检索结果-内蒙古大学图书馆

Methodology for Real-Time Torque Estimation in a Ship Propulsion Digital Twin

JOURNAL OF OFFSHORE MECHANICS AND ARCTIC ENGINEERING-TRANSACTIONS OF THE ASME 2025年第2期147卷 021402页

作者： Purcell, Etienne Nejad, Amir R. Bekker, Anriette Norwegian Univ Sci & Technol Dept Marine Technol N-7050 Trondheim Norway German Aerosp Ctr Inst Maritime Energy Syst Dept Virtual Ship D-21502 Geesthacht Germany Stellenbosch Univ Dept Mech & Mechatron Engn ZA-7600 Stellenbosch South Africa

The safe operation of ships requires the condition of propulsion components to be maintained. Digital twins are a promising alternative for intelligent monitoring of these complex systems. Digital twins require models which ensure that the digital representation is able to mimic the behavior of the physical system. Alternate modeling solutions must be found when intellectual property restrictions or lack of available information limit the usability of physics-based models. This paper considers such a case where a system model of the propulsion system requires a real-time capable model of the propeller hydrodynamic torque. The creation of a data-driven hydrodynamic torque model based on full-scale, operational measurements is discussed. The described method focuses on the significant challenges associated with data cleaning and preparation while also evaluating whether well-known machine learning methods are suited for this application. The methods use speed-over-ground, heading, course, rotational speed, and propeller pitch as inputs. The outputs of the models are the single quadrant propeller torque coefficient and the amplitude of harmonic torsional excitation. These outputs are then combined to create a holistic prediction of the torque. Results indicate that both a polynomial least-squares fit and a shallow neural network predict the mean and the amplitude of harmonic components of the torque well. This prediction can be used to isolate the hydrodynamic torque when more than one torque source is present or to simulate what-if scenarios in a digital twin environment.

关键词： hydrodynamic propeller torque data-driven modeling digital twin full-scale data

来源：评论

学校读者我要写书评

暂无评论

Towards Faster Multi-Objective Surrogate Optimization using SVR: A Casting Case Study

引用

TRANSACTIONS OF THE INDIAN INSTITUTE OF METALS 2024年第12期77卷 4345-4354页

作者： Inapakurthi, Ravi Kiran Mitra, Kishalay Indian Inst Technol Hyderabad Dept Chem Engn Global Optimizat & Knowledge Unearthing Lab Hyderabad 502284 Telangana India Indian Inst Technol Hyderabad Dept Climate Change Hyderabad 502284 Telangana India Indian Inst Technol Hyderabad Dept Artificial Intelligence Hyderabad 502284 Telangana India

Optimization of energy intensive continuous casting process involving multiple objectives is crucial for ensuring sustainability. Use of physics-based models for such tasks is challenging as they are time expensive in evaluation. Alternatively, support vector regression (SVR) is utilized to develop optimally structured data-based models for the afore-mentioned optimization task. Considering the simultaneous minimization of number of samples for training SVR and prediction error of the SVR models, a plethora of candidate surrogate models were obtained through a novel formulation of data-based model development. This involves a sample size estimation algorithm to select the most informative data points from the training set and exploration of multiple kernels while assigning kernel parameter to each input. Model with least Akaike Information Criterion was used to perform surrogate-assisted optimization. Significant temperature reduction (similar to 31.6%) compared to current operating conditions was observed, where surrogate models could save similar to 79.32% computation time without losing much prediction accuracy.

关键词： Energy-intensive industry data-driven modeling Machine learning Process optimization Multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

On the Sample Complexity of Stabilizing Linear Dynamical Systems from data

引用

FOUNDATIONS OF COMPUTATIONAL MATHEMATICS 2024年第3期24卷 955-987页

作者： Werner, Steffen W. R. Peherstorfer, Benjamin NYU Courant Inst Math Sci 251 Mercer St New York NY 10012 USA

Learning controllers from data for stabilizing dynamical systems typically follows a two-step process of first identifying a model and then constructing a controller based on the identified model. However, learning models means identifying generic descriptions of the dynamics of systems, which can require large amounts of data and extracting information that are unnecessary for the specific task of stabilization. The contribution of this work is to show that if a linear dynamical system has dimension (McMillan degree) n, then there always exist n states from which a stabilizing feedback controller can be constructed, independent of the dimension of the representation of the observed states and the number of inputs. By building on previous work, this finding implies that any linear dynamical system can be stabilized from fewer observed states than the minimal number of states required for learning a model of the dynamics. The theoretical findings are demonstrated with numerical experiments that show the stabilization of the flow behind a cylinder from less data than necessary for learning a model.

关键词： Model reduction Dynamical systems Numerical linear algebra data-driven control data-driven modeling Scientific machine learning

来源：评论

学校读者我要写书评

暂无评论

Design optimization and closed-loop operational planning to achieve sustainability goals in buildings

引用

COMPUTERS & CHEMICAL ENGINEERING 2024年 181卷

作者： Risbeck, Michael J. Cyrus, Saman Zhang, Chenlu Lee, Young M. Johnson Controls Int Plc AI Prod Dev Grp Milwaukee WI 53201 USA

Given the significant energy consumption and carbon emissions associated with buildings, there is increasing interest in improving sustainability of building operations to reduce impact on climate change. A common goal is to operate buildings as "net-zero'' energy users in which all energy consumed by the building is balanced against renewable energy purchased from the grid or produced on site. To achieve net-zero status, many buildings will require significant retrofit so as to both reduce energy consumption in the absolute sense and provide the remainder without consuming fossil fuels. Thus, multi-year planning is required to ensure that goals can be met on time. In addition, due to the inherent uncertainty associated with energy consumption and generation, actually achieving net-zero energy use may require discretionary curtailment actions to be taken, and deciding whether such actions are necessary can be challenging. To address these challenges, we propose in this paper a design optimization and operational planning strategy to make the decisions needed to achieve sustainability goals in buildings. The strategy can be applied to schedule design changes over a long horizon to meet annual targets, and it can also be applied in closed loop on a shorter horizon to determine whether curtailment is needed to stay on track. We discuss the formulation of the optimization problem, solution methods, and modeling approaches for key parameters. Application of the strategy is illustrated via examples. Overall, this approach will help automate planning that is often done manually, allowing buildings to take a significant leap forward toward achieving their sustainability goals.

关键词： Building sustainability Design optimization Operational planning data-driven modeling

来源：评论

学校读者我要写书评

暂无评论

data-Light Physics-Informed modeling for the Modulation Optimization of a Dual-Active-Bridge Converter

引用

IEEE TRANSACTIONS ON POWER ELECTRONICS 2024年第7期39卷 8770-8785页

作者： Li, Xinze Lin, Fanfan Zhang, Xin Ma, Hao Blaabjerg, Frede Nanyang Technol Univ Sch Elect & Elect Engn Nanyang 639798 Singapore Zhejiang Univ Univ Illinois Urbana Champaign Inst Hangzhou 310027 Peoples R China Zhejiang Univ Coll Elect Engn Hangzhou 310027 Peoples R China Aalborg Univ Dept Energy Technol DK-9220 Aalborg Denmark

In modulation optimization, power converter modeling is pivotal for performance evaluation. However, mainstream knowledge-based approaches suffer from low accuracy and heavy computation burden, while emerging data-driven methods are data-intensive and opaque black-box models. Even state-of-the-art physics-informed artificial intelligence (AI) is improper for modulation optimization due to resource-intensive retraining for new predictions. Hence, a physics-in-architecture recurrent neural network (PA-RNN), which customizes recurrent neurons to integrate physical laws into the structure, is proposed, tailoring for modulation optimization of power converters. The PA-RNN model reveals diverse circuit insights, exhibiting data-light merit and on-call prediction capability. Modulation optimization via PA-RNN involves two stages. First, PA-RNN constructs converter models in the time domain for direct performance evaluation. Second, an optimization algorithm interacts with the PA-RNN model to minimize current stress while realizing full-range soft switching. Two design cases are presented: first, modeling buck converters;second, optimizing dual-active-bridge converters under a triple phase-shift modulation or a five-degree-of-freedom modulation. Algorithm experiments and 1-kW hardware experiments have comprehensively validated the merits and feasibility of the proposed PA-RNN. Broadly speaking, this article strives to increase the penetration of AI in power electronics.

关键词： Integrated circuit modeling Modulation Optimization Predictive models Power electronics Phase modulation Artificial intelligence Artificial intelligence (AI) data-driven modeling dual-active-bridge (DAB) converter modulation physics-in-architecture physics-informed AI soft switching triple phase shift

来源：评论

学校读者我要写书评

暂无评论

Weak-form latent space dynamics identification

引用

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING 2024年 427卷

作者： Tran, April He, Xiaolong Messenger, Daniel A. Choi, Youngsoo Bortz, David M. Univ Colorado Dept Appl Math Boulder CO 80309 USA ANSYS Inc Livermore CA 94551 USA Lawrence Livermore Natl Lab Ctr Appl Sci Comp Livermore CA 94550 USA

Recent work in data -driven modeling has demonstrated that a weak formulation of model equations enhances the noise robustness of a wide range of computational methods. In this paper, we demonstrate the power of the weak form to enhance the LaSDI (Latent Space Dynamics Identification) algorithm, a recently developed data -driven reduced order modeling technique. We introduce a weak form -based version WLaSDI (Weak -form Latent Space Dynamics Identification). WLaSDI first compresses data, then projects onto the test functions and learns the local latent space models. Notably, WLaSDI demonstrates significantly enhanced robustness to noise. With WLaSDI, the local latent space is obtained using weak -form equation learning techniques. Compared to the standard sparse identification of nonlinear dynamics (SINDy) used in LaSDI, the variance reduction of the weak form guarantees a robust and precise latent space recovery, hence allowing for a fast, robust, and accurate simulation. We demonstrate the efficacy of WLaSDI vs. LaSDI on several common benchmark examples including viscid and inviscid Burgers', radial advection, and heat conduction. For instance, in the case of 1D inviscid Burgers' simulations with the addition of up to 100% Gaussian white noise, the relative error remains consistently below 6% for WLaSDI, while it can exceed 10,000% for LaSDI. Similarly, for radial advection simulations, the relative errors stay below 16% for WLaSDI, in stark contrast to the potential errors of up to 10,000% with LaSDI. Moreover, speedups of several orders of magnitude can be obtained with WLaSDI. For example applying WLaSDI to 1D Burgers' yields a 140X speedup compared to the corresponding full order model. Python code to reproduce the results in this work is available at (https://***/ MathBioCU/PyWSINDy_ODE) and (https://***/MathBioCU/PyWLaSDI).

关键词： data-driven modeling Weak form Equation learning Reduce order modeling Latent space learning

来源：评论

学校读者我要写书评

暂无评论

Online Optimization and Ambiguity-Based Learning of Distributionally Uncertain Dynamic Systems

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2024年第9期69卷 6153-6168页

作者： Li, Dan Fooladivanda, Dariush Martinez, Sonia Univ Calif San Diego Dept Mech & Aerosp Engn La Jolla CA 92092 USA Univ Calif Berkeley Dept Elect Engn & Comp Sci Berkeley CA 94720 USA

This article proposes a novel approach to construct data-driven online solutions to optimization problems (P) subject to a class of distributionally uncertain dynamical systems. The introduced framework allows for the simultaneous learning of distributional system uncertainty via a parameterized, control-dependent ambiguity set using a finite historical dataset, and its use to make online decisions with probabilistic regret function bounds. Leveraging the merits of machine learning, the main technical approach relies on the theory of distributional robust optimization (DRO), to hedge against uncertainty and provide less conservative results than standard robust optimization approaches. Starting from recent results that describe ambiguity sets via parameterized, and control-dependent empirical distributions as well as ambiguity radii, we first present a tractable reformulation of the corresponding optimization problem while maintaining the probabilistic guarantees. We then specialize these problems to the cases of 1) optimal one-stage control of distributionally uncertain nonlinear systems, and 2) resource allocation under distributional uncertainty. A novelty of this work is that it extends DRO to online optimization problems subject to a distributionally uncertain dynamical system constraint, handled via a control-dependent ambiguity set that leads to online-tractable optimization with probabilistic guarantees on regret bounds. Further, we introduce an online version of the Nesterov's accelerated-gradient algorithm, and analyze its performance to solve this class of problems via the dissipativity theory.

关键词： Optimization Uncertainty Dynamical systems Probabilistic logic Heuristic algorithms Vectors data models Adaptive learning data-driven modeling optimization under distributional uncertainties

来源：评论

学校读者我要写书评

暂无评论

TossNet: Learning to Accurately Measure and Predict Robot Throwing of Arbitrary Objects in Real Time With Proprioceptive Sensing

引用

IEEE TRANSACTIONS ON ROBOTICS 2024年 40卷 3232-3251页

作者： Chen, Lipeng Lu, Weifeng Zhang, Kun Zhang, Yizheng Zhao, Longfei Zheng, Yu Tencent Robot X Shenzhen 518057 Peoples R China City Univ Hong Kong Dept Biomed Engn Hong Kong Peoples R China Hong Kong Univ Sci & Technol Dept Elect & Comp Engn Hong Kong Peoples R China

Accurate measuring and modeling of dynamic robot manipulation (e.g., tossing and catching) is particularly challenging, due to the inherent nonlinearity, complexity, and uncertainty in high-speed robot motions and highly dynamic robot-object interactions happening in very short distances and times. Most studies leverage extrinsic sensors such as visual and tactile feedback toward task or object-centric modeling of manipulation dynamics, which, however, may hit bottleneck due to the significant cost and complexity, e.g., the environmental restrictions. In this work, we investigate whether using solely the on-board proprioceptive sensory modalities can effectively capture and characterize dynamic manipulation processes. In particular, we present an object-agnostic strategy to learn the robot toss dynamics of arbitrary unknown objects from the spatio-temporal variations of robot toss movements and wrist-force/torque (F/T) observations. We then propose TossNet, an end-to-end formulation that jointly measures the robot toss dynamics and predicts the resulting flying trajectories of the tossed objects. Experimental results in both simulation and real-world scenarios demonstrate that our methods can accurately model the robot toss dynamics of both seen and unseen objects, and predict their flying trajectories with superior prediction accuracy in nearly real-time. Ablative results are also presented to demonstrate the effectiveness of each proprioceptive modality and their correlations in modeling the toss dynamics. Case studies show that TossNet can be applied on various real robot platforms for challenging tossing-centric robot applications, such as blind juggling and high-precise robot pitching.

关键词： Robots Robot sensing systems Trajectory Dynamics Manipulator dynamics Propioception Accuracy data-driven modeling dynamic manipulation proprioceptive sensing trajectory prediction

来源：评论

学校读者我要写书评

暂无评论

Physics-Informed Active Learning With Simultaneous Weak-Form Latent Space Dynamics Identification

引用

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING 2025年第1期126卷

作者： He, Xiaolong Tran, April Bortz, David M. Choi, Youngsoo ANSYS Inc Livermore CA 94551 USA Univ Colorado Dept Appl Math Boulder CO USA Lawrence Livermore Natl Lab Ctr Appl Sci Comp Livermore CA USA

The parametric greedy latent space dynamics identification (gLaSDI) framework has demonstrated promising potential for accurate and efficient modeling of high-dimensional nonlinear physical systems. However, it remains challenging to handle noisy data. To enhance robustness against noise, we incorporate the weak-form estimation of nonlinear dynamics (WENDy) into gLaSDI. In the proposed weak-form gLaSDI (WgLaSDI) framework, an autoencoder and WENDy are trained simultaneously to discover intrinsic nonlinear latent-space dynamics of high-dimensional data. Compared with the standard sparse identification of nonlinear dynamics (SINDy) employed in gLaSDI, WENDy enables variance reduction and robust latent space discovery, therefore leading to more accurate and efficient reduced-order modeling. Furthermore, the greedy physics-informed active learning in WgLaSDI enables adaptive sampling of optimal training data on the fly for enhanced modeling accuracy. The effectiveness of the proposed framework is demonstrated by modeling various nonlinear dynamical problems, including viscous and inviscid Burgers' equations, time-dependent radial advection, and the Vlasov equation for plasma physics. With data that contains 5%-10%$$ \% $$ Gaussian white noise, WgLaSDI outperforms gLaSDI by orders of magnitude, achieving 1%-7%$$ \% $$ relative errors. Compared with the high-fidelity models, WgLaSDI achieves 121 to 1779x$$ \times $$ speed-up.

关键词： autoencoder data-driven modeling latent space dynamics learning physics-informed active learning reduced-order modeling weak form

来源：评论

学校读者我要写书评

暂无评论

Neural transition system abstraction for neural network dynamical system models and its application to Computational Tree Logic verification

引用

NEURAL NETWORKS 2025年 186卷 107261页

作者： Yang, Yejiang Wang, Tao Xiang, Weiming Augusta Univ Sch Comp & Cyber Sci Augusta GA 30912 USA Southwest Jiaotong Univ Sch Elect Engn Chengdu Peoples R China

This paper proposes an explainable abstraction-based verification method that prioritizes user interaction and enhances interpretability. By partitioning the system's state space using a data-driven process, we can abstract the dynamics into words consisting of state labels. When given a trained neural network model, a set- valued reachability analysis method is introduced to estimate the relationship between each subsystem. We construct the neural transition system abstraction with the neural network model and the relationships between partitions. Then, the abstracted model can be verified through Computational Tree Logic (CTL), enabling formal verification of the system's behavior. This approach greatly enhances the interpretability and verification of data-driven models, as well as the ability to validate against the specification. Finally, examples of the Maglev model and handwritten model abstractions are given to illustrate our proposed model verification framework, which demonstrates that the proposed framework has advantages in enhancing model interpretability and verifying user-specified properties based on CTL.

关键词： Model verification Transition system data-driven modeling Reachability analysis CTL specification Model abstraction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：