Speaker: Prof. Bin Pei, Northwestern Polytechnical University, China
Date & Time: Monday, 27th April, 10:00 - 10:40
Location: IRZ-Library
Title: Analytical Responses of Rotor-Nacelle Systems Subjected to Non-white Aerodynamic Moments
Abstract
Tilt-rotor aircraft may be subjected to long-correlated random loads in its complex flight environments, potentially inducing dynamics distinct from those based on deterministic force or idealized white noise assumptions in existing studies. In this paper, we will explore the whirl flutter of tilt-rotor aircraft subjected to non-white random aerodynamic moments, modeled as two-degree-of-freedom nonlinear rotor-nacelle systems with fractional Gaussian noise. Firstly, a harmonic balance method is employed to obtain the amplitude response characteristics of the deterministic rotor-nacelle systems. Subsequently, a dimensionality reduction approach based on diffusion approximation theory is proposed for the analysis of stochastic rotor-nacelle systems. To achieve a precise and efficient calculation of system responses, a memory-dependent Fokker-Planck-Kolmogorov equation method is further introduced. Finally, numerical simulations are conducted to verify the effectiveness of the proposed solution methodology. In addition, the statistical characteristics of the responses of stochastic rotor-nacelle systems are examined in detail, and the influence of system parameters on the responses is systematically studied. We discover that the second-order central moments of the system responses increase with the decreasing of the Hurst index of the FGN, while increase with the increasing of the freestream-to-blade-tip velocity ratio, the rotor angular velocity, and the noise intensity. The results reveal that both structural and noise parameters have significant impacts on the dynamic stability of the systems, further threatening the safety of tilt-rotor aircraft.
Biography:
He is a full-time professor and doctoral supervisor at Northwestern Polytechnical University, a Humboldt Research Fellow, and a JSPS Fellow. His research focuses on applied probability and statistics, stochastic and nonlinear dynamical systems, non-Markovian dynamics, and stochastic control, with applications in aerospace engineering. He has led multiple national and provincial research projects, including three grants from the National Natural Science Foundation of China. He has published over 20 high-quality SCI papers in recent years, including one ESI Highly Cited Paper. His honors include the Humboldt Fellowship, JSPS Fellowship, and several provincial and national awards. He also serves as Executive Director of the Shaanxi Society of Vibration Engineering and Young Editorial Board Member of related journals.
Speaker: Dr. Dan Zhao, Humboldt University of Berlin, Germany
Date & Time: Monday, 27th April, 10:40 - 11:20
Location: IRZ-Library
Title: Generator-Wise Vulnerability and Disturbance Propagation in the Italian Power Grid under Intermittent Renewable Fluctuations
Abstract:
This talk examines how localized intermittent renewable-like fluctuations affect frequency synchronization in the Italian transmission grid. By injecting a wind-like signal into one generator at a time, we construct generator-wise vulnerability maps and identify the most critical locations for disturbance amplification. The results show that the system response depends more strongly on where fluctuations are injected than on their intensity over the parameter range considered. Near critical coupling, the dynamics also become strongly history dependent because of multi-stability and non the intendiy are the hysteresis. Overall, the study highlights the key role of spatial heterogeneity in renewable-driven stability risks.
Biography:
Dan Zhao is a postdoctoral researcher at the Potsdam Institute for Climate Impact Research (PIK) and Humboldt University of Berlin, Germany, supported by the CSC-DAAD Postdoctoral Scholarship. She received her Ph.D. in Mathematics from Northwestern Polytechnical University, China. Her research interests include stochastic dynamical systems, higher-order interactions, Lévy noise, and extreme events in complex networks. She has authored and co-authored 7 papers, including publications in Communications Physics, Communications in Nonlinear Science and Numerical Simulation, Chaos, and Science China Technological Sciences.
Speaker: Prof Yong Xu, Northwestern Polytechnical University, China
Date & Time: Monday, 27th April, 11:20 - 12:00
Location: IRZ-Library
Title: Efficient Solution of the Fokker-Planck Equation via Deep Learning
Abstract:
The Fokker-Planck equation (FPE) governs the evolution of probability densities in stochastic systems, and its efficient solution is crucial for stochastic analysis. However, traditional numerical methods suffer from the curse of dimensionality, making them intractable for high-dimensional systems. Moreover, they typically require separate solutions for each set of parameters, hindering a comprehensive exploration of high-dimensional parameter spaces. To address these challenges, we develop a series of numerical methods that integrate stochastic dynamics theory with deep learning techniques. First, the DL-FP method is proposed, in which the stationary solution of the FPE is represented by a neural network. By transforming the solution process into a learning problem, physical and probabilistic constraints can be effectively enforced. This method is further extended to handle fractional-order and high-dimensional systems. Second, a pseudo-analytical probabilistic solution (PAPS) is introduced by constructing a deep learning mapping from system parameters to their corresponding stationary distributions. This enables the parallel solution of FPEs across the continuous parameter space. Together, these methods provide a powerful tool for the dynamical analysis of multi-parameter, high-dimensional stochastic systems.
Biography:
Yong Xu is a professor at Northwestern Polytechnical University, PR China. He got the experienced Alexander von Humboldt Research Fellow (Germany) in 2016. He research interests include stochastic and nonlinear dynamics, aircraft structural integrity, and applied probability and statistics. He has led over 20 provincial-and national-level research projects. As the lead contributor, he has received the First Prize of the Shaanxi Province Science and Technology Award, the Second Prize of Natural Science Award of Ministry of Education et al. Y. Xu is the first or corresponding author of more than 200 publications in journals and conference proceedings. He currently serves as Director of the Key Laboratory of Complexity Science in Aerospace of Ministry of Education. He also serves as an associate editor or editorial board member for several international journals, including Nonlinear Dynamics, Chaos, and Probabilistic Engineering Mechanics.
If you would like to participate online via Webex, please contact Mengze Lyu before the presentation.
