检索结果-内蒙古大学图书馆

Ensemble learning predicts glass-forming ability under imbalanced datasets

COMPUTATIONAL MATERIALS SCIENCE 2025年 248卷

作者： Cheng, Duan-jie Liang, Yong-chao Pu, Yuan-wei Chen, Qian Guizhou Univ Coll Adv Optoelect Mat & Technol Sch Big Data & Informat Engn Guiyang 550025 Peoples R China

With the development of artificial intelligence, machine learning (ML) is widely used to predict glass-forming ability (GFA). However, GFA experimental data usually exhibits a long-tailed distribution, and the similarity between the enhanced dataset and the original dataset is unclear. In terms of modeling, although model fusion provides better prediction results than individual learners, it also faces the risk of overfitting. Therefore, two preprocessing methods designed for regression problems WEighted Relevance-based Combination Strategy (WERCS) and Synthetic Minority Over-sampling technique with Gaussian Noise (SMOGN) are employed. The best strategy is selected by Pairwise correlation difference (PCD). Based on the screening results, this paper further proposes a multi-layer stacking ensemble learning model (MLS) for predicting GFA. Considering model accuracy and diversity together, the base model and meta-model combinations are optimized by bayesian optimization algorithm (BOA). The results show that MLS achieves R2 = 0.79 in prediction accuracy, which is better than other models and criteria discussed in this paper. In addition, the generalization ability of the MLS model is verified in the Cu-Mg-Ca alloy system. To explain the MLS model, SHapley Additive exPlanation (SHAP) is introduced. With the help of MLS and SHAP methods, the formation law of bulk metallic glasses (BMGs) is revealed, and the BMGs of Zr-Cu-Al-Ag series alloys are successfully designed.

关键词： Glass-forming ability Data Enhancement Ensemble learning bayesian optimization algorithm Design of bulk metallic glasses

来源：评论

学校读者我要写书评

暂无评论

Holographic memory-based bayesian optimization algorithm (HM-BOA) in dynamic environments

引用

Science China(Information Sciences) 2013年第9期56卷 110-126页

作者： KAEDI Marjan GHASEM-AGHAEE Nasser AHN Chang Wook Department of Information technology Engineering University of Isfahan Department of Computer Engineering University of Isfahan Department of Computer Engineering Sungkyunkwan University

This paper presents a new evolutionary dynamic optimization algorithm, holographic memory-based bayesian optimization algorithm (HM-BOA), whose objective is to address the weaknesses of sequential memory-based dynamic optimization approaches. To this end, holographic associative neural memory is applied to one of the recent successful memory-based evolutionary methods, DBN-MBOA (memory-based BOA with dynamic bayesian networks). Holographic memory is appropriate for encoding environmental changes since its stimulus and response data are represented by a vector of complex numbers such that the phase and the magnitude denote the information and its confidence level, respectively. In the learning process in HM-BOA, holographic memory is trained by probabilistic models at every environmental change. Its weight matrix contains abstract information obtained from previous changes and is used for constructing a new probabilistic model when the environment changes. The unique features of HM-BOA are: 1) the stored information can be generalized, and 2) a small amount of memory is required for storing the probabilistic models. Experimental results adduce grounds for its effectiveness especially in random environments.

关键词： dynamic optimization bayesian optimization algorithm holographic memory dynamic bayesian networks.

来源：评论

学校读者我要写书评

暂无评论

bayesian algorithm optimized deep model for multi-signal fault identification of hydraulic pumps

引用

ALEXANDRIA ENGINEERING JOURNAL 2025年 121卷 465-483页

作者： Tang, Shengnan Jiang, Yixuan Zhou, Tao Zhu, Yong Lim, Kian Meng Jiangsu Univ Inst Adv Mfg & Modern Equipment Technol Sch Mech Engn Zhenjiang 212013 Peoples R China Zhejiang Univ State Key Lab Fluid Power & Mechatron Syst Hangzhou 310013 Peoples R China Natl Univ Singapore Dept Mech Engn Singapore 117575 Singapore Jiangsu Univ Natl Res Ctr Pumps Zhenjiang 212013 Peoples R China LEO Grp Co Ltd Wenling 317500 Peoples R China

Hydraulic pump (HP) supplies the power for a hydraulic transmission system. It is critical for the fault diagnosis of HP to guarantee the healthy operation of systems. The rapid progress of artificial intelligence and measuring technique empowers data-driven methods. Diagnosis methods based on machine learning are limited in timeconsuming parameter adjustment. Moreover, current methods based on one measured signal achieve fault diagnosis only by using limited feature information. To address the two problems, this research proposes an intelligent fault diagnosis method for HP. First, original signals from multiple sources are obtained using multiple sensors. Due to the effective solution of frequency aliasing and the good robustness to the background noise of the signals, an intelligent method combining an improved deep model and synchrosqueezing wavelet transform is further established. The introduction of bayesian algorithm completes the adaptive selection of main parameters and contributes to reduce the computing time. The optimal model is selected by updating hyperparameter groups and evaluating the performance. The optimized diagnostic model presents good convergence and stability. The diagnostic accuracies achieve 96.69 %, 97.58 %, and 99.89 % for three signals. The proposed diagnostic method presents the effectiveness in mining fault characteristic information and intelligent identification of common fault types.

关键词： Hydraulic pump Intelligent defect identification Deep model bayesian optimization algorithm Synchrosqueezing wavelet transform

来源：评论

学校读者我要写书评

暂无评论

Hybrid machine learning modeling of nitrogen removal from wastewater using gas-liquid-solid circulating fluidized bed riser

引用

CHEMICAL ENGINEERING RESEARCH & DESIGN 2024年 207卷 295-307页

作者： Razzak, Shaikh Abdur Sultana, Nahid Hossain, S. M. Zakir Rahman, Muhammad Muhitur Yuan, Yue Hossain, Mohammad Mozahar Zhu, Jesse King Fahd Univ Petr & Minerals Dept Chem Engn Dhahran 31261 Saudi Arabia King Fahd Univ Petr & Minerals Interdisciplinary Res Ctr Refining & Adv Chem Dhahran Saudi Arabia Imam Abdulrahman Bin Faisal Univ Coll Comp Sci & Informat Technol Dept Comp Sci Dammam Saudi Arabia Univ Bahrain Coll Engn Dept Chem Engn Zallaq Bahrain King Faisal Univ Dept Civil & Environm Engn Al Hufuf Saudi Arabia Western Univ Dept Chem & Biochem Engn London ON Canada

Municipal wastewater generally contains high amounts of nitrogen (N), approximately 23-28 mg/L of total Kjeldahl Nitrogen (TKN), which causes eutrophication and pollution of groundwater. This article reports the generation of empirical models for estimating the nitrogen removal efficacy of two outputs (NH4-N 4-N and NO3-N) 3-N) from the anoxic riser of a highly efficient municipal wastewater treatment system involving a gas-liquid-solid circulating fluidized bed (GLSCFB) bioreactor. To identify the important variables and discard irrelevant ones, the ReliefF algorithm is applied initially. Out of 10 independent variables, this method indicates that only three, such as flow rate, total Kjeldahl Nitrogen (TKN), and downer effluent of NO3-N, are meaningful for anoxic nitrogen removal. Then, a hybrid bayesian optimization algorithm and Support Vector Regression (BOA-SVR) is implemented, considering only the essential variables. In this regard, real-life laboratory data are utilized to develop the models. The BOA approach and k-fold cross-validation technique are explicitly applied to optimize the model's hyperparameters and avoid overfitting. The established models provided sufficiently accurate predictions via their comparisons to the experimental observations. Besides, the BOA-SVR model outperforms the classical multiple linear regression (MLR). The BOA-SVR models' forecasted results for both the predicted amount of NH4-N 4-N and NO3-N 3-N are favorably compared with the laboratory trials (R2> 2 > 99 %). A comprehensive evaluation is further conducted to validate the performance of this model by calculating the root mean square error, mean absolute percentage error, fractional bias, and computational efficiency. Gaussian white noise is used to add three distinct noise levels (20 %, 40 %, and 60 %) to the testing data to evaluate the model's robustness. Moreover, the plots of relative deviations and residuals were dispersed across the zero-reference line with moderate

关键词： Nitrogen Removal Municipal Wastewater Twin Fluidized Bed Bioreactor Feature Selection bayesian optimization algorithm SVR Model

来源：评论

学校读者我要写书评

暂无评论

A rolling bearing fault diagnosis technique based on recurrence quantification analysis and bayesian optimization SVM

引用

APPLIED SOFT COMPUTING 2024年 156卷

作者： Wang, Bing Qiu, Wentao Hu, Xiong Wang, Wei Shanghai Maritime Univ Shanghai 201306 Peoples R China

A rolling bearing fault diagnosis technique is proposed based on Recurrence Quantification Analysis (abbreviated as RQA) and bayesian optimized Support Vector Machine (abbreviated as RQA-Bayes-SVM). Firstly, analyzing the vibration signal with recurrence plot and the nonlinear feature parameters are extracted with RQA, constructing a feature matrix describing the fault mode and fault degree comprehensively. Finally, bayesian optimization algorithm is introduced for searching the best penalty factor C and kernel function parameter g of SVM and establishing an optimal Bayes-SVM model. Bearing datasets from CWRU is imported for diagnosis on fault mode and fault degree. The results show that the technique presents a good performance on fault mode diagnosis as well as fault degree distinction. Compared with common k-Nearest Neighbor (abbreviated as KNN) and Random Forest (abbreviated as RF) diagnosis models, Bayes-SVM has the best accuracy and stability, which indicates a potential value for engineering applications.

关键词： Rolling bearing RQA Fault diagnosis SVM bayesian optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

EVOLUTIONARY optimization IN DYNAMIC ENVIRONMENTS: BRINGING THE STRENGTHS OF DYNAMIC bayesian NETWORKS INTO bayesian optimization algorithm

引用

INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL 2013年第6期9卷 2485-2503页

作者： Kaedi, Marjan Ghasem-Aghaee, Nasser Ahn, Chang Wook Univ Isfahan Dept Informat Technol Engn Esfahan *** Iran Univ Isfahan Dept Comp Engn Esfahan *** Iran Sungkyunkwan Univ Dept Comp Engn 2066 Seobu Ro Suwon 440746 South Korea

In this paper, a new evolutionary algorithm termed DBN-MBOA (Memory based BOA with Dynamic bayesian Networks) is proposed for the dynamic optimization. In DBN-MBOA, the knowledge obtained from previously solved problems is encoded in some structures called network translators. The network translators defined on non-stationary Dynamic bayesian Networks (nsDBNs) describe the correlation between conditional dependencies of candidate solution variables before and after environmental changes. The network translators constructed for the changes are stored in memory. When any change occurs in the environment, a relevant network translator is retrieved from the memory and is used for modifying the dependencies of the current bayesian network. In the retrieve stage, unlike existing memory-based methods, the relevant network translator is selected based on the characteristic of the change itself, not that of the new environmental state. Experimental results show that DBN-MBOA achieves better performance in random environments as well as cyclic environments.

关键词： bayesian optimization algorithm Dynamic bayesian network Dynamic optimization Environmental change Memory-based method Network translator

来源：评论

学校读者我要写书评

暂无评论

Multi-output discrete grey model tailored for electricity consumption forecast

引用

APPLIED MATHEMATICAL MODELLING 2025年 139卷

作者： Zhang, Xu Dang, Yaoguo Ding, Song Wang, Haowei Ding, Feipeng Nanjing Univ Aeronaut & Astronaut Coll Econ & Management Nanjing 211106 Peoples R China Zhejiang Univ Finance & Econ Sch Econ Hangzhou 310018 Peoples R China Natl Univ Singapore Dept Ind Syst Engn & Management Singapore Singapore Jiangxi Univ Finance & Econ Sch Stat & Data Sci Nanchang 330013 Peoples R China

To address the limitation of many conventional grey forecasting methods that prioritize temporal dynamics analysis while often overlooking the essential spatial characteristics, we developed a multi-output discrete grey model tailored for forecasting electricity consumption in China's Yangtze River Delta, incorporating spatial effects. Specifically, we design a dynamic spatial interaction matrix to precisely capture the spatial correlations among neighboring areas, and integrate multiple sets of fixed-effect parameters to adeptly address spatial heterogeneity. Additionally, we implement regularization technique to prevent overfitting and utilize the bayesian optimization algorithm for hyper-parameter adjustment, considerably enhancing the model's stability of parameter estimation and its resilience to noise. Finally, a comprehensive case study, benchmarked against seven other models, highlights our model's outstanding multi-step predictive accuracy and robustness against input data uncertainty, exceeding the performance of existing methods.

关键词： Grey forecasting Spatial correlation and spatial heterogeneity bayesian optimization algorithm Regularization Electricity consumption forecasting

来源：评论

学校读者我要写书评

暂无评论

A Novel Deep Learning Based Classification of Building Roof Types Using Point Cloud Data

引用

JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING 2025年第3期53卷 815-826页

作者： Yildirim, Merve Karsli, Fevzi Karadeniz Tech Univ Engn Fac Dept Geomat Engn Trabzon Turkiye

The classification of various building roof types within point clouds has become a significant research topic, as it constitutes a crucial step for numerous applications, including the automatic detection and extraction of buildings, urban area mapping, and energy potential analysis. This study aims to develop a novel deep learning architecture that uses point clouds as input data to classify different types of building roofs. For this purpose, the proposed architecture, whose hyperparameters are optimized using the bayesian optimization algorithm, is trained and tested on the RoofN3D dataset. This architecture classified three different types of roof on the test data with an overall classification accuracy of 95%. In addition, the classification performance of the proposed architecture is compared with the well-known PointNet architecture. The outputs indicate that the classification performance of the two architectures is almost identical, with only a small difference due to the fact that the original design of the proposed architecture has fewer layers and thus a shorter training time. Therefore, the proposed architecture is an important method for automatically classifying different building roof types from point cloud data.

关键词： Point cloud Roof type classification Deep learning bayesian optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Enhanced Prediction of Geothermal Heat Flow Using an Improved GBRT Model with Genetic algorithm

引用

NATURAL RESOURCES RESEARCH 2025年 1-27页

作者： Chen, Yang Li, Kewen Zhang, Han Gu, Chao He, Jifu Lu, Wei China Univ Geosci Beijing Sch Energy Resources Beijing 100083 Peoples R China Henan Acad Geol Zhengzhou 450016 Henan Peoples R China SINOPEC Res Inst Petr Engn Co Ltd Beijing Peoples R China Ecol Environm Geoserv Ctr Henan Geol Bur Zhengzhou 450053 Henan Peoples R China

Geothermal heat flow (GHF) comprises comprehensive data on geothermal temperature gradients, rock thermal conductivity, and crustal/mantle heat flow, which is crucial for evaluating regional geothermal resources and conducting other studies. However, the traditional hyperparameter tuning methods such as grid search and bayesian optimization (BO) are computationally intensive and susceptible to local optima, thereby complicating the accurate prediction of GHF. Moreover, the risk of overfitting poses a significant challenge in developing robust predictive models of GHF. To address these challenges, this study employed a genetic algorithm (GA) to tune the hyperparameters of the gradient boosted regression tree (GBRT) model. This research integrated the GBRT model with GA to establish a GA-GBRT hybrid model for predicting GHF more accurately. The GA-GBRT hybrid model incorporates 17 geological and geophysical features from Henan Province and its surroundings for sample data training. It is demonstrated that the GA-GBRT model improves the generalization performance of the test dataset and increases the R2 by 11.75% compared with the GBRT model after BO. Through prediction performance analysis, the GA-GBRT model outperformed the existing optimized random forest (RF), deep neural network (DNN), and traditional interpolation methods in all of the studied cases. Based on the GA-GBRT model, a new GHF map was created and it demonstrated a more rational representation of the GHF distribution within the area of interest compared to traditional interpolation outcomes. The superiority of the GA-GBRT model to predict GHF was validated by the geological background, geodynamics, geophysical information, and high-temperature hot springs data from Henan Province.

关键词： Geothermal heat flow Machine learning Genetic algorithm bayesian optimization algorithm Henan Province

来源：评论

学校读者我要写书评

暂无评论

Time series forecasting of oil production in Enhanced Oil Recovery system based on a novel CNN-GRU neural network

GEOENERGY SCIENCE AND ENGINEERING

引用

GEOENERGY SCIENCE AND ENGINEERING 2024年 233卷

作者： Chen, Guangxu Tian, Hailong Xiao, Ting Xu, Tianfu Lei, Hongwu Jilin Univ Key Lab Groundwater Resources & Environm Minist Educ Changchun 130021 Peoples R China Jilin Univ Jilin Prov Key Lab Water Resources & Environm Changchun 130021 Peoples R China Univ Utah Dept Civil & Environm Engn Salt Lake City UT USA Univ Utah Energy & Geosci Inst Salt Lake City UT 84108 USA Chinese Acad Sci Inst Rock & Soil Mech State Key Lab Geomech & Geotech Engn Wuhan 430071 Peoples R China Jilin Univ Key Lab Groundwater Resources & Environm Minist Educ Changchun 130021 Peoples R China

An accurate prediction of oil production is critical for the oilfield development, and many deep learning models have been widely employed for this purpose. However, those methods show insufficiencies in extracting complex features from multivariable time series datasets, which leaves the prediction of oil production still challenging. In this study, a novel CNN-GRU model combining Convolutional Neural Networks (CNN) and Gate Recurrent Unit (GRU) neural network was proposed to accurately predict oil production for Enhanced Oil Recovery (EOR) performance. The CNN layer can extract the features from variables affecting oil production, and the GRU layer models temporal information using the transmitted features for prediction. The bayesian optimization algorithm (BO) was employed to design the optimal hyper-parameters of CNN-GRU. For evaluation purpose, two case studies were carried out with the production data from a CO2-EOR project and a waterflooding project. The prediction performance of the proposed approach was compared with typical deep learning methods and a hybrid (statistical and machine learning) method. The results of experiments and comparisons indicate that the proposed CNN-GRU model outperforms other prediction approaches. The CNN-GRU model provides future oil production of wells, enabling engineers to make informed decisions in development plan of reservoirs.

关键词： Oil production forecasting Deep learning Convolutional neural network Gate recurrent unit EOR bayesian optimization 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

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：