arXiv subjects

Zhen Song

Publications and source records attributed to Zhen Song.

22 records · Page 2Linked to original sources

arXiv2024

Forecasting the Future with Future Technologies: Advancements in Large Meteorological Models

The field of meteorological forecasting has undergone a significant transformation with the integration of large models, especially those employing deep learning techniques. This paper reviews the advancements and applications of these models in weather prediction, emphasizing their role in transforming traditional forecasting methods. Models like FourCastNet, Pangu-Weather, GraphCast, ClimaX, and FengWu have made notable contributions by providing accurate, high-resolution forecasts, surpassing the capabilities of traditional Numerical Weather Prediction (NWP) models. These models utilize advanced neural network architectures, such as Convolutional Neural Networks (CNNs), Graph Neural Networks (GNNs), and Transformers, to process diverse meteorological data, enhancing predictive accuracy across various time scales and spatial resolutions. The paper addresses challenges in this domain, including data acquisition and computational demands, and explores future opportunities for model optimization and hardware advancements. It underscores the integration of artificial intelligence with conventional meteorological techniques, promising improved weather prediction accuracy and a significant contribution to addressing climate-related challenges. This synergy positions large models as pivotal in the evolving landscape of meteorological forecasting.

cs.LG↗

arXiv2024

LightSword: A Customized Virtual Reality Exergame for Long-Term Cognitive Inhibition Training in Older Adults

The decline of cognitive inhibition significantly impacts older adults' quality of life and well-being, making it a vital public health problem in today's aging society. Previous research has demonstrated that Virtual reality (VR) exergames have great potential to enhance cognitive inhibition among older adults. However, existing commercial VR exergames were unsuitable for older adults' long-term cognitive training due to the inappropriate cognitive activation paradigm, unnecessary complexity, and unbefitting difficulty levels. To bridge these gaps, we developed a customized VR cognitive training exergame (LightSword) based on Dual-task and Stroop paradigms for long-term cognitive inhibition training among healthy older adults. Subsequently, we conducted an eight-month longitudinal user study with 12 older adults aged 60 years and above to demonstrate the effectiveness of LightSword in improving cognitive inhibition. After the training, the cognitive inhibition abilities of older adults were significantly enhanced, with benefits persisting for 6 months. This result indicated that LightSword has both short-term and long-term effects in enhancing cognitive inhibition. Furthermore, qualitative feedback revealed that older adults exhibited a positive attitude toward long-term training with LightSword, which enhanced their motivation and compliance.

cs.HC↗

arXiv2024

High-order BDF convolution quadrature for stochastic fractional evolution equations driven by integrated additive noise

The numerical analysis of stochastic time fractional evolution equations presents considerable challenges due to the limited regularity of the model caused by the nonlocal operator and the presence of noise. The existing time-stepping methods exhibit a significantly low order convergence rate. In this work, we introduce a smoothing technique and develop the novel high-order schemes for solving the linear stochastic fractional evolution equations driven by integrated additive noise. Our approach involves regularizing the additive noise through an $m$-fold integral-differential calculus, and discretizing the equation using the $k$-step BDF convolution quadrature. This novel method, which we refer to as the ID$m$-BDF$k$ method, is able to achieve higher-order convergence in solving the stochastic models. Our theoretical analysis reveals that the convergence rate of the ID$2$-BDF2 method is $O(τ^{α+ γ-1/2})$ for $1< α+ γ\leq 5/2$, and $O(τ^{2})$ for $5/2< α+ γ<3$, where $α\in (1, 2)$ and $γ\in (0, 1)$ denote the time fractional order and the order of the integrated noise, respectively. Furthermore, this convergence rate could be improved to $O(τ^{α+ γ-1/2})$ for any $α\in (1, 2)$ and $γ\in (0, 1)$, if we employ the ID$3$-BDF3 method. The argument could be easily extended to the subdiffusion model with $α\in (0, 1)$. Numerical examples are provided to support and complement the theoretical findings.

math.NA↗

arXiv2018

Battery Life-Cycle Optimization and Runtime Control for Commercial Buildings Demand Side Management: A New York City Case Study

In metropolitan areas populated with commercial buildings, electric power supply is stringent especially during business hours. Demand side management using battery is a promising solution to mitigate peak demands, however long payback time creates barriers for large scale adoption. In this paper, we have developed a design phase battery life-cycle cost assessment tool and a runtime controller for the building owners, taking into account the degradation of battery. In the design phase, perfect knowledge on building load profile is assumed to estimate ideal payback time. In runtime, stochastic programming and load predictions are applied to address the uncertainties in loads for producing optimal battery operation. For validation, we have performed numerical experiments using the real-life tariff model serves New York City, Zn/MnO2 battery, and state-of-the-art building simulation tool. Experimental results shows a small gap between design phase assessment and runtime control. To further examine the proposed methods, we have applied the same tariff model and performed numerical experiments on nine weather zones and three types of commercial buildings. On contrary to the common practice of shallow discharging battery for preventing phenomenal degradation, experimental results show promising payback time achieved by optimally deep discharge a battery.

eess.SP↗