检索结果-内蒙古大学图书馆

Optimized deep learning vision system for human action recognition from drone images

MULTIMEDIA TOOLS AND APPLICATIONS 2023年第1期83卷 1143-1164页

作者： Samma, Hussein Bin Sama, Ali Salem King Fahd Univ Petr & Minerals SDAIA KFUPM Joint Res Ctr Artificial Intelligence Dhahran Saudi Arabia King Faisal Univ Coll Business Adm Dept Management Informat Syst MIS Al Hasa Eastern Provinc Saudi Arabia

There are several benefits to constructing a lightweight vision system that is implemented directly on limited hardware devices. Most deep learning-based computer vision systems, such as YOLO (You Only Look Once), use computationally expensive backbone feature extractor networks, such as ResNet and Inception network. To address the issue of network complexity, researchers created SqueezeNet, an alternative compressed and diminutive network. However, SqueezeNet was trained to recognize 1000 unique objects as a broad classification system. This work integrates a two-layer particle swarm optimizer (TLPSO) into YOLO to reduce the contribution of SqueezeNet convolutional filters that have contributed less to human action recognition. In short, this work introduces a lightweight vision system with an optimized SqueezeNet backbone feature extraction network. Secondly, it does so without sacrificing accuracy. This is because that the high-dimensional SqueezeNet convolutional filter selection is supported by the efficient TLPSO algorithm. The proposed vision system has been used to the recognition of human behaviors from drone-mounted camera images. This study focused on two separate motions, namely walking and running. As a consequence, a total of 300 pictures were taken at various places, angles, and weather conditions, with 100 shots capturing running and 200 images capturing walking. The TLPSO technique lowered SqueezeNet's convolutional filters by 52%, resulting in a sevenfold boost in detection speed. With an F1 score of 94.65% and an inference time of 0.061 milliseconds, the suggested system beat earlier vision systems in terms of human recognition from drone-based photographs. In addition, the performance assessment of TLPSO in comparison to other related optimizers found that TLPSO had a better convergence curve and achieved a higher fitness value. In statistical comparisons, TLPSO surpassed PSO and RLMPSO by a wide margin.

关键词： Deep learning Human action recognition YOLO SqueezeNet optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Stacking ensemble approach in data mining methods for landslide prediction

引用

JOURNAL OF SUPERCOMPUTING 2023年第8期79卷 8583-8610页

作者： Abdollahizad, Solmaz Balafar, Mohammad Ali Feizizadeh, Bakhtiar Sangar, Amin Babazadeh Samadzamini, Karim Islamic Azad Univ Dept IT & Comp Engn Urmia Branch Orumiyeh *** Iran Univ Tabriz Fac Elect & Comp Engn Tabriz *** Iran Univ Tabriz Dept Remote Sensing & GIS Tabriz *** Iran Univ Coll Nabi Akram Dept Comp Engn Tabriz *** Iran

Landslide susceptibility mapping is still an ongoing requirement for variety of applications such as land use management plans. The central objective of the present research was to investigate the effect of using ensemble machine learning methods for developing accurate landslide prediction. We aimed to explore and compare three techniques, namely the random forests, support vector machine and multiple-layer neural networks with an adaptive neuro-fuzzy inference system, which incorporates three metaheuristic methods including grey wolf optimization, particle swarm optimization, and shuffled frog leaping algorithm for landslide susceptibility assessment in the East Azerbaijan of Iran. Also, two ensemble ways (voting and stacking) were used in final decision stage. A sum of 766 locations with landslide inventory was recognized in the context of the study. Then the all models were trained using tenfold cross-validation technique. Lastly, the receiver operating characteristic and statistical procedures were employed to validate and contrast the predictive capability of the models. The findings of the study show the ANFIS-PSO model had high performance on the validation dataset (AUC = 0.89). Besides, the study revealed that using stacking ensemble technique could increase the predicting capability of all models (AUC = 0.911).

关键词： Stacking ensemble Adaptive neuro-fuzzy inference system Machine learning optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

A Markovian Incremental Stochastic Subgradient Algorithm

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2023年第1期68卷 124-139页

作者： Massambone, Rafael Costa, Eduardo Fontoura Helou, Elias Salomao Demeter Agrosolut BR-04536001 Sao Paulo Brazil Univ Sao Paulo Inst Ciencias Math & Computacao BR-13560970 Sao Carlos Brazil

In this article, a stochastic incremental subgradient algorithm for the minimization of a sum of convex functions is introduced. The method sequentially uses partial subgradient information, and the sequence of partial subgradients is determined by a general Markov chain. This makes it suitable to be used in networks, where the path of information flow is stochastically selected. We prove convergence of the algorithm to a weighted objective function, where the weights are given by the Cesaro limiting probability distribution of the Markov chain. Unlike previous works in the literature, the Cesaro limiting distribution is general (not necessarily uniform), allowing for general weighted objective functions and flexibility in the method.

关键词： Incremental subgradient algorithms optimization optimization algorithms randomized algorithms

来源：评论

学校读者我要写书评

暂无评论

Convolutional neural network-multi-kernel radial basis function neural network-salp swarm algorithm: a new machine learning model for predicting effluent quality parameters

引用

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH 2023年第44期30卷 99362-99379页

作者： Khozani, Zohreh Sheikh Ehteram, Mohammad Mohtar, Wan Hanna Melini Wan Achite, Mohammed Chau, Kwok-wing Alfred Wegener Inst Helmholtz Ctr Polar & Marine Res Paleoclimate Dynam Grp D-27570 Bremerhaven Germany Semnan Univ Dept Water Engn Semnan Iran Univ Kebangsaan Malaysia Fac Engn & Built Environm Dept Civil Engn Ukm Bangi 43600 Selangor Malaysia Hassiba Benbouali Univ Chlef Water & Environm Lab BP 78 COuled Fares Chlef 02180 Algeria Hong Kong Polytech Univ Dept Civil & Environm Engn Hung Hom Kowloon Hong Kong Peoples R China

A wastewater treatment plant (WWTP) is an essential part of the urban water cycle, which reduces concentration of pollutants in the river. For monitoring and control of WWTPs, researchers develop different models and systems. This study introduces a new deep learning model for predicting effluent quality parameters (EQPs) of a WWTP. A method that couples a convolutional neural network (CNN) with a novel version of radial basis function neural network (RBFNN) is proposed to simultaneously predict and estimate uncertainty of data. The multi-kernel RBFNN (MKRBFNN) uses two activation functions to improve the efficiency of the RBFNN model. The salp swarm algorithm is utilized to set the MKRBFNN and CNN parameters. The main advantage of the CNN-MKRBFNN-salp swarm algorithm (SSA) is to automatically extract features from data points. In this study, influent parameters (if) are used as inputs. Biological oxygen demand (BODif), chemical oxygen demand (CODif), total suspended solids (TSSif), volatile suspended solids (VSSif), and sediment (SEDef) are used to predict EQPs, including CODef, BODef, and TSSef. At the testing level, the Nash-Sutcliffe efficiencies of CNN-MKRBFNN-SSA are 0.98, 0.97, and 0.98 for predicting CODef, BODef, and TSSef. Results indicate that the CNN-MKRBFNN-SSA is a robust model for simulating complex phenomena.

关键词： Wastewater treatment plant Deep learning optimization algorithms Artificial neural networks

来源：评论

学校读者我要写书评

暂无评论

A New Protection System Design of Active MV Distribution Network

引用

TEHNICKI VJESNIK-TECHNICAL GAZETTE 2023年第1期30卷 131-137页

作者： Provci, Igor Stojkov, Marinko Maric, Predrag Duracic, Ivan Croatian Transmiss Syst Operator Ltd Transmiss Area Osijek Vukovarska cesta 217 Osijek 31000 Croatia Univ Slavonski Brod Mech Engn Fac Trg I B Mazuran 2 Slavonski Brod 35000 Croatia Univ Osijek Fac Elect Engn Comp Sci & Informat Technol Osijek Kneza Trpimira 2b Osijek 31000 Croatia

The increasing implementation of renewable energy sources (RES), along with the diversity of energy source types, has additionally imposed significant operational and management problems in distribution networks. These problems are manifested in voltage regulations, system stability and coordination of protection, both in the distribution and transmission networks. For the medium voltage (MV) network, this includes new energy sources and higher amounts of fault currents, invisibility of several faults in the existing protection scheme, reduction of the range of protection devices and reduction of the possibility of detecting small fault currents with existing protection relays. Such changes significantly reduce the possibility of proper distribution system protection, both in subordinate and superior networks. The subject of this paper is the presentation of a new concept of the use of automation in the management and arrangement of power system protection dependent on the scheme and configuration of an active MV network. The goals of this analysis and research are to find and define the necessary architecture in which the scheme and appearance of the MV network should be automatically detected, and based on network topology to establish new settings of protection devices (ground fault, overcurrent and short circuit protection). The contributions of generation units of RES in the MV network must be considered. This paper specifically analyses the problems of power system management with simultaneous harmonization of protection systems both in the transmission and in the radial distribution network, offering optimization algorithms that have the ability of achieving the optimal solution. The implementation of the proposed technique was tested on a radial connection integrated with a microgrid (MG) which has the possibility of two-way power supply. The obtained results indicate that the proposed technique can solve described problems in the coordination of protection

关键词： automated fault analysis digital fault recorder distributed production fuzzy expert system guidance automation optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on Information Dissemination and Monitoring of Online Public Opinion Under an Intelligent Algorithm

引用

International Journal of Network Security 2025年第2期27卷 330-337页

作者： Zhang, Huidan Li, Qingjuan Shen, Hongyan Xia Qing Communication School Handan University No. 530 North College Road Hanshan District Hebei Handan056000 China School of Chinese and History Handan University Hebei Handan056005 China Departpment of Propaganda Handan University Hebei Handan056005 China

Under the influence of network popularization, the information dissemination speed of online public opinion is faster, and public opinion events appear more frequently. It is necessary to effectively monitor them. This paper firstly analyzed the information dissemination of online public opinion, selected some relevant indicators, and designed an improved sooty tern optimization algorithm-back propagation neural network (ISTOABPNN) algorithm to realize online public opinion monitoring. Through tests on three test functions, it was found that compared with the SOA, GWO, and other optimization algorithms, the ISTOA obtained better solutions, which verified its optimization ability. Through a case analysis, it was found that compared with the support vector machine and BPNN algorithms, the output result of the ISTOA-BPNN algorithm was in better agreement with the expected output. The accuracy of the ISTOA was 95% for 20 test samples. The experimental results verify that the ISTOA-BPNN algorithm is reliable in online public opinion monitoring and can be applied in practice. © (2024), (Femto Technique Co. Ltd.). All Rights Reserved.

关键词： optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Powering up microgrids: A comprehensive review of innovative and intelligent protection approaches for enhanced reliability

引用

ENERGY REPORTS 2023年 10卷 1899-1924页

作者： Alasali, Feras Saad, Saad M. Saidi, Abdelaziz Salah Itradat, Awni Holderbaum, William El-Naily, Naser Elkuwafi, Fatima F. Hashemite Univ Fac Engn Dept Elect Engn POB 330127 Zarqa 13133 Jordan Coll Elect & Elect Technol Benghazi Benghazi Libya King Khalid Univ Coll Engn Elect Engn Dept Abha 61421 Saudi Arabia Hashemite Univ Fac Engn Dept Comp Engn POB 330127 Zarqa 13133 Jordan Univ Reading Sch Biomed Engn Reading RG6 6AY England Benghazi Univ Sch Grad Studies Benghazi 3332 MV4 Libya

Microgrid protection schemes play a vital role in ensuring the reliability and efficiency of power distribution in urban and rural areas, especially as renewable distributed energy resources are increasingly integrated. This paper aims to provide a comprehensive analysis of existing microgrid protection schemes, discussing their advantages and limitations and highlighting key challenges and opportunities for future research. As microgrid systems become increasingly common, the management of power flow in small-community networks equipped with intelligent electronic devices, non-linear loads, and multiple distributed generation sources becomes more complicated. In order to address these challenges, coordination of protective schemes is required to prevent overload and damage to equipment. Firstly, the study discusses microgrid definitions and functional categories, highlighting their benefits and drawbacks. An analysis of microgrid protection literature includes adaptive protection systems as intelligent methods to address coordination challenges. Secondly, this review classifies microgrid protection techniques as modified, new knowledge-based and conventional schemes and provides a systematic analysis of optimization approaches. The study also examines the essential problems associated with the coordination of protective relays within microgrids. Finally, examining the current state of microgrid protection to identify the key research directions and opportunities for future development in this rapidly advancing field. The findings of this comprehensive analysis highlight the importance of effective microgrid protection in ensuring a stable and sustainable energy future.

关键词： Microgrid Power protection Overcurrent relays Coordination Renewable energy Intelligent devices optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Optimal dispatch of generators based on network constrained to enhance power deliverable using the heuristic approach

引用

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH 2023年第44期30卷 99076-99087页

作者： Selladurai, Ravindran Chelladurai, Chinnadurrai Jayakumar, Muralidharan KPR Inst Engn & Technol Dept EEE Coimbatore Tamil Nadu India Bannari Amman Inst Technol Dept EEE Erode India Schneider Elect India Pvt Ltd Maintenance Dept Chennai Tamil Nadu India

In the rapid development of the electric power industry, there is a very steep increase in load. In the era of the modern power sector, renewable-based distributed generation has been forced to incorporate into the network. The classical generator dispatch method concentrates only on economic and environmental concerns during dispatch. These approaches are not scheduled considering the capacity benefit approach. The proposed work proposes a fourth-angle view of generator dispatch to meet the demand considering the available transfer capability of the network. This novel method gives an edge in the overloading of the network over classical economic-based load dispatch methods. Since the problem formulated is based on network loading, the objective functions are non-linear and particle swarm optimization is implemented to solve the solutions. This paper presents the comparative result analysis based on economical based optimal power flow (EPOF) and capacity benefited optimal power flow (COPF). From the case studies, we could find the available transfer capability in IEEE 30 bus system has increased by 13% when compared with solving in classical optimal power flow.

关键词： Optimal power flow Network loading Economic dispatch optimization algorithms Available transfer capability

来源：评论

学校读者我要写书评

暂无评论

Predicting Hydropower Production Using Deep Learning CNN-ANN Hybridized with Gaussian Process Regression and Salp Algorithm

引用

WATER RESOURCES MANAGEMENT 2023年第9期37卷 3671-3697页

作者： Ehtearm, Mohammad Zadeh, Hossein Ghayoumi Seifi, Akram Fayazi, Ali Dehghani, Majid Semnan Univ Dept Water Engn Semnan Iran Vali e Asr Univ Rafsanjan Dept Elect Engn Rafsanjan Iran Vali e Asr Univ Rafsanjan Coll Agr Dept Water Sci & Engn POB 515 Rafsanjan Iran Vali e Asr Univ Rafsanjan Fac Civil Engn Dept Tech & Engn POB 518 Rafsanjan Iran

The hydropower industry is one of the most important sources of clean energy. Predicting hydropower production is essential for the hydropower industry. This study introduces a hybrid deep learning model to predict hydropower production. Statistical methods are unsuitable for modeling hydropower production because their accuracy depends on seasonal and periodic fluctuations. For accurate predictions, deep learning models can capture daily, weekly, and monthly patterns. Since ANNs may not capture latent and nonlinear patterns, we use deep learning models to predict hydropower production. We used Convolutional Neural Network-Multilayer Perceptron-Gaussian Process Regression (CNNE-MUPE-GPRE) to extract key features and predict outcomes. The main advantages of the hybrid model are the quantification of production uncertainty, the accurate prediction of hydropower production, and the extraction of features from input data. We use a binary SSOA to select optimal input scenarios. The new model is benchmarked against the long short term memory neural network (LSTM), Bi directional LSTM (BI-LSTM), MUPE, GPRE, MUPE-GPRE, CNNE-GPRE, and CNNE-MUPE models. The models are used to predict 1-, 2-, and 3-day ahead power. The root mean square error values of CNNE-MUPE-GPRE, CNNE-MUPE, CNNE-GPRE, MUPE-GPRE, BI-LSTM, LSTM, CNNE, MUPE, GPRE were 578, 615, 832, 861, 914, 934, 1436, 1712, and 1954 KW at the 1-day prediction horizon. The RMSE of the CNNE-MUPE-GPRE was 595, 600, and 612 at the 1-day, 2-days, and 3-days prediction horizons. Extending the prediction horizon degrades accuracy. The uncertainty of the CNNE-MUPE-GPRE model was lower than that of the other models. The CNNE-MUPE-GPRE model is recommended for more accurate hydropower production predictions.

关键词： Hydropower Deep learning models optimization algorithms Power generation

来源：评论

学校读者我要写书评

暂无评论

Emperor penguin optimizer: A comprehensive review based on state-of-the-art meta-heuristic algorithms

引用

ALEXANDRIA ENGINEERING JOURNAL 2023年 63卷 487-526页

作者： Khalid, Othman Waleed Isa, Nor Ashidi Mat Sakim, Harsa Amylia Mat Univ Sains Malaysia Sch Elect & Elect Engn Engn Campus George Town 14300 Malaysia

Meta heuristics is an optimization approach that works as an intelligent technique to solve optimization problems. Evolutionary algorithms, human-based algorithms, physics-based algorithms and swarm intelligence are categorized under meta-heuristic algorithms. This study pre-sents a critical review of meta-heuristic algorithms for future reference, including concepts, applica-tions, advantages and disadvantages, before focusing on one specific meta-heuristic algorithm, namely, Emperor Penguin Optimizer (EPO). It is an intelligent algorithm developed after observing the behaviour of emperor penguins during cold winters. This technique was introduced by Dhiman in 2018 and adopted to solve optimization problems. The study reviews the algorithm variants start-ing from its invention in 2018 until 2022. The literature is comprehensively reviewed to reflect on the progress of the algorithm's adoption, highlighting a new area for improvement. The most significant result is that the proposed algorithm has been proven an effective technique. The merits and demer-its of the algorithm are explored to provide valuable perspectives for future research. This study answers the question regarding meta-heuristic algorithms' effectiveness, especially EPO. Both begin-ners and experts of EPO research can use the findings of this study as guidelines for enhancing cur-rent concepts and applications of state-of-the-art algorithms for future development works.(c) 2022 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Alexandria University. This is an open access article under the CC BY-NC-ND license (http://***/ licenses/by-nc-nd/4.0/).

关键词： Meta heuristic Emperor Penguin Optimizer EPO variants optimization algorithms Huddling behaviour Swarm intelligence

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

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

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：