检索结果-内蒙古大学图书馆

2nd IEEE International Conference on Contemporary Computing and Informatics (IC3I)

作者： Nagori, Viral GLS Univ Fac Comp Technol Ahmadabad Gujarat India

ISBN: (纸本)9781509052561

The back propagation algorithm has wide range of applications for training of feed forward neural networks. Over the years, many researchers have used back propagation algorithm to train their neural network based systems without emphasizing on how to fine tune the parameters of the algorithm. The paper throws the light on how researchers can manipulate and experiment with the parameters of the back propagation algorithm to achieve the optimum learning performance. The paper presents the results of the laboratory experiments of fine tuning the parameters of the back propagation algorithm. The process of fine tuning the parameters was applied on the neural network based expert system prototype. The prototype aims to analyze and design customized motivational strategies based on employees' perspective. The laboratory experiments were conducted on the following parameters of back propagation algorithm: learning rate, momentum rate and activation functions. Learning performance are measured and recorded. At the same time, the impact of activation function on the final output is also measured. Based on the results, the values of the above parameters which provide the optimum learning performance is chosen for the full scale system implementation.

关键词： back propagation algorithm neural network expert system motivational strategies learning performance activation function

来源：评论

学校读者我要写书评

暂无评论

Super resolution of Astronomical Objects using back propagation algorithm.

Super resolution of Astronomical Objects using Back Propagat...

引用

International Conference on Inventive Computation Technologies

作者： Heena Mhatre Varsha Bhosale Department of Information Technology Vidyalankar Institute of Technology Mumbai India

ISBN: (纸本)9781509012862

In the past years modern arithmetical methods for image investigation have led to a rebellion in many fields, from computer vision to scientific imaging. Though, some recently developed image processing techniques successfully oppressed by other sections have been infrequently, if ever, experimented on celestial observations. Here we present a new idea of super resolution of Astronomical objects using back propagation algorithm."Super-resolution "is efficient in improving the excellence of analysis of diffused sources formerly unobserved by the background noise, efficiently rising the depth of obtainable observations. Higher-resolution image out of a set of low resolution frames can be obtained through super-resolution. Super-resolution is viable only for point sources which have negligible dimensions, then for wideranging objects the knowledge about intensity vacillation at angular prevalence is irreversibly mislaid. Again obtaining super resolved image for extended sources(e.g. comets, meteoroids, etc) is a new challenge if the speed of the object is very high. Acquiring High resolution images of celestial objects from ground based telescopes is intricate and often requires computational post processing techniques to remove blur caused by atmospheric commotion. Even images obtained through satellite imaging are compressed and sent to earth. So there is need for Super resolution of those compressed or noisy images. So, here we simply implement Super-resolution for Astronomical objects using back propagation algorithm to overcome lost information and challenges for high speedy celestial objects. The purpose is to super resolve high speedy celestial objects whose analysis may in future help to prevent collisions of such celestial objects with earth and also avoid future solar system damage

关键词： Super resolution Astronomical observations back propagation algorithm Low resolution images Gradient descent function mean square error function Gaussian kernel Linear Kernel Blur Sigma factor sparse toeplitz matrix bicubic interpolation High resolution image

来源：评论

学校读者我要写书评

暂无评论

Optical character recognition using template matching and back propagation algorithm

Optical character recognition using template matching and ba...

引用

IEEE International Conference on Inventive Computation Technologies (ICICT)

作者： Desai, Swapnil Singh, Ashima Thapar Univ CSED Patiala Punjab India

Building an effective methodology to detect characters from images with less error rate is the great task. Our aim is to furnish such an algorithm that will be able to generate error free recognition of text from the given input image which will help in document digitizing and prevention to the hand written text recognition. OCR has been in the intensive research topic for more than 4 decades, it is probably the most time consuming and labor intensive work of inputting the data through keyboard. This paper discuss about mechanical or electronic conversion of scanned images, text which contain graphics, image captured by camera, scanned images and the recognition of images where characters may be broken or smeared. The optical character recognition is the desktop based application developed using Java IDE and mysql as a database. We have gain 91.82% accuracy when applied on different data sets, in pre-processing we used different techniques to remove noise from the image in post processing we used dictionary for the characters which are not recognized during classification, in classification we have used the back propagation algorithm for the training of neural network, feature extraction has been performed by template matching and hamming distance. All the algorithms have been developed in java technology.

关键词： artificial intelligence back propagation algorithm template matching ocr machine learning java api

来源：评论

学校读者我要写书评

暂无评论

Adaptive memetic differential evolution-back propagation-fuzzy neural network algorithm for robot control

引用

INFORMATION SCIENCES 2023年第1期637卷

作者： Zheng, Kunming Zhang, Qiuju Peng, Li Zeng, Shuisheng Jiangnan Univ Sch Mech Engn 1800 Lihu Ave Wuxi 214122 Peoples R China Jiangsu Key Lab Adv Food Mfg Equipment & Technol 1800 Lihu Ave Wuxi 214122 Peoples R China Jiangnan Univ Sch Internet Things Engn 1800 Lihu Ave Wuxi 214122 Peoples R China Changzhou Gucoi Intelligent Equipment Technol Res Changzhou 213001 Peoples R China

This study established an adaptive memetic differential evolution-back propagation-fuzzy neural network (AMDE-BP-FNN) control method to achieve high-efficiency and precise control of robots with complex dynamic characteristics while reducing control costs. The adaptive differential evolution (ADE) method was applied to search the optimal parameters in the global scope and delimited the pseudo-global search scope. The memetic differential evolution (MDE) method was used to search for optimal parameters in the pseudo-global scope, and the probability factor was set to decide whether to use the back propagation (BP) algorithm for online optimization. Finally, simulations, experiments, and real-world applications were conducted. The results indicated the high efficiency, high precision, and viability of the proposed AMDE-BP-FNN method.

关键词： Robot Memetic evolution algorithm Differential evolution algorithm back propagation algorithm Fuzzy neural network

来源：评论

学校读者我要写书评

暂无评论

Design of PSS for multi-machine system using extreme learning machine algorithm

引用

MEASUREMENT 2025年 247卷

作者： Suman, M. Rao, M. Venu Gopala Veerendra, A. S. Mopidevi, Subbarao Marquez, Fausto Pedro Garcia Vignans Lara Inst Technol & Sci Dept Elect & Elect Engn Vadlamudi 522213 India Prasad V Potluri Siddhartha Inst Technol Dept Elect & Elect Engn Kanuru 520007 India Manipal Inst Technol Manipal Acad Higher Educ Dept Elect & Elect Engn Manipal 576104 India Vignans Fdn Sci Technol & Res Guntur Andhra Pradesh India Univ Castilla La Mancha Ingn Res Grp Albacete Spain

A novel Extreme Learning Machine algorithm is used to train the neural network for Power System Stabilizer (PSS) to minimize low-frequency oscillations. The use of rapid-acting exciters, the interconnection of various power systems, and disturbances like faults and load changes all contribute to the generation of low-frequency oscillations. If sufficient damping is not provided, these oscillations generate and sustain, and eventually cause the power system to shut down entirely. The lead-lag power system stabilizer is a conventional device used but it is slow in operation and can apply to linear systems only. Artificial intelligence techniques like, fuzzy and neural networks are used to overcome the bottlenecks. The neural networks are trained using backpropagation and extreme learning algorithms. The operation of the designed power system stabilizers is verified on a 7-machine, 29-bus system, 4-machine, 11-bus system, and SMIB System. The proposed controller has been providing better damping performance compared to other controllers in terms of Integral Time Squared Error (ITSE) and Integral Squared Error (ISE). The proposed system is designed and validated through the R2023b MATLAB/Simulink Software environment.

关键词： Power System Stabilizer Power System Stability Fuzzy Controller back propagation algorithm Extreme Learning Machine algorithm

来源：评论

学校读者我要写书评

暂无评论

Research on the Prediction of Drilling Rate in Geological Core Drilling Based on the BP Neural Network

引用

APPLIED SCIENCES-BASEL 2024年第21期14卷 9959页

作者： Gong, Da Zu, Yutong Zhou, Zheng Jia, Mingrang Liu, Jiachen Hu, Yuanbiao China Univ Geosci Sch Mat Sci & Technol Beijing 100083 Peoples R China

The development of automation and intelligence in geological core drilling is not yet mature. The selection and improvement of drilling parameters rely mainly on experience, and adjustments are often made after drilling by evaluating the core, which introduces a lag and reduces the drilling efficiency. Therefore, this study first establishes a geological core drilling experiment platform to collect drilling data. Through the constructed geological core drilling experimental platform, the practical data at the drill bit can be directly obtained, solving the problem of the data from the surface equipment in the practical drilling differing from the practical data. Second, the back propagation (BP) algorithm is used to perform the ROP prediction, with weight on bit (WOB), torque (TOR), flow rate (Q), and rotation speed (RPM) as input parameters, and rate of penetration (ROP) as the output. Subsequently, correlation analysis is used to perform the feature parameter optimization, and the effects of bit wear and bit cutting depth on the experiment are considered. Finally, comparison with algorithms such as ridge regression, SVM and KNN shows that the ROP prediction model using the BP neural network has the highest prediction accuracy of 94.1%. The results provide a reference for ROP prediction and the automation of geological core drilling rigs.

关键词： geological core drilling back propagation algorithm drilling speed prediction BP neural network ROP

来源：评论

学校读者我要写书评

暂无评论

Image-based random rotation for preserving the data in data mining process

引用

SIGNAL IMAGE AND VIDEO PROCESSING 2024年第4期18卷 3893-3902页

作者： Sunil, Nadella Narsimha, G. Aditya Coll Engn & Technol Dept Comp Sci & Engn Surampalem Andhra Pradesh India JNTUH Univ Coll Engn Dept Comp Sci & Engn Sultanpur Telungana India

The privacy and security of big data have become a major concern in recent years, necessitating privacy-preserving data mining strategies to preserve the balance between data value and privacy. The application of data mining techniques to the web is known as web mining. The majority of consumers seek complete anonymity when using web apps and engaging in online activities, which raises privacy problems. Condensation, randomness, tree structure, and other traditional approaches are employed to maintain privacy. Existing techniques have limitations in that they are unable to balance data usefulness and may have privacy and scalability issues. Privacy-preserving tools such as encryption and machine learning techniques, among others, can be used to protect and classify the data stream. To overcome this, here, the data are transformed into image, and then, the random transformation is performed using enhanced particle swarm optimization. Here, the optimization is performed to identify the optimal random rotation of the data for both protection and better classification. The classification is performed through back propagation algorithm, and the perturbed data are tested against independent component analysis attack. The classification accuracy, computation time, and error rate of the classifier are measured, and it is compared with the existing method. The comparison result shows the achieved result of proposed method. This proposed system is done with the help of MATLAB 2021a.

关键词： Data mining Privacy protection Random rotation perturbation back propagation algorithm Particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

Distribution locational marginal price computation based on active power loss and emissions in active distribution system using nucleolus theory

引用

EVOLUTIONARY INTELLIGENCE 2024年第1期17卷 249-265页

作者： Veeramsetty, Venkataramana SR Engn Coll Ctr AI & Deep Learning Dept Elect & Elect Engn Warangal Telangana India

In this study, a Nucleolus Theory (NT) based iterative method has been presented to compute Distribution Locational Marginal Price (DLMP) at buses where embedded generator (EG) units were installed in the distribution network. The NT-based iterative method provides financial incentives to EG owners as per their contribution to loss reduction and emission reduction for specified loading conditions. In this study, DLMP values depend on the decision-maker preference among loss reduction, emission reduction, and distribution company additional benefit. The main objective of the proposed method is to optimize the active power loss and emissions based on DISCO decision-maker priority subjected to EG capacity. DLMP value for each EG unit is a variable that has to compute based on DISCO decision-maker priority. This proposed NT based iterative method has been implemented on Taiwan power company Distribution System (TPCDS) consisting of 84 buses with 15 EG units, and 201 bus radial distribution networks with 15 EG units under MATLAB environment and computed DLMP values based on the contribution of EG units on loss and emission reduction. It was inferred from the results that the proposed method enables EG owners to receive more financial benefit and also enables DISCO decision-makers can operate the network with more optimal in terms of loss and emissions.

关键词： Embedded generator Distribution locational marginal price Loss reduction Emission reduction Nucleolus theory back propagation algorithm Artificial neural network Distribution company

来源：评论

学校读者我要写书评

暂无评论

Rock mass quality classification based on deep learning:A feasibility study for stacked autoencoders

引用

Journal of Rock Mechanics and Geotechnical Engineering 2023年第7期15卷 1749-1758页

作者： Danjie Sheng Jin Yu Fei Tan Defu Tong Tianjun Yan Jiahe Lv Faculty of Engineering China University of GeosciencesWuhan430074China CCCC Second Harbor Engineering Compasny Ltd. Wuhan430040China

Objective and accurate evaluation of rock mass quality classification is the prerequisite for reliable sta-bility *** develop a tool that can deliver quick and accurate evaluation of rock mass quality,a deep learning approach is developed,which uses stacked autoencoders(SAEs)with several autoencoders and a softmax net *** rock parameters of rock mass rating(RMR)system are calibrated in this *** model is trained using 75%of the total database for training sample *** SAEs trained model achieves a nearly 100%prediction *** comparison,other different models are also trained with the same dataset,using artificial neural network(ANN)and radial basis function(RBF).The results show that the SAEs classify all test samples correctly while the rating accuracies of ANN and RBF are 97.5%and 98.7%,repectively,which are calculated from the confusion ***,this model is further employed to predict the slope risk level of an abandoned *** proposed approach using SAEs,or deep learning in general,is more objective and more accurate and requires less human *** findings presented here shall shed light for engineers/researchers interested in analyzing rock mass classification criteria or performing field investigation.

关键词： Rock mass quality classification Deep learning Stacked autoencoder(SAE) back propagation algorithm

来源：评论

学校读者我要写书评

暂无评论

Design of Hybrid Neural Controller for Nonlinear MIMO System Based on NARMA-L2 Model

引用

IETE JOURNAL OF RESEARCH 2023年第5期69卷 3038-3051页

作者： El Hamidi, K. Mjahed, M. El Kari, A. Ayad, H. El Gmili, N. Cadi Ayyad Univ Lab Elect Syst Energy Efficiency & Telecommun BP 549Av Abdelkarim Elkhattabi Gueliz 4000 Marrakesh Morocco Royal Sch Aeronut Math & Syst Dept Marrakech 40000 Morocco Univ Hassan II Casablanca Condensed Matter Phys & Renewable Energy Lab Mohammedia BP 146 Mohammadia 28806 Morocco

This paper introduces a nonlinear adaptive controller of unknown nonlinear dynamical systems based on the approximate models using a multi-layer perceptron neural network. The proposal of this study is to employ the structure of the Multi-Layer Perceptron (MLP) model into the NARMA-L2 structure in order to construct a hybrid neural structure that can be used as an identifier model and a nonlinear controller for the MIMO nonlinear systems. The big advantage of the proposed control system is that it doesn't require previous knowledge of the model. Our ultimate goal is to determine the control input using only the values of the input and output. The developed NARMA-L2 neural network model is tuned for its weights employing the backpropagation optimizer algorithm. Nonlinear autoregressive-moving average-L2 (NARMA-L2) neural network controller, based on the inputs and outputs from the nonlinear model, is designed to perform control action on the nonlinear for the attitude control of unmanned aerial vehicles (UAVs) model. Once the system has been modeled efficiently and accurately, the proposed controller is designed by rearranging the generalized submodels. The controller performance is evaluated by simulation conducted on a quadcopter MIMO system, which is characterized by a nonlinear and dynamic behavior. The obtained results show that the NARMA-L2-based neural network achieved good performances in modeling and control.

关键词： back propagation algorithm MIMO system MLP neural network NARMA-L2 quadcopter

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：