检索结果-内蒙古大学图书馆

Smart Systems for applications in Electrical Sciences (ICSSES), International Conference on

作者： Anuj Kumar Shveta Mahajan Dept. of Computer Science and Engineering Dr. B.R. Ambedkar National Institute of Technology Jalandhar India

ISBN: (数字)9798350364040

ISBN: (纸本)9798350364057

Nondestructive identification of seed varieties from different origins is crucial in optimizing crop improvement, plant growth, and advancement in breeding methods. Evaluating varietal purity is essential for seed quality. Traditional barley seed classification relies on visual inspection, which is a subjective and error-prone process and is time-consuming. Moreover, chemical testing involves the breaking of seeds which highlights the necessity for exploring non-destructive methods. The integration of Near-infrared (NIR) spectroscopy with neural networks offers a promising alternative. In this study, we have used a hyperspectral imaging dataset consisting of 34 different Indian barley varieties (1008 per variety). Six spectral pre-processing techniques were used to extract and pre-treat the mean reflectance spectrum associated with each seed, namely Savitzky-Golay (SG) smoothing, SG first derivative, SG second derivative, detrending, standard normal variate (SNV), and multiplicative scatter correction (MSC). Further, we have used five different classifiers: K-nearest neighbors (KNN), partial least squares discrimination analysis (PLS-DA), artificial neural networks (ANN), support vector machine techniques (SVM), and convolutional neural networks (CNN). CNN demonstrated superior classification accuracy when implementing it on unprocessed data, whereas the ANN model outperformed it when combined based on the SG2 pre-processing method. Their respective classification accuracy rates were $98.44 \%$ and $98.88 \%$. The current study examined the use of an ANN model coupled with the NIR-HSI technology to accurately, quickly, and nondestructively classify different types of barley seeds.

关键词： Support vector machines Visualization Spectroscopy Accuracy Data models Seeds (agriculture) Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

LLM4PlC: Harnessing large Language Models for Verifiable Programming of PlCs in Industrial Control Systems

LLM4PlC: Harnessing large Language Models for Verifiable Pro...

引用

IEEE/ACM International Conference on Software Engineering: Software Engineering in Practice Track (ICSE-SEIP)

作者： Mohamad Fakih Rahul Dharmaji Yasamin Moghaddas Gustavo Quiros Araya Oluwatosin Ogundare Mohammad Abdullah Al Faruque Dept. of Electrical Engineering and Computer Science University of California Irvine CA USA Siemens Technology Princeton NJ USA

ISBN: (数字)9798400705014

ISBN: (纸本)9798350352177

Although Large Language Models (LLMs) have established pre-dominance in automated code generation, they are not devoid of shortcomings. The pertinent issues primarily relate to the absence of execution guarantees for generated code, a lack of explainabil-ity, and suboptimal support for essential but niche programming languages. State-of-the-art LLMs such as GPT-4 and LLaMa2 fail to produce valid programs for Industrial Control Systems (ICS) op-erated by Programmable Logic Controllers (PLCs). We propose LLM4PLC, a user-guided iterative pipeline leveraging user feed-back and external verification tools - including grammar checkers, compilers and SMV verifiers - to guide the LLM's generation. We further enhance the generation potential of LLM by employing Prompt Engineering and model fine-tuning through the creation and usage of LoRAs. We validate this system using a FischerTech-nik Manufacturing TestBed (MFTB), illustrating how LLMs can evolve from generating structurally-flawed code to producing verifiably correct programs for industrial applications. We run a complete test suite on GPT-3.5, GPT-4, Code Llama-7B, a fine-tuned Code Llama-7B model, Code Llama-34B, and a fine-tuned Code Llama-34B model. The proposed pipeline improved the generation success rate from 47% to 72%, and the Survey-of-Experts code quality from 2.25/10 to 7.75/10. To promote open research, we share the complete experi-mental setup, the LLM Fine-Tuning Weights, and the video demonstrations of the different programs on our dedicated webpage 1 1 https://***/***/llm4plc/home.

关键词： Codes Industrial control Pipelines Programmable logic devices Programming Reliability engineering Manufacturing

来源：评论

学校读者我要写书评

暂无评论

Probabilistic Dynamic Line Rating Forecasting with Line Graph Convolutional LSTM

arXiv

引用

arXiv 2024年

作者： Kim, Minsoo Dvorkin, Vladimir Kim, Jip Dept. of Energy Engineering Korea Institute of Energy Technology Korea Republic of Dept. of Electrical Engineering and Computer Science University of Michigan Ann Arbor United States

Dynamic line rating (DLR) is a promising solution to increase the utilization of transmission lines by adjusting ratings based on real-time weather conditions. Accurate DLR forecast at the scheduling stage is thus necessary for system operators to proactively optimize power flows, manage congestion, and reduce the cost of grid operations. However, the DLR forecast remains challenging due to weather uncertainty. To reliably predict DLRs, we propose a new probabilistic forecasting model based on line graph convolutional LSTM. Like standard LSTM networks, our model accounts for temporal correlations between DLRs across the planning horizon. The line graph-structured network additionally allows us to leverage the spatial correlations of DLR features across the grid to improve the quality of predictions. Simulation results on the synthetic Texas 123-bus system demonstrate that the proposed model significantly outperforms the baseline probabilistic DLR forecasting models regarding reliability and sharpness while using the fewest parameters. Copyright © 2024, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach Towards Gas Turbine Emission Reduction by Using Neural Networks

A Novel Approach Towards Gas Turbine Emission Reduction by U...

引用

2023 International Conference on Sustainable Communication Networks and Application, ICSCNA 2023

作者： Roy, Atanu Pramanik, Sabyasachi Mitra, Kalyan Chakraborty, Manashi Sarisa West Bengal 743368 India Haldia Institute of Technology Dept. of Computer Science and Engineering Hatiberia West Bengal Haldia721657 India Indian Maritime University Dept. of Maritime Studies P - 19 Taratalla Road Kolkata700088 India Sarisa West Bengal 743368 India

ISBN: (纸本)9798350313987

Optimizations of controlling parameters are the key factors to achieve effectual output and emission reduction from machinery running. This study relates prediction technology for gas turbine's (GT's) running optimization and emission reduction. The tool can identify upcoming disputes and alter the engine control setting to achieve highly efficient running operations. This research objects to get a machine learning approach to predict GT performance on emissions. The forecasting capabilities of this system are based on real-time data collected from a GT plant. A neural network is employed to analyze and predict the emissions of exhaust pollutants, primarily carbon monoxide and nitrogen oxides. The model performance is studied by considering the values of MSE, MAE, and residuals. The model's accuracy and precision are revealed by its low MSE (0.0035) and MAE (0.043). This approach is a cost-effective way to gage emissions accurately and can be used to confirm compliance with environmental regulations. It is also a helpful tool for monitoring the health of the turbine and identifying any potential issues with its operation. © 2023 IEEE.

关键词： Convolutional Neural Networks Emission Gas Turbine Predictive Emission Monitoring

来源：评论

学校读者我要写书评

暂无评论

Face Recognition from Partial Face Data and Fraud Detection

Face Recognition from Partial Face Data and Fraud Detection

引用

International Conference on Computational Intelligence and Communication Networks (CICN)

作者： Thavisala Veneela Sonti Vivek Sai Yeturu Vamshidhar Reddy Vishnu Karthik Varma Kothapally dept. of Assistant professor/computer science and engineering B V RAJU INSTITUTE OF TECHNOLOGY Narsapur Medak Telangana India

ISBN: (数字)9798331505264

ISBN: (纸本)9798331505271

Face recognition is a critical era with applications starting from safety to personalization. However, traditional face Recognition systems frequently conflict with partial face information, which can occur because of occlusions or picture excellent g troubles. I t consists of a novel technique to stand reputation using partial face statistics, specializing in fraud detection scenarios. This approach leverages deep getting to know strategies, mainly convolutional neural networks (CNNs), to extract functions from partial face snap shots. These features are then used to create a compact illustration of the face, which is robust to occlusions and noise. A similarity measure is employed to evaluate those representations, enabling the identification of individuals even from partial face records. In the context of fraud detection, it can be used to verify the identity of people based totally on partial face data captured from surveillance cameras or different sources. Experimental consequences on benchmark datasets show the effectiveness of the proposed approach, outperforming conventional face recognition procedures in eventualities with partial face information. It gives a promising solution for improving face recognition accuracy in challenging conditions, with capability packages in security, regulation enforcement, and personal identification systems. It also analyzes facial images with masks using Multi-Task Cascaded Convolutional Neural Networks (MTCNN). FaceNet algorithm is also used which adds more embeddings and verifications to face recognition. Support vector Machine (SVM) algorithm labels the data sets to produce a reliable prediction probability and along with that it also detects the frauds.

关键词： Support vector machines Accuracy Face recognition Surveillance Prediction algorithms Vectors Robustness Fraud Safety Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Wild Animal Recognition: A Vocal Analysis

Wild Animal Recognition: A Vocal Analysis

引用

International Conference on Computing for Sustainable Global Development (INDIACom)

作者： P Sudharshan Duth N Manohar R Suresha M Priyanka Tian Jipeng Dept. of Computer Science Amrita Vishwa Vidyapeetham Amrita School of Computing Mysuru Campus Mysuru India Computer Science College Zhongyuan University of Technology Zhengzhou Henan Province China

The Animal Voice Recognition system has become an important application in the field of animal sound retrieval and classification; it is highly beneficial for applications involving bio-acoustics and audio retrieval. Currently existing work lacks wild animal vocal dataset and requires appropriate vocal feature engineering mechanism even in the presence of varied background noise is necessary to conduct vocal analysis for efficient recognition. In this research work, initially wild animal vocal database was created, then fusion of vocal features were extracted for developing the animal recognition system based on Multiple vocal features such as the Mel-Frequency Cepstral Coefficients (MFCC), Zero-Cross-Rate (ZCR), Spectral Roll Off, Spectral Centroid and Spectral Contrast features. Further, PCA is used to extract the transformed discriminating features. Finally, the performance of this model has been analyzed by extracting features of the collected dataset using number of conventional classifiers to identify the specific animal. Among which SVM and Random Forest classifier obtained 96.89% and 97.46% recognition accuracy respectively and it proves to be cost-effective.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Secure Host Login Technique based Key Agreement protocol for Software Defined Network 4

Secure Host Login Technique based Key Agreement protocol for...

引用

4th International Conference on Inventive Research in Computing Applications, ICIRCA 2022

作者： Bhardwaj, Aditya Mutaher, Hamza International Inistitute of Technology Dept. Computer Science and Engineering Bhubaneswar India School of Computing and Innovative Technologies British University Vietnam Hanoi Viet Nam

ISBN: (数字)9781665497077

ISBN: (纸本)9781665497077

Scalability is one of the Software Defined Network features that makes it more efficient than the traditional network. Due to this feature, a huge number of hosts connect to the network and require network rules and policies from the network controller. The huge number of connected hosts makes the network vulnerable to unauthorized access, which may cause network damage. In this paper, a secure hosts login technique is proposed to assure the authenticity of the hosts that need to access the network and use its resources. The proposed technique uses the key agreement protocol to allow the hosts and controller to establish secure communication. Then the procedure moves to the registration phase and login phase, which are responsible for creating and verifying the hosts' credentials while claiming network access. The secrecy and authenticity of the proposed technique was simulated, verified, and proved using AVISPA tool. The result of the simulation also proved the immunity of the proposed technique against MITM and replay attacks. © 2022 IEEE.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

DeepStock Forecast: Unveiling Market Movements through Advanced Deep Learning Models

DeepStock Forecast: Unveiling Market Movements through Advan...

引用

Emerging Systems and Intelligent Computing (ESIC), International Conference on

作者： Smruti Rekha Das Debahuti Mishra Arundhati Lenka Kailash Shaw Dept. Computer Science and Engineering GITAM Deemed to be University Visakhapatnam Andhra Pradesh India Dept. Computer Science and Engineering Siksha ‘O’ Anusandhan Deemed to be University Odisha India Odisha India dept. of AIML Symbiosis Institute of Technology Symbiosis International (Deemed University) Pune India

This study uses deep neural network (DNN) methodologies, including auto encoder, deep belief network (DBN), and backpropagation neural network (BPNN), to predict stock prices over 15 and 30 days. Utilizing BSE Sensex and NSE Sensex datasets with diverse technical indicators, the research highlights DNN’s superior training accuracy, measured by minimal mean squared error (MSE). The study’s key contribution is applying DNN to predict stock market volatility. While acknowledging the limitation of relying solely on training error, the study incorporates additional measures like RMSE, MAPE, MAE, and ARV during testing. Notably, DNN’s MAPE results fall within a narrow range (0.0221 to 0.0255), indicating minimal error deviation. In a comprehensive evaluation across all datasets, DNN consistently outperforms DBN and BPNN in both training and testing performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Based Hybrid Technique for Heart Disease Prediction

Machine Learning Based Hybrid Technique for Heart Disease Pr...

引用

2022 International Conference on Advances in Computing, Communication and Materials, ICACCM 2022

作者： Negi, Sourav Singh Memoria, Minakshi Kumar, Rajiv Joshi, Kapil Pandey, Shiv Dayal Gupta, Ashulekha Uttaranchal Institute of Technology Uttaranchal University Department of Computer Science and Engineering Dehradun India Graphic Era University Dept. of Management Studies Dehradun India

ISBN: (纸本)9781665474399

There is a lot of information in the medical services industry. With such a big amount of data, the illness can often be identified, predicted, or reduced. Infections such as cardiovascular sickness, malignant development, tumours, and so on pose a significant threat to humanity. In this research, we try to focus on coronary sickness prediction using AI approaches. Coronary illness is frequently predicted. Information such as pulse, hypertension, diabetes, and cigarette smoking is collected, and these highlights are then displayed for forecasting. The calculations like K-nearest neighbor, Random Forest and Decision tree are used. We have also proposed a hybrid model of combining Decision Tree and Random Forest. The precision of the model is to investigate utilizing every one of the calculations. At that point with the more exactness is taken as a result of the model for anticipating the daringness infection. © 2022 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Optimizing Age-of-Information in Adversarial Environments with Channel State Information 14

Optimizing Age-of-Information in Adversarial Environments wi...

引用

14th International Conference on COMmunication Systems and NETworkS, COMSNETS 2022

作者： Mandal, Avijit Bhattacharjee, Rajarshi Sinha, Abhishek Indian Institute of Technology Madras Dept. of Electrical Engineering Chennai600036 India University of Massachusetts at Amherst College of Information and Computer Sciences United States

ISBN: (纸本)9781665421041

This paper considers a multi-user downlink scheduling problem with access to the channel state information at the transmitter (CSIT) to minimize the Age-of-Information (AoI) in a non-stationary environment. The non-stationary environment is modelled using a novel adversarial framework. In this setting, we propose a greedy scheduling policy, called MA-CSIT, that takes into account the current channel state information. We establish a finite upper bound on the competitive ratio achieved by the MA-CSIT policy for a small number of users and show that the proposed policy has a better performance guarantee than a recently proposed greedy scheduler that operates without CSIT. In particular, we show that access to the additional channel state information improves the competitive ratio from 8 to 2 in the two-user case and from 18 to 8/3 in the three-user case. Finally, we carry out extensive numerical simulations to quantify the advantage of knowing CSIT in order to minimize the Age-of-Information for an arbitrary number of users. © 2022 IEEE.

关键词： Channel state information

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：