检索结果-内蒙古大学图书馆

Convolutional Neural Network- ANN- E (Tanh): A New Deep Learning Model for Predicting Rainfall

WATER RESOURCES MANAGEMENT 2023年第4期37卷 1785-1810页

作者： Nia, Mahdie Afshari Panahi, Fatemeh Ehteram, Mohammad Univ Kashan Fac Nat Resources & Earth Sci Kashan Iran Semnan Univ Dept Water Engn Semnan Iran

The prediction of rainfall is essential for monitoring droughts and floods. The purpose of this paper is to develop a deep learning model for predicting monthly rainfall. The new model is used to predict rainfall in the Kashan plain of Iran. This study combines a deep learning model with an artificial neural network (ANN) model to predict rainfall. In this study, a convolutional neural network (CONV) is used as a deep learning model. The paper also introduces a new activation function called E-Tanh to develop ANN models. The new model has two main advantages. The model automatically determines key features. In addition, the new activation function can enhance the precision of ANN models. Lagged rainfall values are inserted into the models to predict rainfall. This study uses a bat optimization algorithm to choose inputs. At the training level, the mean absolute percentage errors (MAPES) of CONV-ANN-ANN-E-Tanh, CONV, and ANN-E-Tanh were 0.5%, 1%, and 2%, respectively. At the testing level, the MAPEs of CONV-ANN -E-Tanh, CONV, and ANN-E-Tanh were 1%, 3%, and 4%, respectively. The E-Tanh performed better than other activation functions based on error function values. Also, the CONV-ANN-E-Tanh can reduce CPU time. Our results show that the new hybrid model is a reliable tool for simulating complex phenomena.

关键词： Hydrological predictions Hybrid model Water resource management optimization algorithm New activation function

来源：评论

学校读者我要写书评

暂无评论

Research on Stick-Slip Vibration Suppression Method of Drill String Based on Machine Learning optimization

引用

Sound & Vibration 2023年第1期57卷 97-117页

作者： Kanhua Su Jian Wei Meng Li Hao Li Wenghao Da Lang Zhang School of Petroleum Engineering Chongqing University of Science and TechnologyChongqing401331China

During the drilling process,stick-slip vibration of the drill string is mainly caused by the nonlinear friction gen-erated by the contact between the drill bit and the *** eliminate the fatigue wear of downhole drilling tools caused by stick-slip vibrations,the Fractional-Order Proportional-Integral-Derivative(FOPID)controller is used to suppress stick-slip vibrations in the drill *** the FOPID controller can effectively suppress the drill string stick-slip vibration,its structure isflexible and parameter setting is complicated,so it needs to use the cor-responding machine learning algorithm for parameter *** on the principle of torsional vibration,a simplified model of multi-degree-of-freedom drill string is established and its block diagram is *** continuous nonlinear friction generated by cutting rock is described by the LuGre friction *** adaptive learning strategy of genetic algorithm(GA),particle swarm optimization(PSO)and particle swarm optimization improved(IPSO)by arithmetic optimization(AOA)is used to optimize and adjust the controller parameters,and the drill string stick-slip vibration is suppressed to the greatest *** results show that:When slight drill string stick-slip vibration occurs,the FOPID controller optimized by machine learning algorithm has a good effect on suppressing drill string stick-slip ***,the FOPID controller cannot get the drill string system which has fallen into serious stick-slip vibration(stuck pipe)out of trouble,and the machine learning algorithm is required to mark a large amount of data on adjacent Wells to train the *** a reasonable range of drilling parameters(weight on bit/drive torque)in advance to avoid severe stick-slip vibration(stuck pipe)in the drill string system.

关键词： Stick-slip vibration machine learning fractional order proportional integral derivative(FOPID)control optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Nonlinear simulating of the Proton Exchange Membrane Fuel Cells utilizing Ridgelet Neural Network optimized using a hybrid form of Northern Goshawk Optimizer

引用

APPLIED ENERGY 2024年 360卷

作者： Li, Ruiheng Tian, Hao Di, Yi Hossain, Sarmistha Hubei Univ Econ Sch Informat & Engn Wuhan 430205 Hubei Peoples R China Hubei Key Lab Digital Finance Innovat Wuhan 430205 Hubei Peoples R China Hubei Univ Econ Hubei Internet Finance Informat Engn Technol Res C Wuhan 430205 Hubei Peoples R China Chittagong Univ Engn & Technol Chittagong 4349 Bangladesh

The utilization of modeling techniques for Proton Exchange Membrane Fuel Cells (PEMFC) presents a viable and proficient approach for comprehending, enhancing, and increasing the efficiency of these sustainable energy sources. This is of paramount importance in tackling worldwide anxieties regarding ecological contamination and the lessening of non-renewable resources of energy. The authors of this research paper introduce a novel approach for the efficient detection of output voltage in PEMFC. This approach involves the utilization of an optimized Ridgelet Neural Network (RNN) in conjunction with a Hybrid Northern Goshawk optimization (HNGO) algorithm. The primary aim of the proposed approach is to reduce the discrepancy among the determined and anticipated resulted voltages of the PEMFC. This is a vital aspect in increasing the fuel cells' efficacy. The authors conducted simulations to assess the efficacy of the proposed approach. The outcomes designated that the RNN/HNGO method outperforms existing methods when it comes to accuracy in tracking voltage signal predictions. Moreover, the proposed approach yields a comparatively lower level of forecast error compared to previously published studies. The aforementioned results underscored the possibility of utilizing the suggested approach in enhancing the performance of PEMFC through practical means. Additionally, the authors propose that the methodology they have presented has the potential to be expanded to encompass other performance metrics of PEMFCs, including power and efficiency. The authors propose the integration of the RNN/HNGO method with additional optimization techniques or its utilization in hybrid systems as a means of augmenting its performance. The authors concluded by acknowledging the need for additional research to experimentally validate the proposed methodology and optimizing its parameters in order to ensure its reliability and applicability in real-world scenarios.

关键词： Ridgelet Neural Network (RNN) Hybrid Northern Goshawk optimization (HNGO) Output voltage prediction Nonlinear modeling optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

A framework to reduce energy consumption in a press shop floor based on industrializable IIoT (I3oT)

引用

HELIYON 2024年第9期10卷 e29432页

作者： Peinado-Asensi, Ivan Montes, Nicolas Garcia, Eduardo CEU Univ Univ Cardenal Herrera CEU Dept Math Phys & Technol C San Bartolome 55 Valencia 46115 Spain Ford Spain Poligono Ind Ford S-N Valencia 46440 Spain

This article presents a framework to reduce energy consumption in a floor shop press based on Industrializable Industrial Internet of Things (I3oT). The I3oT proposes the development of IIoT tools using the information available in the system, without adding any additional sensors. Based on this philosophy, we proposed to develop the C360 criterion in our previous works, which allowed to extract all the information available in the stamping presses for the development of I3oT applications. In this article, we propose the development of a framework to optimize the parameters accessible from the C360 criterion for energy saving in the stamping process. Regarding the three parameters that can be modified and that affect energy consumption, that is, counterbalance pressure, tonnage and press speed, we will work with the first two in this paper. At the end of the article, the results obtained from the presses installed at Ford factory in Almussafes (Valencia) are shown based on their adjustment.

关键词： Industrializable IIoT(I3oT) Energy management optimization algorithm Automotive manufacturing Press shop

来源：评论

学校读者我要写书评

暂无评论

Optimizing energy costs and reliability: A multi-objective framework with learning-enhanced manta ray foraging for hybrid PV/battery systems

引用

ENERGY 2024年 291卷

作者： Alghamdi, Ali S. Majmaah Univ Coll Engn Dept Elect Engn Al Majmaah 11952 Saudi Arabia

This paper presents a multi-objective framework for designing a hybrid photovoltaic energy and battery storage system (PV/Battery) with the aim of minimizing the cost of electricity, loss of energy expectation, and loss of load expectation. Focusing on meeting the energy demand of a commercial complex in Al-Jubail industrial city, Saudi Arabia, utilizing real-world data, the proposed framework employs the multi-objective improved manta ray foraging optimization algorithm in conjunction with fuzzy decision-making. To enhance algorithm performance and prevent premature convergence, a Learning-based Hunting Movement Strategy is incorporated. Results highlight the superior performance of Case 3, achieving an optimal balance between cost and reliability. The obtained values for cost of electricity, loss of energy expectation, and loss of load expectation are 0.2255 $/kWh, 170.67 kWh/yr, and 14 h/yr, respectively. This study emphasizes the strengthened capability of the algorithm, augmented through the Learning-based Hunting Movement Strategy, establishing its superiority over established multi-objective methods with a higher percentage of dominant solutions.

关键词： Hybrid photovoltaic/battery energy system Multi-objective design framework Cost of electricity Reliability Learning-based hunting movement strategy Multi-objective improved manta ray foraging optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Application Of Chimp-based ANFIS Model For Forecasting The Compressive Strength Of The Improved High-performance Concrete

JOURNAL OF APPLIED SCIENCE AND ENGINEERING

引用

JOURNAL OF APPLIED SCIENCE AND ENGINEERING 2024年第4期27卷 2295-2306页

作者： Yuan, Yan Zhengzhou Univ Sci & Technol Fac Civil & Architectural Engn Zhengzhou 450064 Henan Peoples R China

In order to assess the compressive strength (CS) of high-performance concrete (HPC) prepared with fly ash and blast furnace slag, several artificial-based analytics were applied. This study, it was employed the Chimp optimizer ( CO) to identify optimal values of determinative factors of Support vector regression (SVR) and adaptive neuro-fuzzy inference system (ANFIS), which could be adjusted to improve performance. The suggested approaches were established using 1030 tests, eight inputs (a primary component of mix designs, admixtures, aggregates, and curing age), and the CS as the forecasting objective. The outcomes were then contrasted with those found in the body of existing scientific literature. Calculation results point to the potential benefit of combining CO - SVR and CO - ANFIS study. When compared to the CO - SVR, the CO - ANFIS showed much higher R2 and lower Root means square error values. Comparing the findings shows that the created CO- ANFIS is superior to anything that has previously been published. In conclusion, the suggested CO - ANFIS analysis might be used to determine the proposed approach for estimating the CS of HPC augmented with blast furnace slag and fly ash.

关键词： High-performance concrete Compressive strength Support vector regression Adaptive neuro-fuzzy inference system optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Prediction of miRNA-disease association based on multisource inductive matrix completion

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-15页

作者： Wang, Yawei Yin, Zhixiang Shanghai Univ Enginneering Sci Inst Frontier Med Technol Ctr Intelligent Comp & Appl Stat Sch Math Phys & Stat Shanghai 201620 Peoples R China

MicroRNAs (miRNAs) are endogenous non-coding RNAs approximately 23 nucleotides in length, playing significant roles in various cellular processes. Numerous studies have shown that miRNAs are involved in the regulation of many human diseases. Accurate prediction of miRNA-disease associations is crucial for early diagnosis, treatment, and prognosis assessment of diseases. In this paper, we propose the Autoencoder Inductive Matrix Completion (AEIMC) model to identify potential miRNA-disease associations. The model captures interaction features from multiple similarity networks, including miRNA functional similarity, miRNA sequence similarity, disease semantic similarity, disease ontology similarity, and Gaussian interaction kernel similarity between miRNAs and diseases. Autoencoders are used to extract more complex and abstract data representations, which are then input into the inductive matrix completion model for association prediction. The effectiveness of the model is validated through cross-validation, stratified threshold evaluation, and case studies, while ablation experiments further confirm the necessity of introducing sequence and ontology similarities for the first time.

关键词： miRNA-disease association Multi-source information Autoencoder Inductive matrix completion Ablation experiment optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Efficiency, optimality, and selection in a rigid actuation system with matching capabilities for an assistive robotic exoskeleton

引用

ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH 2024年 51卷

作者： Ghaffar, Asim Rahman, Muhammad Zia Ur Leiva, Victor Martin-Barreiro, Carlos Ali, Imran Cabezas, Xavier Castro, Cecilia Univ Engn & Technol Lahore Dept Mech Mechatron & Mfg Engn Faisalabad Campus Faisalabad Pakistan Pontificia Univ Catolica Valparaiso Escuela Ingn Ind Valparaiso Chile ESPOL Escuela Super Politecn Litoral Fac Ciencias Nat & Matemat Guayaquil Ecuador Univ Espiritu Santo Fac Ingn Samborondon Ecuador Escuela Super Politecn Litoral ESPOL Ctr Estudios Invest Estadist Guayaquil Ecuador Univ Minho Ctr Math Braga Portugal

Selecting the right actuator for a portable exoskeleton involves a comprehensive evaluation of various design characteristics. In this study, we introduce a methodology for actuator selection based on specific tasks, enhancing the practical adoption of portable exoskeletons. By examining a range of candidate actuators designed for lower limb exoskeletons, our objective is to engineer a system that is both lightweight and power-efficient. These candidate actuators, developed by integrating diverse motors and transmission systems, were rigorously tested against defined tasks. Our methodology, applied to an assistive exoskeleton catered to the elderly, showed its potential in tailoring an efficient system with matching capabilities. The obtained results indicated that the ideal configuration achieved reductions in weight and power requirements by 35% and 80%, respectively. The present research delineates a strategic approach for actuator selection in portable exoskeletons, contributing to the evolution of high-performing assistive devices.

关键词： Actuators Assistive robotics Dynamic modeling Mathematical simulation MATLAB software optimization algorithm Rehabilitation

来源：评论

学校读者我要写书评

暂无评论

Guaranteed Cost Gain-Scheduled Control of Typical Aeroelastic Section Model

引用

JOURNAL OF GUIDANCE CONTROL AND DYNAMICS 2025年第3期48卷 714-722页

作者： Pereira, Renan L. Gonzalez, Pedro J. Silvestre, Flavio J. Aeronaut Inst Technol Elect Engn Div BR-12228900 Sao Jose Dos Campos Brazil Tech Univ Berlin Flight Control & Aeroelast FMRA D-10587 Berlin Germany Tech Univ Berlin Flight Control & Aeroelast FMRA Marchstr 12 D-10587 Berlin Germany

来源：评论

学校读者我要写书评

暂无评论

Evaluation of machine learning applications in building life cycle processes for energy efficiency: A systematic review

引用

ENERGY REPORTS 2025年 13卷 4900-4916页

作者： Kaya, Gevher Nesibe Beyhan, Figen Ilerisoy, Zeynep Yesim Cudzik, Jan Siirt Univ Fac Fine Arts & Design Dept City & Reg Planning Siirt Turkiye Gazi Univ Fac Architecture Dept Architecture TR-06530 Ankara Turkiye Gdansk Univ Technol Fac Architecture Dept Urban Architecture & Waterside Spaces Gdansk Poland

In recent years, machine learning has been increasingly applied to achieve energy efficiency in buildings. This study analyzes the utilization of machine learning across the building life cycle by reviewing literature on building energy efficiency. In this context, a systematic literature search was conducted using the Web of Science (WOS) search engine, and 868 publications were found. The publications were analyzed according to their year, subject scope, and qualification results, and 84 publications were selected. These publications were discussed under five categories: objective function and control variables, programs, simulations, machine learning, and optimization algorithms. The relationships between these categories and each phase of the building life cycle were examined. The findings suggest that machine learning can effectively optimize factors related to energy efficiency and building sustainability throughout the life cycle, and it is anticipated that interdisciplinary studies incorporating machine learning will experience exponential growth in the future.

关键词： Machine learning Life cycle process Energy efficiency Building design optimization algorithm Systematic analysis

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：