy storage for floating photovoltaic across China’s decarbonizing power grids
- 10 July 2026
- Published by Springer Nature in ENGINEERING Environment
- Vol. 20 (11) , 164
Climate value of energy storage for floating photovoltaic across China’s decarbonizing power grids
- 25 June 2026
- Published by Springer Nature
- p. 1253-1263
Design of a Small Handling Robot for Non-standard Components
- 25 June 2026
- Published by Springer Nature
- p. 1253-1263
Design of a Small Handling Robot for Non-standard Components
- 12 June 2026
- Published by Elsevier in Journal of Environmental Management
- Vol. 411, 130177
Algal community composition drives lake greenhouse gas emissions via dissolved organic matter transformation and microbial processing
- 12 June 2026
- Published by Elsevier in Journal of Environmental Management
- Vol. 411, 130177
Algal community composition drives lake greenhouse gas emissions via dissolved organic matter transformation and microbial processing
- 27 February 2026
- Published by Springer Nature in Acta Geophysica
- Vol. 74 (2) , 96
A knowledge map-based review of GRACE satellite applications across multiple domains: challenges and future directions
- 27 February 2026
- Published by Springer Nature in Acta Geophysica
- Vol. 74 (2) , 96
A knowledge map-based review of GRACE satellite applications across multiple domains: challenges and future directions