Shengjun Ju

Shengjun Ju

Auto-generated profile, based on publicly available information

Shengjun Ju

Chinese Academy of Sciences Institute of Mechanics, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2023 in Physics of Fluids
Verified
Journal Article
Closed access
PDF file
CYChang Yan
Chang Yan
SXShengfeng Xu
Shengfeng Xu
Zhenxu SunZhenxu Sun
Zhenxu Sun
DGDilong Guo
Dilong Guo
SJShengjun Ju
Shengjun Ju
GYGuowei Yang
Guowei Yang
Proper orthogonal decomposition (POD) enables complex flow fields to be decomposed into linear modes according to their energy, allowing the key features of the flow to be extracted. However, traditional POD requires high-quality inputs, namely, high-resolution spatiotemporal data. To alleviate the dependence of traditional POD on the quality and quantity of data, this paper presents a POD method that is strengthened by a physics-informed neural network (PINN) with an overlapping domain decomposition strategy. The loss function and convergence of modes are considered simultaneously to determine the convergence of the PINN-POD model. The proposed framework is applied to the flow past a two-dimensional circular cylinder at Reynolds numbers ranging from 100 to 10 000 and achieves accurate and robust extraction of flow structures from spatially sparse observation data. The spatial structures and dominant frequency can also be extracted under high-level noise. These results demonstrate that the proposed PINN-POD method is a reliable tool for extracting the key features from sparse observation data of flow fields, potentially shedding light on the data-driven discovery of hidden fluid dynamics.

1 December 2023 in Physics of Fluids
Verified
Journal Article
Open access
PDF file
YWYeteng Wang
Yeteng Wang
Zhenxu SunZhenxu Sun
Zhenxu Sun
SJShengjun Ju
Shengjun Ju
DGDilong Guo
Dilong Guo
BYBo Yin
Bo Yin
GYGuowei Yang
Guowei Yang
Micro-riblet is an efficient passive method for controlling turbulent boundary layers, with the potential to reduce frictional drag. In various applications within the transportation industry, flow separation is a prevalent flow phenomenon. However, the precise drag reduction performance of riblets in the presence of flow separation remains unclear. To address this, an inclined forward step model is proposed to investigate the interaction between riblet and upstream flow separation. The large eddy simulation (LES) method is applied to simulate the flow over geometries with different step angles and riblet positions. The results show riblets still reduce wall frictional resistance when subjected to the upstream flow separation. Remarkably, as the angle of the step increases from 0° to 30°, the drag reduction experiences an increment from 9.5% to 12.6%. From a turbulence statistics standpoint, riblets act to suppress the Reynold stress in the near-wall region and dampen ejection motions, thus weakening momentum exchange. Quadrant analysis reveals that with the augmentation of flow separation, the Q2 motion within the flow field intensifies, subsequently enhancing the riblet-induced drag reduction. Moreover, the position of the riblets has a significant impact on the pressure drag. Riblets close to the point of separation enhance flow separation, altering the surface pressure distribution and thus increasing the resistance. The results reveal that when the riblets are positioned approximately 160 riblet heights away from the step, their effect on the upstream flow separation becomes negligible. The precise performance of riblets under complex flow conditions is important for their practical engineering application.

1 July 2025 in Physics of Fluids
Verified
Journal Article
Closed access
YWYeteng Wang
Yeteng Wang
Zhenxu SunZhenxu Sun
Zhenxu Sun
DGDilong Guo
Dilong Guo
SJShengjun Ju
Shengjun Ju
GYGuowei Yang
Guowei Yang
Micro riblet surfaces have demonstrated significant potential for reducing skin-friction drag in turbulent boundary layers, making them attractive candidates for aerodynamic optimization in high-speed transportation. To address the high computational cost arising from the large disparity between riblet scale and engineering scale in numerical simulations, modeling riblet effects in Reynolds-averaged Navier–Stokes (RANS) computations has become an effective strategy. However, existing methods often suffer from excessive mesh sensitivity and overprediction of drag increases for oversized riblets, limiting their engineering applicability. The present study proposes an improved RANS-based modeling approach for simulating riblet effects. The core of the model involves a correction to the boundary condition of the specific dissipation rate (ω), extending a formulation previously introduced by other researchers. The modeling method consists of three key components. First, an empirical relation is established between riblet geometry and the riblet-induced velocity profile shift in the logarithmic region, accounting for variations in the aspect ratio of V-shaped riblets and accurately capturing both drag-reducing and drag-increasing regimes. Second, a boundary correction for ω is developed through numerical experiments to reproduce the velocity profile shift, incorporating a mesh-sensitivity analysis and introducing a correction factor related to grid resolution. Third, a pressure-gradient correction factor is introduced to simulate the influence of adverse pressure gradients on riblet performance. The proposed method is implemented in the OpenFOAM solver and validated through simulations of flat-plate boundary layers and a wind turbine airfoil, demonstrating good agreement with experimental results.

1 April 2026 in AIP Advances
Verified
Journal Article
Open access
BPBoyu Pan
Boyu Pan
DGDilong Guo
Dilong Guo
SJShengjun Ju
Shengjun Ju
Zhenxu SunZhenxu Sun
Zhenxu Sun
JWJunxiang Wang
Junxiang Wang
GYGuowei Yang
Guowei Yang
Deep Reinforcement Learning (DRL) is a powerful paradigm for discovering non-linear control strategies in complex fluid dynamics. However, its application to high-fidelity simulations is often hampered by prohibitive sample inefficiency and the risks of “cold-start” exploration. To overcome these limitations, an Expert-Guided Soft Actor-Critic (EG-SAC) framework is proposed, which synergistically integrates prior knowledge from a classical, particle swarm optimization-optimized proportional-integral-derivative controller into the DRL agent. The framework employs a two-stage learning process, beginning with an offline phase where behavioral cloning initializes the policy and expert demonstrations pre-fill the replay buffer and an online fine-tuning stage where a composite loss function, featuring a decaying expert-regularization term, provides continuous guidance to the agent. This framework is applied to active flow control for a circular cylinder at a Reynolds number (Re) of 100. The results demonstrate a threefold advantage: first, the EG-SAC agent achieves immediate operational safety, exhibiting near-expert performance from the first epoch, while the standard SAC agent initially suffers from detrimental, high-drag explorations; second, the sample efficiency is significantly enhanced, with asymptotic convergence accelerated by ∼19%; finally, the converged EG-SAC policy achieves a drag reduction of 7.93%, surpassing the 5.21% achieved by the best linear controller and demonstrating the ability to discover superior non-linear control laws. The superior performance stems from an enhanced base pressure recovery via a further elongated recirculation bubble. This work presents a robust and efficient methodology for applying DRL to fluid mechanics, bridging the gap between the stability of classical control and the high-performance adaptability of data-driven methods.