检索结果-内蒙古大学图书馆

A reinforcement learning-based ranking teaching-learning-based optimization algorithm for parameters estimation of photovoltaic models

引用

SWARM AND EVOLUTIONARY COMPUTATION 2025年 93卷

作者： Wang, Haoyu yu, xiaobing Lu, Yangchen Nanjing Univ Informat Sci & Technol Res Inst Risk Governance & Emergency Decis Making Sch Management Sci & Engn Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Collaborat Innovat Ctr Forecast & Evaluat Meteorol Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Sch Management Sci & Engn Nanjing Peoples R China

Accurate parameter control and optimization are vital issues in the process of utilizing solar energy through photovoltaic systems, which poses significant challenges due to the inherent complexity of photovoltaic systems. The paper proposes a new algorithm, reinforcement learning-based ranking teaching-learning-based optimization, for identifying photovoltaic model parameters. Parameters play a major role in the performance of many optimization algorithms, and the same parameter is not appropriate for all problems. Reinforcement learning adjusts parameters by accumulating returns to meet the requirements of the environmental model. The proposed algorithm divides learners into superior and inferior groups based on fitness rankings and uses a reinforcement learning approach to dynamically adjust learner classification divisions to ensure adaptability in different optimization scenarios. In the teacher phase, superior learners emulate top performers, while inferior learners engage in guided mutual learning to enhance global search capabilities. In the learner phase, superior learners communicate with better peers, while inferior learners seek a wider range of information sources, balancing exploration and exploitation. In the experimental evaluation of five different photovoltaic models, the comparative analysis of eleven established algorithms verified its superior performance in accuracy, convergence speed, and complexity.

关键词： Photovoltaic system Parameter identification Teaching-learning-based optimization Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

An optimized extreme learning machine composite framework for point, probabilistic, and quantile regression forecasting of carbon price

引用

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION 2025年 195卷

作者： Wang, Xuming Zhou, Jiaqi yu, xiaobing Nanjing Univ Informat Sci & Technol Engn Training Ctr Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Sch Management Sci & Engn Nanjing 210044 Peoples R China

The increasing impact of carbon dioxide emissions on the ecological environment has aroused widespread concern. Precise estimation of carbon prices is essential for rational planning of carbon emissions and mitigation of environmental crises. Carbon price series exhibit nonlinear features, and capturing these features is challenging. Therefore, this paper proposes a carbon price forecasting framework (EEMD-ELM-EGWO) based on signal decomposition and Extreme Learning Machine (ELM). Under this forecasting framework, we realize point forecasting, probabilistic forecasting, and quantile regression interval forecasting. Firstly, the carbon price is decomposed into multiple Intrinsic Mode Functions (IMFs) using Ensemble Empirical Mode Decomposition (EEMD). Subsequently, an intelligently optimized ELM is used to forecast each IMF. Finally, the individual forecasts are aggregated to produce the final forecast. To optimize the hidden layer matrices of the ELM, an Enhanced Grey Wolf Optimizer (EGWO) algorithm is developed. Experimental results show that the proposed method provides more accurate point forecasting results and exhibits superior nonlinear fitting ability compared to other signal processing techniques and intelligent algorithms. In addition, the proposed framework is able to more accurately characterize the uncertainty of carbon price series. Together, these three forecasting models provide carbon market participants with more comprehensive and accurate information on future price trends, which helps to formulate rational policy planning and market strategies.

关键词： Carbon emission trading price Time series forecasting Point forecasting Probabilistic forecasting Interval forecasting

来源：评论

学校读者我要写书评

暂无评论

Multi-objective optimal power flow problem using constrained dynamic multitasking multi-objective optimization algorithm

引用

SWARM AND EVOLUTIONARY COMPUTATION 2025年 93卷

作者： Zhu, Junhua yu, xiaobing Wang, Feng Mao, Yaqi Nanjing Univ Informat Sci & Technol Res Inst Risk Governance & Emergency Decis Making Sch Management Sci & Engn Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Collaborat Innovat Ctr Forecast & Evaluat Meteorol Nanjing 210044 Peoples R China

The multi-objective optimal power flow (MOOPF) problem involves conflicting objectives and complex constraints, presenting a significant challenge for existing optimization methods. To address constrained multi- objective optimization problems (CMOPs), a recent evolutionary multi-tasking (EMT) framework has been proposed, involving a primary task and several auxiliary tasks running in parallel. The design of these auxiliary tasks is critical for supporting the solution of the primary task. This paper introduces a novel constrained dynamic multitasking multi-objective optimization algorithm (CDMTMO) to solve CMOPs. The proposed algorithm comprises three populations, each assigned a specific task: the first population focuses on solving the primary CMOP, the second population tackles a constraint-relaxed problem, and the third population gradually transitions from solving an unconstrained problem to the CMOP. To ensure effective collaboration among auxiliary tasks and the main task, CDMTMO incorporates an improved epsilon-constrained method and an enhanced dual ranking method. Furthermore, a pre-selection strategy for solution sets is integrated to discern promising individuals and facilitate knowledge transfer. CDMTMO has been evaluated using two IEEE standard systems, to demonstrate its capability and suitability in efficiently tackling the MOOPF problem. A thorough analysis of CDMTMO's results was performed, comparing it with seven state-of-the-art algorithms: AGEMOEA, CCMO, DSPCMDE, CMOEMT, ToP, EMCMO and DEST. After evaluating across eight test cases, CDMTMO achieved the best inverted generational distance plus (IGD+) and hypervolume (HV) values in seven cases. Furthermore, CDMTMO achieves a feasible rate (FR) of 1 in all cases, demonstrating its consistent ability to find feasible solutions.

关键词： Optimal power flow Constrained multi-objective optimization Constraint handling technique Auxiliary task Knowledge transfer

来源：评论

学校读者我要写书评

暂无评论

Convolutional neural network combined with reinforcement learning-based dual-mode grey wolf optimizer to identify crop diseases and pests

引用

SWARM AND EVOLUTIONARY COMPUTATION 2025年 94卷

作者： Lu, Yangchen yu, xiaobing Hu, Zhengpeng Wang, Xuming Nanjing Univ Informat Sci & Technol Sch Management Sci & Engn Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Res Inst Risk Governance & Emergency Decis Making Sch Management Sci & Engn Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Collaborat Innovat Ctr Forecast & Evaluat Meteorol Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Engn Training Ctr Nanjing 210044 Peoples R China

Agriculture is crucial for national food security, but crop pests and diseases pose significant threats. Traditional manual methods for detection are subjective, costly, and less accurate. Deep learning, especially convolutional neural network, is revolutionizing crop pest and disease identification, manual hyperparameter tuning can lead to suboptimal results. In contrast, grey wolf optimizer has demonstrated effective global search capabilities in hyperparameter optimization, improving model performance. Therefore, a reinforcement learning-based dualmode grey wolf optimizer is introduced to enhance the performance of the original algorithm in hyperparameter optimization and identify the optimal hyperparameters, which combines a dynamic elite learning strategy and a dual-mode adaptive strategy well balanced with the exploration and exploitation of populations, while the integration of the reinforcement learning technique strengthens the information feedback. To validate the effectiveness of the proposed algorithm, additional ablation experiments were conducted, and experiments using CPU time as the termination criterion were included to increase rigor and ensure fairness. The main hyperparameters of convolutional neural network optimized by the proposed algorithm is utilized for the recognition of the pentatomidae stinkbug pests and corn diseases, with experimental results compared against six other intelligent optimization algorithms. Results from two sets of experiments indicate that the proposed algorithm improves the recognition accuracy of the original convolutional neural networks model, achieving the highest accuracy on the pest dataset at 95.83 % and on the corn disease dataset at 96.51 %.

关键词： Crop disease and pest Convolutional neural network Grey wolf optimizer Hyperparameter optimization (HPO) Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Images Change Detection Based on Dense Multi-scale Attention for Land Resource Auditing

引用

KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS 2025年第3期19卷 907-925页

作者： Zhang, Jinjin Dai, Ranchen Tang, Yongsheng Zhan, Tianming yu, xiaobing Nanjing Audit Univ Nanjing Sch Comp Sci Nanjing 211815 Peoples R China Nanjing Audit Univ Nanjing Sch Intelligence Audit Nanjing 211815 Peoples R China Nanjing Brain Hosp Nanjing 210024 Peoples R China Nanjing Chest Hosp Nanjing 210029 Peoples R China

Land resource audit is an important aspect of natural resource conservation, but achieving full coverage is challenging due to the vast geographical area. The development of hyperspectral imaging technology in recent years has effectively addressed the difficulties in land resource audit. However, further research is needed on how to accurately extract change information from hyperspectral images. Over the past period, Deep learning methods have been widely applied in tasks pertaining to hyperspectral images change detection, yielding commendable outcomes. This article introduces a dense multi-scale attention-based approach for detecting multiple classes of changes in hyperspectral images. This method makes the most of spectral details regarding hyperspectral images and introduces multi-scale structures and dense links in spatial modules, effectively improving change detection results. Accurate. The outcomes from experiments conducted across various hyperspectral datasets certify that approach surpasses most present methodologies in multi-category change detection.

关键词： Hyperspectral image change detection multiscale structure dense net attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Research on Compensation Approach for Motion Error of Rotating Accelerometer Gravity Gradiometer

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Liang, yu yu, Mingbiao yu, xiaobing Ke, Bohai Jiang, Tao yu, Li Fan, Ji Hu, Chenyuan Tu, Liangcheng Liu, Huafeng Zhou, Zebing Huazhong Univ Sci & Technol MOE Key Lab Fundamental Phys Quant Measurement Hubei Key Lab Gravitat & Quantum Phys NGL Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Phys Wuhan 430074 Peoples R China Sun Yat sen Univ TianQin Res Ctr Gravitat Phys Frontiers Sci Ctr TianQin Sch Phys & AstronMOE Key Lab TianQin Mission Zhuhai Campus Zhuhai 519082 Peoples R China Sun Yat sen Univ CNSA Res Ctr Gravitat Waves Zhuhai Campus Zhuhai 519082 Peoples R China

The rotating accelerometer gravity gradiometer (RAGG) is a gravity gradient measuring instrument designed for moving bases, and as such, the primary technical challenge in RAGG development lies in compensating for motion error. Nevertheless, the existing open research is limited in number and exhibits several limitations: a lack of systematic design for monitoring systems, inadequate model interpretability, missing parameters, and lack of experimental validation, among others. To address these issues, this article proposes a novel approach for compensating RAGG motion error, which encompasses two key contributions: 1) development of a multiparameter compensation model for RAGG motion errors based on the motion transfer mechanism and accelerometer output model;and 2) by leveraging RAGG's internal accelerometers, only four external sensors are utilized to achieve six-degree-of-freedom motion monitoring, thereby streamlining the motion monitoring system. Furthermore, a six-degree-of-freedom movement platform is utilized to construct an experimental test platform, validating the method's effectiveness and practical applicability through real experiments. The proposed method is also subjected to comparative experimentation with existing approaches. The experimental results demonstrate that the method presented in this study exhibits superior performance, achieving over three orders of magnitude of compensation capability within the 0.35-0.65-Hz bandwidth, and attaining a motion error compensation level of 99.91%.

关键词： Accelerometers Gravity Monitoring Geologic measurements Error compensation Physics Measurement uncertainty Extraterrestrial measurements Coordinate measuring machines Angular velocity Earth gravitational field error compensation rotating accelerometer gravity gradiometer (RAGG)

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional choroidal characteristics in different subtypes of central serous chorioretinopathy using swept-source optical coherence tomography angiography

引用

GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY 2025年第3期263卷 689-697页

作者： Zhang, yue Wang, Jianing Liu, Jing Gu, Xiaoya Song, Shuang yu, xiaobing Chinese Acad Med Sci Beijing Hosp Inst Geriatr Med Natl Ctr GerontolDept Ophthalmol Beijing Peoples R China Peking Union Med Coll Grad Sch Beijing Peoples R China

Purpose To assess choroid vascular characteristics in four subtypes of central serous chorioretinopathy (CSC) eyes under the new classification system. Methods There were 83 subjects in total for analysis including 16 individuals with acute CSC, 13 with non-resolving CSC, 12 with recurrent CSC, 16 with chronic CSC, and 26 healthy control eyes. We utilized the integrated software of SS-OCTA to acquire measurements of the central choroidal thickness (CT), the choriocapillaris perfusion area (CCPA), three-dimensional choroidal vascularity index (CVI) and three-dimensional choroidal vessel volume (CVV). The SNK-q test was conducted for pairwise comparisons among four subgroups of CSC and healthy control eyes. A multiple linear regression model was employed to investigate the relationship between CVI, CCPA and other factors. Results CT, CVI, and CCPA of the chronic CSC subgroup were significantly lower than the other three CSC subgroups. Lower CVI was significantly correlated with the subgroup of chronic CSC (beta = -0.140, P = 0.016), lower CT (beta = 0.0014, P < 0.01), and higher CCPA (beta = -0.141, P < 0.01). Lower CCPA was significantly correlated with the subgroup of chronic CSC (beta = -0.937, P < 0.01), diabetes (beta = -0.118, P = 0.015), higher CVV (beta = -0.414, P = 0.014), and higher CVI (beta = -0.764, P < 0.01). Conclusions This is the first study to evaluate the choroidal characteristics of four subtypes of CSC under the new classification system in detail. It seems that chronic CSC displays distinct pathophysiological alterations in choroidal characteristics.

关键词： Central serous chorioretinopathy Swept-source optical coherence tomography angiography Choroidal vascularity index Choroidal vasculature

来源：评论

学校读者我要写书评

暂无评论

Gravity gradient signal extraction based on time-frequency feature threshold method

引用

JOURNAL OF APPLIED GEOPHYSICS 2025年 234卷

作者： Jiang, Tao Ke, Bohai yu, xiaobing yu, Li Yang, Meng Fan, Ji Hu, Chenyuan Feng, Wei Liu, Huafeng Zhong, Min Tu, Liangcheng Zhou, Zebing Huazhong Univ Sci & Technol Key Lab Fundamental Phys Quant Measurement Minist Educ MOE Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Hubei Key Lab Gravitat & Quantum Phys Natl Precise Grav Measurement Facil PGMF Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Phys Wuhan 430074 Peoples R China Sun Yat Sen Univ Sch Geospatial Engn & Sci Zhuhai 519082 Peoples R China Sun Yat Sen Univ TianQin Res Ctr Gravitat Phys MOE Key Lab TianQin Mission Zhuhai Campus Zhuhai 519082 Peoples R China Sun Yat Sen Univ Frontiers Sci Ctr TianQin CNSA Res Ctr Gravitat Waves Sch Phys & Astron Zhuhai Campus Zhuhai 519082 Peoples R China Southern Marine Sci & Engn Guangdong Lab Zhuhai Zhuhai 519082 Peoples R China

Gravity gradient, defined as the spatial derivative of gravitational acceleration, represents the detailed variation of the gravity field and plays a significant role in resource exploration and auxiliary navigation. Airborne gravity gradient measurement is an important method for obtaining high-precision gravity gradient data. Compared to traditional gravity measurement techniques, it offers superior measurement efficiency and provides high- frequency information of gravity field which reveals small-scale terrain features. However, it also introduces more high-frequency measurement noise. Due to the randomness of Earth's topography and resource distribution, the high-frequency characteristics of gravity gradient signals are mainly concentrated in specific time and frequency ranges, exhibiting distinct time-frequency features. This paper employs the VMD and Hilbert transform to obtains the time-frequency features of gravity gradient signal, and proposes a time-frequency feature threshold method to extract gravity gradient signal. This method can accurately analyze and extract the gravity gradient signals components without being affected by the same frequency noise, suppressing 78 % of the noise in simulation data. Additionally, the proposed method is validated in the gravity gradient data of actual terrain, demonstrating over 19 % improvement compared to traditional smooth filter and wavelet threshold methods. Furthermore, it effectively retains the high-frequency characteristics of gravity gradient signal, enhancing the advantages of airborne gravity gradient signals

关键词： Gravity gradient Signal extraction method Time-frequency feature

来源：评论

学校读者我要写书评

暂无评论

Decoupling compensation accuracy from sensor misalignment: A generalized black-box model for gravity gradiometers

引用

ISA Transactions 2025年 161卷 228-242页

作者： yu, Mingbiao Liang, yu yu, xiaobing Hu, Chenyuan Hou, Zhenlong yu, Li Zhao, Wei Liu, Huafeng Zhang, Liang Fan, Ji Sun, yuanhang Jiang, Congmei Tu, Liangcheng Zhou, Zebing Key Laboratory of Advanced Manufacturing Technology of the Ministry of Education School of Mechanical Engineering GuiZhou University Guiyang550025 China MOE Key Laboratory of Fundamental Physical Quantities Measurement & Hubei Key Laboratory of Gravitation and Quantum Physics PGMF and School of Physics Huazhong University of Science and Technology Wuhan430074 China Zhuhai519082 China School of Resources and Civil Engineering Northeastern University Shenyang Liaoning110819 China

Moving-base gravity gradiometers are widely used in applications such as mineral exploration and passive navigation;however, their high sensitivity to platform motion presents significant challenges in compensating for motion-induced errors. Existing methods rely on precise alignment between motion sensors and the gradiometer's coordinate system, making compensation accuracy heavily dependent on sensor installation precision. To address this issue, we propose a novel generalized black-box modeling method for motion error compensation in gravity gradiometers. By incorporating sensor misalignment directly into the model, our approach effectively decouples compensation accuracy from installation precision. Constructed based on the gradiometer's dynamic characteristics and the excitation and input–output behavior of its sensitive components, the black-box model inherently avoids redundant operational terms, reduces multicollinearity, and lowers computational resource requirements. Validation through numerical simulations in a dynamic 0.1 g environment demonstrates that the output consistency between the black-box and analytical models reaches 10−10, achieving an compensation accuracy of 0.1 E. Furthermore, experimental results using a gravity gradiometer prototype confirm the model's effectiveness in real-world conditions. The proposed method significantly reduces motion errors induced by 1 mg linear motion and 0.001 rad/s angular motion, lowering the noise level to 4 ng/Hz, approaching the static inherent noise level of 1 ng/Hz. These results verify the black-box model's effectiveness and robustness in dynamic environments, highlighting its potential to enhance the practical performance of gravity gradiometers by mitigating motion-induced errors without stringent requirements on sensor installation precision. © 2025

关键词： Mineral exploration

来源：评论

学校读者我要写书评

暂无评论

An improved reinforcement learning-based differential evolution algorithm for combined economic and emission dispatch problems

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 140卷

作者： Wang, yuan yu, xiaobing Zhang, Wen Nanjing Univ Informat Sci & Technol Res Inst Risk Governance & Emergency Decis Making Sch Management Sci & Engn Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Collaborat Innovat Ctr Forecast & Evaluat Meteorol Nanjing 210044 Peoples R China

To overcome challenges posed by escalating environmental pollution and climate change, the combined economic and emission dispatch problem is proposed to balance economic efficiency with emission cost. The primary objective of the problem is to ensure that emissions are minimized while optimal economic costs are achieved simultaneously. However, due to the nonlinear and nonconvex characteristics of the model, the optimization is confronted with many difficulties. Hence, an innovative improved reinforcement learning-based differential evolution algorithm is proposed in this article, with reinforcement learning seamlessly integrated into the differential evolution algorithm. Q-learning from reinforcement learning technique is utilized to dynamically adjust parameter settings and select appropriate mutation strategies, thereby boosting the algorithm's adaptability and overall performance. The effectiveness of the proposed algorithm is tested on thirty testing functions and combined economic and emission dispatch problems in comparison with the other five algorithms. According to the experimental results of testing functions, superior performance is consistently achieved by the proposed algorithm, with the highest adaptability exhibited and an average ranking of 1.4167. Its superiority is further demonstrated through Wilcoxon tests on results of testing functions and combined economic and emission dispatch problems with the proportion of 100%, and the proposed algorithm is significantly better than other algorithms at a 0.05 significance level. The superiority of the proposed algorithm in optimizing combined economic and emission dispatch problems demonstrates that the proposed algorithm is shown to be adaptable to complex optimization environments, which proves useful for industrial applications and artificial intelligence.

关键词： Combined economic emission dispatch Differential evolution Power system optimization Q-learning

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：