检索结果-内蒙古大学图书馆

Wind Speed Forecasting Using a Two-Stage Forecasting System With an Error Correcting and Nonlinear Ensemble Strategy

引用

IEEE ACCESS 2019年 7卷 176000-176023页

作者： Zhang, Lifang Dong, Yao Wang, Jianzhou Dongbei Univ Finance & Econ Sch Stat Dalian 116025 Peoples R China Jiangxi Univ Finance & Econ Sch Stat Nanchang 330013 Jiangxi Peoples R China Jiangxi Univ Finance & Econ Appl Stat Res Ctr Nanchang 330013 Jiangxi Peoples R China

Precise wind speed prediction is increasingly practical for sustained and stable wind energy utilization considering the growing portion of wind energy in the global electric grid. Although plenty of wind speed forecasting approaches have been devoted to improving forecasting performance, the majority have neglected the utilization of error information, integrated the forecasting value of every component with simple ensemble approaches, and ignored forecasting stability, which may make the forecasting results poor. Considering the above drawbacks, a two-stage forecasting system is performed in our study based on the data preprocessing approach, improved multi-objective optimization algorithm, error correction and nonlinear ensemble strategy. The developed system effectively overcomes the shortcomings of previous models and observably prompts wind speed forecasting capacity. For investigating the prediction capacity of the developed system, six wind speed series obtained from China and Spain are applied as case sites. The forecasting consequences indicate that the developed system is more conducive to enhancing forecasting precision and stability than other involved models, which can provide beneficial assistance for wind speed forecasting.

关键词： Wind speed forecasting multi-objective optimization algorithm ensemble strategy forecasting accuracy

来源：评论

学校读者我要写书评

暂无评论

Scalable approaches to Selecting Key Entities in Large Networked Infrastructure Systems

Scalable Approaches to Selecting Key Entities in Large Netwo...

引用

IEEE International Conference on Big Data

作者： Arun V Sathanur Arif Khan Pacific Northwest National Laboratory Seattle WA USA Pacific Northwest National Laboratory Richland WA US

ISBN: (纸本)9781665445993

This work aims at bringing advances in discrete optimization algorithms to solving practical engineering problems at scale. Often times, in many engineering design problems, there is a need to select a small set of influential or representative elements from a large ground set of entities in an optimal fashion. Submodular optimization provides for a formal way to solve such problems. Common examples with infrastructure systems involve sensor placement and identification of key entities with certain objectives. However, scaling these approaches to large infrastructure systems can be challenging because of the high computational complexity of the overall framework that include the optimization algorithms as well as high-complexity compute-oracles that provide the necessary objective function values. In this work, we explore a well-studied and widely-applicable paradigm, namely leader-selection in a multi-agent networked setting in the context of scalable methodologies. We demonstrate novel frameworks that utilize variations of accelerated submodular optimization algorithms along with linear-algebraic methods that can help accelerate the oracle computations. We further explore this combination in conjunction with graph partitioning paradigms to take advantage of the accelerated algorithms in a distributed setting. Finally we demonstrate the key findings on a practical problem in an operational setting. For this, we leverage an example road network with approximately 18k nodes and 27k edges in a traffic control application, where we seek a limited number of k = 200 key intersections. This problem can be solved in a serial setting in just under 5 hours providing more than 2 orders of magnitude speed-up over methods that do not consider acceleration techniques.

关键词： Sensor placement Roads Conferences Big Data Traffic control Linear programming Partitioning algorithms

来源：评论

学校读者我要写书评

暂无评论

A filled function which has the same local minimizer of the objective function

引用

optimization LETTERS 2019年第4期13卷 761-776页

作者： Lin, Hongwei Gao, Yuelin Wang, Xiaoli Su, Shoubao Jinling Inst Technol Fac Sci Nanjing 211169 Jiangsu Peoples R China Beifang Univ Nationalities Informat & Syst Sci Yinchuan 750021 Peoples R China Xidian Univ Sch Comp Sci & Technol Xian 710071 Shaanxi Peoples R China Jinling Inst Technol Sch Comp Engn Nanjing 211169 Jiangsu Peoples R China

Auxiliary function methods have been considered to be practical approaches for finding the global minimizer of multi-model *** function methods, as a typical representative of auxiliary function methods, obtain a global minimizer by minimizing the objective function and the filled function cyclically. In order to improve the efficiency of the filled function, this paper presents a new filled function which has the same local minimizers of the objective function, and these minimizers are all better than the current minimizer of the objective function. Therefore, it does not need to minimize the objective function except for the first iteration in the filled function method. Additionally, the proposed filled function excludes some disadvantages of conventional filled functions and a classical local optimization method can be applied directly to the new filled function to obtain a better minimizer of the original problem. Finally, numerical experiments are made and the results show the effectiveness of the proposed method.

关键词： Global optimization Filled function method Global minimizer Local minimizer

来源：评论

学校读者我要写书评

暂无评论

Face Alignment via multi-Regressors Collaborative optimization

引用

IEEE ACCESS 2019年 7卷 4101-4112页

作者： Liu, Caifeng Feng, Lin Wang, Huibing Wu, Bin Dalian Univ Technol Fac Elect Informat & Elect Engn Dalian 116024 Peoples R China Dalian Univ Technol Sch Innovat & Entrepreneurship Dalian 116024 Peoples R China Dalian Maritime Univ Coll Informat Sci & Technol Dalian 116024 Peoples R China Neusoft Dept Prod Platform Dalian 116024 Peoples R China

Face alignment is a fundamental step in facial image analysis. To solve the non-convex optimization problem, most cascade-based regression approaches conventionally utilize a single regressor to cover the entire optimization space. These measures are prone to average conflicting gradient directions, especially when applied to faces in the unconstrained condition with various poses and expressions. In this paper, we present an effective face alignment approach based on the multi-regressors collaborative optimization. The foundation of our method is the cascaded regression (CR) that has recently established itself as one of the most practical and effective frameworks for localizing the facial landmarks. CR is interpreted as a learning-based approach to iteratively optimize an objective function. On this basis, in all iterations, the proposed algorithm further divides the sample space into several clusters, in each of which, samples with similar gradient directions and one separate local regressor are learned. During the prediction stage, the unseen landmarks of a face image are evaluated by a linear combination of estimations from all cluster regressors with different weights. The experimental results demonstrate the advantages of our method on the general unconstrained images.

关键词： Face alignment multi-regressors cascaded shape regression

来源：评论

学校读者我要写书评

暂无评论

A hybrid constrained multi-objective algorithm for dynamic economic emission dispatch

引用

International Journal of Electrical Power and Energy Systems 2025年 169卷

作者： Qian, Shuqu He, Huihui Wu, Huihong Fournier-Viger, Philippe Li, Hui Huang, Shoude Anshun University Anshun China No. 1 Middle School of Kaiyang County Kaiyang China Shenzhen University Shenzhen China

Dynamic Economic Emission Dispatch (DEED) problems present a critical multi-objective optimization challenge in power systems, requiring simultaneous minimization of conflicting objectives of costs and emissions while addressing complex operational constraints. This problem becomes particularly intricate when considering the Valve-Point Effect Cost (VPEC) in generation units, which exhibits non-smooth, non-convex characteristics that significantly complicate solution landscapes. To address these challenges, this paper proposes a Hybrid Constrained multi-objective Algorithm (HCMOA) incorporating three innovative mechanisms. First, a clonal selection mechanism enhances local exploitation through systematic generation of high-quality non-dominated solutions. Second, a most-crowded neighborhood update strategy combats premature convergence in high-dimensional DEED scenarios while optimizing archive diversity and computational efficiency. Third, a step-by-step repair strategy ensures operational feasibility through systematic output refinement. Comprehensive evaluations on 10-unit and 15-unit systems with varying VPEC configurations demonstrate HCMOA's superiority over ten state-of-the-art algorithms. Experimental results reveal that HCMOA achieves the Pareto-optimal set with superior spread and distribution uniformity compared to existing approaches. These findings collectively establish HCMOA as an effective solution for practical DEED implementations, particularly in scenarios requiring rigorous consideration of non-convex VPEC characteristics. © 2025 The Authors

关键词： multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Modeling, optimization, and thermal management strategies of hydrogen fuel cell systems

引用

Results in Engineering 2025年

作者： Abubakar Unguwanrimi Yakubu Liu Qingsheng Meng Kai Chen Jinwei Omer Abbaker Ahmed Mohammed Jiahao Zhao Qi Jiang Xuanhong Ye Junyi Liu Qinglong Yu Muhammad Aurangzeb Shusheng Xiong College of Energy Engineering Zhejiang University Hangzhou 310027 China College of Agriculture and Environmental Science Kaduna State University Kaduna 800283 Nigeria Longquan Industrial Innovation Research Institute Longquan 323700 China Provincial Key Laboratory of New Energy Vehicles Thermal Management Longquan 323700 China Jiashan Power Supply Company State Grid Zhejiang Electric Power Co. LTD Department of Electrical and Electronics Engineering Faculty of Engineering Sciences University of Nyala Nyala Sudan

The transition to sustainable energy has intensified the focus on Proton Exchange Membrane Fuel Cells (PEMFCs) due to their high efficiency, zero emissions, and flexibility for both mobile and stationary uses. However, widespread adoption is limited by technical challenges, including efficiency losses from activation and mass transport overpotentials, thermal instability under transient loads, and high system and hydrogen production costs. Traditional static modeling approaches struggle to capture the real-time, coupled dynamics of PEMFCs in practical scenarios. This review synthesizes over one hundred studies published between 2015 and 2025, highlighting advancements in steady-state and dynamic modeling, AI-driven multi-objective optimization, and integrated thermal management. optimization algorithms such as PSO, WOA, MIGA, and NSGA-II have shown promising results, including up to 15% reduction in hydrogen consumption and 20 to 30% improvement in thermal uniformity. The review also explores hybrid physical-AI models, CFD-based surrogate models, and predictive machine-learning methods like LSTM and CNN. Emphasis is placed on AI-enhanced energy management systems (EMS) capable of real-time control by integrating stress, degradation, and load conditions. Economic modeling for green hydrogen production is also included. The paper concludes by offering a unified framework and identifying current limitations, paving the way for scalable, intelligent PEMFC deployment in real-world energy systems.

关键词： Hydrogen fuel cells multi-objective optimization thermal management energy management systems AI and ML in fuel cells

来源：评论

学校读者我要写书评

暂无评论

A differentiable structural analysis framework for high-performance design optimization

引用

STRUCTURES 2025年 78卷

作者： Lee, Keith J. Huang, Yijiang Mueller, Caitlin T. MIT Cambridge MA 02139 USA Swiss Fed Inst Technol Zurich Switzerland

Fast, gradient-based structural optimization has long been limited to a highly restricted subset of problems- namely, density-based compliance minimization-for which gradients can be analytically derived. For other objective functions, constraints, and design parameterizations, computing gradients has remained inaccessible, requiring the use of derivative-free algorithms that scale poorly with problem size. This has restricted the applicability of optimization to abstract and academic problems, and has limited the adoption of these potentially impactful methods in practice. In this paper, we bridge the gap between computational efficiency and the freedom of problem formulation through a differentiable analysis framework designed for general structural optimization. We achieve this by leveraging Automatic Differentiation (AD) to manage the complex computational graph of structural analysis programs, and implementing specific derivation rules for performance critical functions along this graph. This paper provides a complete overview of gradient computation for arbitrary structural design objectives, identifies the barriers to their practical use, and derives key intermediate derivative operations that resolve these bottlenecks. Our framework is then tested against a series of structural design problems of increasing complexity: two highly constrained minimum volume problems, a multi-stage shape and section design problem, and an embodied carbon minimization problem. We benchmark our framework against other common optimization approaches, and show that our method outperforms others in terms of speed, stability, and solution quality.

关键词： Automatic differentiation Structural analysis Structural optimization Computational design

来源：评论

学校读者我要写书评

暂无评论

Changeability and flexibility of assembly line balancing as a multi-objective optimization problem

引用

JOURNAL OF MANUFACTURING SYSTEMS 2019年 53卷 150-158页

作者： Fisel, Johannes Exner, Yannick Stricker, Nicole Lanza, Gisela Wbk Inst Prod Sci Kaiserstr 12 D-76131 Karlsruhe Germany

Current trends, such as customers' demand for individual products and shorter product life cycles, are addressed by companies through a greater variety of products and variants. With regard to the line balancing of flow assembly systems, however, adjustments are associated with high investments, which requires a new planning approach for assembly line balancing. Existing approaches do not consider the reallocation of assembly tasks or the dimensioning of system-inherent flexibility and changeability according to requirements. Furthermore, they neglect the uncertainty of the future market situation. The proposed approach aims at optimizing the line balancing of flow assembly systems, taking into account the potential need for adaptation in order to meet this uncertain planning environment. For this purpose, the exchange of occurring costs as well as flexibility and changeability of the system is focused. Based on scenarios, potential future compositions of the variant mix are investigated and the resulting implications for the assembly system are derived. By applying the approach, an adequate adaptable assembly line balancing is generated by performing a mixed integer linear optimization. Since the evaluation and identification of adequacy are subject to subjective factors, several potentially adequate solutions are generated, which differ in terms of costs, flexibility and changeability. The result of the presented approach is a front of pareto-optimal assembly line balancing configurations. In order to show its practical applicability, a use case in automotive assembly line balancing is presented.

关键词： Assembly line balancing Changeability Flexibility multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

Research on the multi-objective optimization of China's Financial Industrial Structure Based on the Evolutionary Algorithm 4

Research on the Multi-objective Optimization of China's Fina...

引用

4th IEEE International Conference on Big Data Analytics (ICBDA)

作者： Gao, Yang Li, Yunhai Pan, Weiyi Wang, Yaojun Beijing Univ Technol Sch Econ & Management Beijing Peoples R China Chinese Acad Sci Inst Comp Technol Beijing Peoples R China Peking Univ Guanghua Sch Management Beijing Peoples R China

ISBN: (纸本)9781728112824

The optimization of the financial industrial structure (FIS) is one of the critical approaches to promote economic development under the new normal. Using the panel data of 30 provinces and cities except for Tibet from 2007 to 2016, this study analyzes the optimization objective of the FIS and establishes an optimization model of the FIS based on the economic development multi-objectives. Then several multi-objective evolutionary optimization algorithms are applied to solve the optimal FIS, and a possible research direction for the optimization of the multi-objective financial structure is provided. The result is of great significance for the theoretical and practical studies of financial structure optimization.

关键词： Evolutionary algorithm multi-objective optimization Financial industrial structure Big Data

来源：评论

学校读者我要写书评

暂无评论

Parallel Bayesian optimization of multiple noisy objectives with expected hypervolume improvement 21

Parallel Bayesian optimization of multiple noisy objectives ...

引用

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Samuel Daulton Maximilian Balandat Eytan Bakshy Facebook University of Oxford Facebook

ISBN: (纸本)9781713845393

Optimizing multiple competing black-box objectives is a challenging problem in many fields, including science, engineering, and machine learning. multi-objective Bayesian optimization (MOBO) is a sample-efficient approach for identifying the optimal trade-offs between the objectives. However, many existing methods perform poorly when the observations are corrupted by noise. We propose a novel acquisition function, NEHVI, that overcomes this important practical limitation by applying a Bayesian treatment to the popular expected hypervolume improvement (EHVI) criterion and integrating over this uncertainty in the Pareto frontier. We argue that, even in the noiseless setting, generating multiple candidates in parallel is an incarnation of EHVI with uncertainty in the Pareto frontier and therefore can be addressed using the same underlying technique. Through this lens, we derive a natural parallel variant, qNEHVI, that reduces computational complexity of parallel EHVI from exponential to polynomial with respect to the batch size. qNEHVI is one-step Bayes-optimal for hypervolume maximization in both noisy and noiseless environments, and we show that it can be optimized effectively with gradient-based methods via sample average approximation. Empirically, we demonstrate not only that qNEHVI is substantially more robust to observation noise than existing MOBO approaches, but also that it achieves state-of-the-art optimization performance and competitive wall-times in large-batch environments.

关键词：

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：