检索结果-内蒙古大学图书馆

Using Transfer Learning and xgboost for Early Detection of Fires in Offshore Wind Turbine Units

ENERGIES 2024年第10期17卷 2330页

作者： Wan, Anping Du, Chenyu Gong, Wenbin Wei, Chao AL-Bukhaiti, Khalil Ji, Yunsong Ma, Shidong Yao, Fareng Ao, Lizheng Hangzhou City Univ Dept Mech Engn Hangzhou 310015 Peoples R China Huadian Elect Power Res Inst Hangzhou 310030 Peoples R China Southwest Jiaotong Univ Sch Civil Engn Chengdu 610031 Peoples R China Guangdong Huadian Fuxin Yangjiang Offshore Wind Po Yangjiang 529500 Peoples R China

To improve the power generation efficiency of offshore wind turbines and address the problem of high fire monitoring and warning costs, we propose a data-driven fire warning method based on transfer learning for wind turbines in this paper. This paper processes wind turbine operation data in a SCADA system. It uses an extreme gradient-boosting tree (xgboost) algorithm to build an offshore wind turbine unit fire warning model with a multiparameter prediction function. This paper selects some parameters from the dataset as input variables for the model, with average cabin temperature, average outdoor temperature, average cabin humidity, and average atmospheric humidity as output variables. This paper analyzes the distribution information of input and output variables and their correlation, analyzes the predicted difference, and then provides an early warning for wind turbine fires. This paper uses this fire warning model to transfer learning to different models of offshore wind turbines in the same wind farm to achieve fire warning. The experimental results show that the prediction performance of the multiparameter is accurate, with an average MAPE of 0.016 and an average RMSE of 0.795. It is better than the average MAPE (0.051) and the average RMSE (2.020) of the prediction performance of a backpropagation (BP) neural network, as well as the average MAPE (0.030) and the average RMSE (1.301) of the prediction performance of random forest. The transfer learning model has good prediction performance, with an average MAPE of 0.022 and an average RMSE of 1.469.

关键词： transfer learning xgboost algorithm offshore wind turbines data-driven fire warning

来源：评论

学校读者我要写书评

暂无评论

Simulation of LUCC Scenarios and Analysis of the Driving Force of Carbon Stock Supply Changes in the North China Plain in the Context of Urbanization

引用

FORESTS 2024年第8期15卷 1414页

作者： Ma, Dongling Huang, Qingji Wang, Qian Lin, Zhenxin Xu, Hailong Shandong Jianzhu Univ Sch Surveying & Geoinformat Jinan 250101 Peoples R China Yunnan Univ Inst Int Rivers & Ecosecur Kunming 650091 Peoples R China

The North China Plain is the core region of China's economic development, and exploring the impacts of its land use and cover change (LUCC) and different urbanization regional drivers on carbon stocks is conducive to promoting sustainable development and carbon balance within the region. In the study, the North China Plain was selected as the study area, and the Patch-Generating Land Use Simulation (PLUS) model and Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model were comprehensively applied to set up three land use policies, predict land use changes in 2030, and calculate carbon stock changes. Meanwhile, the Extreme Gradient Boosting (xgboost) algorithm was used to analyze the degree of influence of different drivers on the supply of carbon stocks in different urbanization regions. Studies show that if the North China Plain prioritizes economic development, the area of farmland and forests will significantly decrease, leading to a substantial decline in carbon stocks. If ecological protection is the development focus, the reduction in farmland and forests will be less, and carbon stocks will remain relatively stable. If farmland protection is the development focus, the reduction in farmland will be minimal, but there will still be some impact on carbon stocks. The driving forces of carbon stock supply vary significantly across different regions. In underdeveloped regions, population density and vegetation cover have a greater impact on carbon stocks. In developing and urban-rural combined regions, vegetation cover and population migration have a greater impact on carbon stocks. In developed regions, the area of artificial land and gross domestic product (GDP) have a greater impact on carbon stocks. The study results provide scientific evidence for regional land use planning and policy formulation.

关键词： North China Plain urban expansion PLUS-InVEST model xgboost algorithm driving factors analysis

来源：评论

学校读者我要写书评

暂无评论

Toward a comprehensive life cycle aquatic ecotoxicity assessment via machine learning: Application to coal power generation in China

引用

JOURNAL OF CLEANER PRODUCTION 2024年 445卷

作者： Li, Danyu Qin, Ji Hong, Jinglan Hong Kong Polytech Univ Dept Elect & Informat Engn Hong Kong 99907 Peoples R China Engn Univ PAP Sch Cryptog Engn Wujing Rd 1 Xian 710016 Peoples R China Shandong Univ Sch Environm Sci & Engn Shandong Prov Key Lab Water Pollut Control & Resou Shanda South Rd 27 Jinan 250100 Peoples R China

Life cycle aquatic ecotoxicity assessments currently lack characterization factors due to the limited availability of experimental toxicity data. In this study, this problem was solved by designing a rapid and simplified approach using machine learning algorithms. The results were further tested and validated via a case study of coal power generation in China that has been implemented since 1990. Compared with the predicted characterization factors of life cycle aquatic ecotoxicity, the predicted average ecotoxicity values (logEC50) are clearly correlated (R2 = 0.597 and RMSE = 0.64) with the reported value in USEtox database based on chemicals' physicochemical properties, degradation rates, and bioaccumulation factors in fish. Approximately 13.6% of the ecotoxicity impact of China's coal power generation is underestimated due to the missing characterization factors of chemicals. In the case study, the predicted ecotoxicity value of the newly identified essential substance via the xgboost algorithm for the case study was close to the actual value based on the experimental data. Intensifying atmospheric pyrene and chrysene control during the coal power generation stage, increasing the utilization rate of thermal power equipment, and optimizing coal consumption efficiency are efficient approaches for reducing the overall aquatic ecotoxicity from coal power generation in China. The proposed approach can rapidly predict ecotoxicity values and calculate missing characterization factors. Consequently, life cycle aquatic ecotoxicity assessments can be improved, and management strategies can be developed.

关键词： Life cycle aquatic ecotoxicity assessment Characterization factors Average ecotoxicity values xgboost algorithm Coal power

来源：评论

学校读者我要写书评

暂无评论

Prediction of corrosion fatigue crack growth rate in aluminum alloys based on incremental learning strategy

引用

INTERNATIONAL JOURNAL OF FATIGUE 2024年 187卷

作者： Peng, Yufeng Zhang, Yongzhen Zhang, Lijun Yao, Leijiang Guo, Xingpeng Guangzhou Univ Joint Inst Guangzhou 510006 Peoples R China Guangzhou Univ Inst Corros Sci & Technol Guangzhou 510006 Peoples R China Inst Corros Sci & Technol Guangzhou 510530 Peoples R China Northwestern Polytech Univ Sch Comp Sci Xian 710065 Peoples R China Univ Sci & Technol Beijing Natl Ctr Mat Serv Safety Beijing 100083 Peoples R China Northwestern Polytech Univ Lab Sci & Technol UAV Xian 710065 Peoples R China Guangzhou Univ Sch Chem & Chem Engn Guangzhou Key Lab Clean Energy & Mat Guangzhou 510006 Peoples R China

The 2xxx series aerospace aluminum alloys, vital for aircraft structures due to their excellent mechanical strength and corrosion resistance, face challenges with corrosion fatigue failure, impacting material integrity and aircraft safety. Accurately predicting the corrosion fatigue crack growth rate is crucial for these materials' reliability in aerospace applications, ensuring safety and operational efficiency. Various unique incremental learning strategies like structural optimization, regularization, and replay were used to enhance the predictive model's adaptability to continually updated data without the need to retrain the entire model, thus overcoming "catastrophic forgetting." Based on these strategies, a series of xgboost models for the prediction of corrosion fatigue crack growth rate in aerospace aluminum alloys have been developed. The results demonstrate that the replay strategy excels in maintaining the accuracy of historical data, while structural optimization and regularization strategies stand out in adapting to new data and effectively preventing overfitting. Combining these three strategies, the models maintain good accuracy and a low forgetting rate across all steps, with an accuracy rate exceeding 0.95 and a forgetting rate below 0.05 in all incremental steps, showcasing excellent knowledge retention capabilities. The models achieve a balanced performance in learning new tasks and retaining old knowledge.

关键词： Aluminum alloys Corrosion fatigue crack growth rate Incremental learning Catastrophic forgetting xgboost algorithm

来源：评论

学校读者我要写书评

暂无评论

Tool wear state recognition and prediction method based on laplacian eigenmap with ensemble learning model

引用

ADVANCED ENGINEERING INFORMATICS 2024年 60卷

作者： Xie, Yang Gao, Shangshang Zhang, Chaoyong Liu, Jinfeng Jiangsu Univ Sci & Technol Sch Mech Engn Zhenjiang 212000 Peoples R China Huazhong Univ Sci & Technol Sch Mech Sci & Engn Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol State Key Lab Intelligent Mfg Equipment & Technol Wuhan 430074 Peoples R China

Accurate prediction of tool wear status plays a critical role in the digital manufacturing industry, and its health level directly affects machining quality, production costs, and overall productivity. In response to the problems of the high dimensionality of extracted features from tool wear characterization sensors, redundant information, and large individual model errors and biases, a novel method for tool wear status identification and prediction that fuses downscaling dimensionality and ensemble models is proposed. First, a multi-algorithm feature filtering based on Random Forest (RF) and extreme gradient boosting (xgboost) is utilized, and the laplacian eigenmaps (LE) algorithm is combined to perform fusion downscaling on the filtered features. Then, the parameters of the xgboost algorithm are optimized using grid search (GS). Finally, the performance of the proposed method is evaluated by different tool wear experiments for both regression and classification using model prediction accuracy evaluation metrics (R-squared values and f1 values) and prediction time. The experimental results show that for different tools wear experimental data, the R-squared values of the regression model are higher than 0.98, the f1 values of the classification model are above 0.96, and the prediction speed is improved by an order of magnitude compared with other models. The results are analyzed to verify the effectiveness and applicability of the proposed method, which can provide technical support for the automated machining process.

关键词： Tool wear Laplacian eigenmaps Features fusion xgboost algorithm Ensemble method

来源：评论

学校读者我要写书评

暂无评论

Dynamic Forecasting of Solar Energy Microgrid Systems Using Feature Engineering

Dynamic Forecasting of Solar Energy Microgrid Systems Using ...

引用

IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES)

作者： Mohamed, Muamar Mahmood, Farhad E. E. Abd, Mehmmood A. A. Chandra, Ambrish Singh, Bhim Univ Quebec Ecole Technol Super Dept Elect Engn Montreal PQ H3C 1K3 Canada Univ Mosul Dept Elect Engn Mosul 00964 Iraq IBM Corp Austin TX 78758 USA Indian Inst Technol Delhi New Delhi 110016 India

The intermittent and stochastic nature of solar energy generation systems, climate change, and the inefficiency of modern power systems due to zero inertia have created many challenges for on-grid operators. Solar forecasting systems based on machine learning algorithms are an emerging and effective solution that uses Big Data related to weather phenomena. However, the predictive ability of these algorithms is hampered by the sporadic nature of solar energy generation. In this article, a robust hybrid machine learning system that utilizes multiple linear regression (MLR) and a Pearson correlation coefficient (PCC) was tested on solar power plant sites of varying capacities in Germany (100-8500 kW). The volume of Big Data features can be reduced by focusing on the features that significantly improve the reliability of the mid-term forecasting system. In this way, drastic fluctuations in the prediction of photovoltaic (PV) power generation can be avoided. The results of our approach are evaluated regarding real-world data using the extreme gradient boosting (xgboost) with feature engineering, and principal component analysis (PCA), in order to forecast PV energy, rank, and track the importance of feature engineering for different PV capacities. Furthermore, we found that the need for selectivity and reduction of performance error was supported by ridge regression. In addition, the proposed novel xgboost forecast system decreased the root-mean-square error (RMSE) and mean absolute error (MAE) by 30% and 18%, respectively, compared to the Autoencoder and long short-term memory (LSTM) for same datasets. Furthermore, the CoD determination coefficient (R-2) increased by 85% compared to the statistical model's autoregressive integrated moving average (ARIMA).

关键词： Artificial intelligence data mining algorithms forecasting techniques machine learning power system statistical analysis solar microgrid xgboost algorithm

来源：评论

学校读者我要写书评

暂无评论

xgboost based solutions for detecting fraudulent credit card transactions

XGBoost based solutions for detecting fraudulent credit card...

引用

International Conference on Advanced Creative Networks and Intelligent Systems (ICACNIS) - Blockchain Technology, Intelligent Systems, and the Applications for Human Life

作者： El Kafhali, Said Tayebi, Mohammed Hassan First Univ Settat Fac Sci & Tech IR2M Lab Settat Morocco

ISBN: (纸本)9798350334449

Due to the emigration of businesses to the internet, credit cards have become a widely used tool for customers to pay for purchases both online and offline. However, fraudsters try to attack these payment tools using various techniques, and credit card fraud transactions have become dangerous. Credit card payment security is becoming an important research topic, and different ways are proposed to overcome this problem. In this article, we will propose a solution that combines the strengths of three methods to secure credit card fraud transactions. The Differential Evolution (DE) algorithm is used to select the best hyperparameters of the xgboost algorithm (eXtreme Gradient Boosting). Synthetic Minority Oversampling (SMOTE) combined with Edited Nearest Neighbour (ENN) to handle the imbalanced data issues. Finally, we will classify fraud transactions using the optimized xgboost algorithm. The proposed solution is evaluated using the European data set in terms of Accuracy, Recall, F1-score, and AUC. As a result, the proposed solution generally achieved the best score in terms of all the measurements used in this study. In terms of accuracy, we achieved 99.94%, similarly, we reached the best score in terms of the Precision score, which is 80.68%, in addition, we got 86.02% of recall score, 83.27 of F-measure, and 99.21 of AUC score. Thus, we emphasize the superiority of our proposed method over the competitive machine learning model utilized in this work.

关键词： Fraud Transactions Detection Hyperparameters Optimization Differential Evolution algorithm xgboost algorithm Synthetic Minority Oversampling Edited Nearest Neighbor

来源：评论

学校读者我要写书评

暂无评论

Research on Guangxi Yulin's Poverty Return Identification Based on Integrated Learning

Research on Guangxi Yulin's Poverty Return Identification Ba...

引用

IEEE International Conference on Electrical Engineering, Big Data and algorithms (EEBDA)

作者： Long Jing Pan Xiuzhi Zhao Qianrong Li Junjiang Nanning Univ Coll Accounting & Auditing Nanning Guangxi Peoples R China Guangxi Univ Foreign Languages Sch Accounting Nanning Guangxi Peoples R China

ISBN: (纸本)9781665416061

2020 is the end of the poverty alleviation exit effectiveness inspection. To consolidate the achievements of poverty alleviation, promote the Rural Revitalization Strategy, and prevent poverty return will be its key means and important guarantee. At present, although many experts and scholars have studied the related fields of dynamic monitoring of poverty return, most of them focus on macro policy content, and the research on poverty return identification for the poor is still relatively lacking. At the same time, the in-depth mining and application of the accumulated large-scale targeted poverty alleviation data is insufficient. This paper finds relevant literature and combines China's achievements in precise poverty alleviation work and the construction of precise poverty alleviation systems, using Yulin's poor household record-keeping data as the research object. The three integrated learning algorithms, xgboost, LightGBM and CatBoost, were used to identify poor households in three categories, out of poverty, poor and returning to poverty. The results show that the xgboost algorithm model achieves optimal results, with an overall accuracy of 95.55% for three classification recognition, and can better identify the poor categories of the poor population. At the same time, the model's feature importance ranking can reveal the importance of povertycausing characteristics of poor households in Fuping County. By analyzing the current work flow of helpers, it is clear that good dynamic monitoring of poverty return should achieve the needs of dynamic updating of data, dynamic feedback of results and dynamic adjustment of countermeasures.

关键词： Integrated learning Rural revitalization Poverty return identification xgboost algorithm Guangxi

来源：评论

学校读者我要写书评

暂无评论

A machine learning method based on TPE-xgboost model for TRIP/TWIP near-β titanium alloy design

引用

MATERIALS TODAY COMMUNICATIONS 2024年 41卷

作者： Wang, Zheng Zhang, Haoyu Wang, Shengyuan Cheng, Jun Wang, Chuan Zhou, Ge Zhang, Siqian Chen, Lijia Shenyang Univ Technol Sch Mat Sci & Engn Shenyang 110870 Liaoning Peoples R China Shenyang Univ Technol Shenyang Key Lab Adv Struct Mat & Applicat Shenyang 110870 Liaoning Peoples R China Northwest Inst Nonferrous Met Res Shaanxi Key Lab Biomed Met Mat Xian 710016 Peoples R China Liaoning Univ Coll Light Ind Shenyang 110036 Liaoning Peoples R China

The traditional design method for TRIP/TWIP near-beta titanium alloys is time-consuming and expensive. This paper proposes a machine-learning design method for a near-beta titanium alloy that uses the Tree-structured Parzen Estimator (TPE) algorithm to optimize the hyperparameter space of xgboost for improved model prediction accuracy in TRIP/TWIP. The model predicts the mechanical properties and plasticity mechanisms of the alloy based on its composition, key empirical parameters derived from the composition, and solution treatment temperature. The test results of the TPE-xgboost model showed that when predicting the ultimate tensile strength (UTS), the adjusted determination coefficient (Adjusted R-2), root mean square error (RMSE), and mean absolute error (MAE) values were 0.9264, 39.63 MPa and 44.83 MPa, respectively. When predicting elongation (El), the Adjusted R-2, RMSE, and MAE values are 0.9134, 1.97 %, and 1.89 %, respectively. When predicting the plastic mechanisms of the alloy, the F1-score, Recall, Precision, Accuracy, and Area Under the ROC Curve (AUC) were 0.74, 0.75, 0.80, 0.78, and 0.83, respectively. A near-beta titanium alloy, Ti-8V-4Mo-3.5Al-3Cr-2Zr (wt%), along with an associated heat treatment process, has been developed using this model. The experimental results have demonstrated that the model accurately predicts the strength, elongation, and plasticity mechanisms of the alloys. The TRIP/TWIP effects occur due to {332}<113> twinning and secondary alpha" martensite formation during tensile deformation. This study offers an alternative approach for designing TRIP/TWIP near-beta titanium alloys.

关键词： Machine learning Near-beta titanium alloy TRIP TWIP xgboost algorithm

来源：评论

学校读者我要写书评

暂无评论

Sequence Alignment Using Machine Learning-Based Needleman-Wunsch algorithm

引用

IEEE ACCESS 2021年 9卷 109522-109535页

作者： El-Din Rashed, Amr Ezz Amer, Hanan M. El-Seddek, Mervat El-Din Moustafa, Hossam Mansoura Univ Dept Commun & Elect Fac Engn Mansoura 35516 Egypt Taif Univ Coll Comp & Informat Technol At Taif 26571 Saudi Arabia Higher Inst Engn & Technol Mansoura 35516 Egypt

Biological pairwise sequence alignment can be used as a method for arranging two biological sequence characters to identify regions of similarity. This operation has elicited considerable interest due to its significant influence on various critical aspects of life (e.g., identifying mutations in coronaviruses). Sequence alignment over large databases cannot yield results within a reasonable time, power, and cost. heuristic methods, such as FASTA, the BLAST family have been demonstrated to perform 40 times faster than DP-based (e.g., Needleman-Wunsch) techniques they cannot guarantee an optimum alignment result An optimized software platform of a widely used DNA sequence alignment algorithm called the Needleman-Wunsch (NW) algorithm based on a lookup table, is described in this study. This global alignment algorithm is the best approach for identifying similar regions between sequences. This study presents a new application of classical machine learning (ML) to global sequence alignment. Customized ML models are used to implement NW global alignment. An accuracy of 99.7% is achieved when using a multilayer perceptron with the ADAM optimizer, and up to 2912 Giga cell updates per second are realized on two real DNA sequences with a length of 4.1 M nucleotides. Our implementation is valid for RNA/DNA sequences. This study aims to parallelize the computation steps involved in the algorithm to accelerate its performance by using ML algorithms. All datasets used in this study are available from https://***/documents/dna-sequence-alignment-datasets-based-nw-algorithm.

关键词： DNA Machine learning algorithms Heuristic algorithms Classification algorithms RNA Graphics processing units Approximation algorithms Bioinformatics DNA RNA pairwise sequence alignment (PWSA) Needlema-Wunsch (NW) algorithm machine learning (ML) algorithms multilayer perceptron (MLP) xgboost algorithm

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：