检索结果-内蒙古大学图书馆

data preprocessing and machine learning method based on ameliorated mathematical models for inferring the power generation of photovoltaic system

引用

ENERGY CONVERSION AND MANAGEMENT 2025年 333卷

作者： Shin, Woo Gyun Lee, Jin Seok Ju, Young Chul Hwang, Hey Mi Ko, Suk Whan Korea Inst Energy Res Renewable Energy Inst Renewable Energy Syst Lab Daejeon South Korea

Countries worldwide are actively pursuing energy transition efforts to mitigate climate change and promote longterm sustainability. This transition involves shifting to carbon-free power sources, with solar energy playing a crucial role. As the installation of photovoltaic (PV) systems increases, the proportion of electricity these systems contribute to the power grid also rises. However, since weather conditions influence PV power generation, accurately inferring power output is essential for ensuring grid stability and assessing power generation efficiency. This paper presents a data preprocessing method for machine-learning regression models, utilizing a mathematical model to infer PV system power generation based on irradiance and module temperature data. The distinctiveness of the proposed method lies in its normalization process, where measured voltage and current values are divided by the corresponding values computed using the mathematical model. The proposed approach resulted in a highly accurate regression model, achieving coefficients of determination (R2) values of 0.9477, 0.9967, and 0.9969 for DC voltage, DC current, and AC power, respectively, along with normalized root mean squared error (NRMSE) values within 3%.

关键词： Mathematical model data preprocessing Photovoltaic system Power generation Machine learning

来源：评论

学校读者我要写书评

暂无评论

data preprocessing methods for selective sweep detection using convolutional neural networks

引用

METHODS 2025年 233卷 19-29页

作者： Zhao, Hanqing Alachiotis, Nikolaos Univ Twente Drienerlolaan 5 NL-7522 NB Enschede Overijssel Netherlands

The identification of positive selection has been framed as a classification task, with Convolutional Neural Networks (CNNs) already outperforming summary statistics and likelihood-based approaches in accuracy. Despite the prevalence of CNN-based methods that manipulate the pixels of images representing raw genomic data as a preprocessing step to improve classification accuracy, the efficacy of these pixel-rearrangement techniques remains inadequately examined, particularly in the presence of confounding factors like population bottlenecks, migration and recombination hotspots. We introduce a set of pixel rearrangement algorithms aimed at enhancing CNN classification accuracy in detecting selective sweeps. These algorithms are employed to assess the performance of four CNN models for selective sweep detection. Our findings illustrate that the judicious application of rearrangement algorithms notably enhances the overall classification accuracy of a CNN across various datasets simulating confounding factors. We observed that sorting the columns of the genomic matrices has higher on CNN performance than rearranging the sequences. To some extent, these rearrangement algorithms are more robust to misspecified demographic models compared with the utilization of the default preprocessing algorithm as suggested by the respective authors of each CNN architecture. We provide the data rearrangement algorithms as a distinct package available for download at: https://*** /Zhaohq96/ Genetic-data-rearrangement.

关键词： Population genetics Selective sweep Convolutional neural networks data preprocessing

来源：评论

学校读者我要写书评

暂无评论

A comprehensive investigation to enhance predictive accuracy of data-driven models applied to air-cooled condensers: from domain expertise data preprocessing to hyperparameters optimization

引用

APPLIED THERMAL ENGINEERING 2025年 267卷

作者： Mohammadi, Mobin Esfahanian, Vahid Univ Tehran Vehicle Fuel & Environm Res Inst Sch Mech Engn Coll Engn Tehran Iran

Owing to minimal water consumption, air-cooled condensers (ACC) have increasingly been adopted worldwide, particularly within the power generation sector. Recently, utilizing data-driven models instead of equations describing physical phenomena has gained considerable attention in the industry. The present investigation employs three machine learning (ML) methodologies to predict ACC performance. The ML models are developed based on one whole year in-situ data collected from a power plant. Previous approaches often have overlooked the integration of domain knowledge related to the operation of the ACC in data preprocessing. Furthermore, several input features used in existing ML models are not directly measurable, which can introduce additional uncertainty. To address these gaps, the present study introduces a novel approach for ACC pressure prediction that is directed by domain expertise. Results demonstrate that the appropriate selection of practical features and proper outlier detection, informed by field knowledge, significantly enhance the prediction accuracy. The study shows that considering the average speed of the operating parallel and mixed fans in each street enhances accuracy by up to 24 %, in comparison with traditional methods that an average value of fan speeds is utilized. Furthermore, by eliminating data samples with harsh non-condensing effect as outliers, the prediction accuracy is enhanced by 36 %. Then, hyperparameter tuning of the ML models is conducted. The final ML models achieve errors below 2 kPa for over 99 % of the train dataset and 98 % of the test dataset with average errors less than 0.3 and 0.5 kPa for the train and test dataset, respectively.

关键词： Air-cooled condenser Machine learning Domain expertize data preprocessing Power plants

来源：评论

学校读者我要写书评

暂无评论

Cutting-Edge Review of Big data preprocessing in TBM Tunnel Construction

引用

SHOCK AND VIBRATION 2025年第1期2025卷

作者： Guo, Qinghua Xiao, Haohan He, Rongjian Liu, Lipeng Yang, Haopeng Yu, Hongtao Guangdong Yuehai Yuexi Water Supply Co Ltd Zhanjiang Peoples R China China Inst Water Resources & Hydropower Res Key Lab Simulat & Regulat Water Cycle River Basin Beijing 100038 Peoples R China

Tunnel boring machines (TBMs) accumulate vast operational data crucial for analyzing complex rock-machine interactions during construction. Recent advancements in big data mining and machine learning (ML) have spurred significant artificial intelligence (AI) research in tunnel construction. The effectiveness of ML models depends on high-quality tunneling data, prompting increased research emphasis on big data preprocessing. This review synthesizes current practices in data preprocessing for tunnel construction, covering data characteristics, methods for handling anomalies such as discrete values, missing data, and noise. It also addresses challenges and future directions in big data processing. These findings establish a comprehensive foundation for further research in big data preprocessing and ML modeling in TBM construction.

关键词： data preprocessing discrete values missing data noise data TBM tunnel

来源：评论

学校读者我要写书评

暂无评论

Ensemble learning unlocking point load forecasting accuracy: A novel framework based on two-stage data preprocessing and improved multi-objective optimisation strategy

引用

COMPUTERS & ELECTRICAL ENGINEERING 2025年 124卷

作者： Luan, Jingmin Li, Qiyang Qiu, Yuyan Liu, Weican Northeastern Univ Qinhuangdao Qinhuangdao Peoples R China Nanyang Technol Univ Singapore Singapore

Accurate point power forecasts are critical for maintaining the security and stability of the grid. However, Load data is volatile and difficult to predict with high accuracy. To improve the stability and accuracy of the model, we proposed a novel hybrid load forecasting framework consisting of three modules. Firstly, we used Successive Variational Mode Decomposition (SVMD) in the data preprocessing module to extract trend features and denoise the data. To reduce the negative impact of feature redundancy on model predictions, we employed feature selection to identify four essential features to aid model training. Secondly, in the ensemble learning module, we address the limitations of single predictive models by combining the models Back Propagation (BP), Temporal Convolutional Network (TCN), Bidirectional Long Short-Term Memory (BiLSTM), Bidirectional Gated Recurrent Unit (BiGRU) and Transformer. Finally, in the improved multi-objective optimisation algorithm module, we implemented various strategies to enhance the optimisation algorithm. We designed some experiments using three load datasets from Australia. The results demonstrated that the mean absolute percentage error values are below 1.25%, with the best value reaching 0.9256%. In contrast, the best result for the mean absolute percentage error of the baseline models was 1.4431% in New South Wales. This represents a 35.9% improvement in load forecasting performance with our proposed model, highlighting its superior accuracy compared to competing approaches. It shows that our forecasting framework is far better than other rivals.

关键词： Load forecast Ensemble learning data preprocessing Multi-objective optimisation

来源：评论

学校读者我要写书评

暂无评论

Rapid identification of pathogenic bacteria using data preprocessing and machine learning-augmented label-free surface-enhanced Raman scattering

引用

SENSORS AND ACTUATORS B-CHEMICAL 2025年 425卷

作者： Jeon, Youngho Lee, Suji Jeon, Yu-Jin Kim, Dabin Ham, Jun-Hyeok Jung, Dae-Hyun Kim, Hae-Yeong You, Jungmok Kyung Hee Univ Dept Plant & Environm New Resources 1732 Deogyeong Daero Yongin 17104 Gyeonggi Do South Korea Kyung Hee Univ Grad Sch Green Bio Sci Yongin 17104 Gyeonggi Do South Korea Kyung Hee Univ Dept Smart Farm Sci Yongin 17104 Gyeonggi Do South Korea Kyung Hee Univ Inst Life Sci & Resources Yongin 17104 Gyeonggi Do South Korea Kyung Hee Univ Dept Food Sci & Biotechnol Yongin 17104 Gyeonggi Do South Korea

Despite the advantages of label-free surface-enhanced Raman scattering (SERS) for bacterial detection, the effective implementation of this approach is hindered by several intrinsic limitations, such as the uniformity and spectral similarity of bacterial SERS spectra. In this study, we developed a machine learning (ML)-integrated hydrophobic SERS platform for sensitive bacterial detection and classification. The hydrophobic Si wafer was successfully fabricated through treatment with dimethyldichlorosilane (DMDCS), resulting in a high contact angle (> 100 degrees). To achieve the highest SERS signals for bacteria, the density of gold core-silver shell nanoparticles (Au@AgNPs) was adjusted. Subsequently, the SERS activity of the Au@AgNPs under the coffee-ring effect on normal Si substrates was compared with that under the local concentration effect on hydrophobic Si substrates. Notably, the SERS signal for bacteria detected from the hydrophobic SERS platform, which enabled local enrichment of bacteria and Au@AgNPs, was approximately 33 times higher than that obtained from the normal Si substrates. Moreover, this hydrophobic SERS platform was integrated with ML algorithms to enhance the accuracy of bacterial detection. By applying three data preprocessing techniques, including baseline correction, smoothing, and normalization, we achieved 100 % accuracy in classifying SERS signals for four bacterial species. Our simple and versatile hydrophobic SERS platform, combined with ML models, exhibits immense potential for label-free detection and classification of pathogenic bacteria.

关键词： Bacteria classification surface-enhanced Raman scattering local concentration machine learning data preprocessing

来源：评论

学校读者我要写书评

暂无评论

Enhancing Road Traffic Prediction Using data preprocessing Optimization

引用

JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS 2025年第2期34卷

作者： Garg, Tanya Kaur, Gurjinder Rana, Prashant Singh Cheng, Xiaochun St Longowal Inst Engn & Technol Dept Comp Sci & Engn Sangrur 148106 Punjab India Thapar Inst Engn & Technol Comp Sci & Engn Patiala 147004 Punjab India Swansea Univ Comp Sci Engn Dept Bay Campus Swansea SA1 8EN Wales

Traffic prediction is essential for transportation planning, resource allocation, congestion management and enhancing travel experiences. This study optimizes data preprocessing techniques to improve machine learning-based traffic prediction models. data preprocessing is critical in preparing the data for machine learning models. This study proposes an approach that optimizes data preprocessing techniques, focusing on flow-based analysis and optimization, to enhance traffic prediction models. The proposed approach explores fixed and variable orders of data preprocessing using a genetic algorithm across five diverse datasets. Evaluation metrics such as root mean squared error (RMSE), mean absolute error (MAE) and R-squared error assess model performance. The results indicate that the genetic algorithm's variable order achieves the best performance for the ArcGIS Hub and Frementon Bridge Cycle datasets, fixed order one preprocessing for the Traffic Prediction dataset and variable order using the genetic algorithm for the PeMS08 dataset. Fixed order 2 preprocessing yields the best performance for the XI AN Traffic dataset. These findings highlight the importance of selecting the appropriate data preprocessing flow order for each dataset, improving traffic prediction accuracy and reliability. The proposed approach advances traffic prediction methodologies, enabling more precise and reliable traffic forecasts for transportation planning and management applications.

关键词： data preprocessing random forest regression genetic algorithm parameter optimization traffic prediction preprocessing order

来源：评论

学校读者我要写书评

暂无评论

Clinician-driven automated data preprocessing in nuclear medicine AI environments

引用

EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING 2025年 1-11页

作者： Krajnc, Denis Spielvogel, Clemens P. Ecsedi, Boglarka Ritter, Zsombor Alizadeh, H. Hacker, Marcus Papp, Laszlo Med Univ Vienna Ctr Med Phys & Biomed Engn Vienna Austria Med Univ Vienna Dept Biomed Imaging & Image Guided Therapy Div Nucl Med Vienna Austria Georgia Inst Technol Atlanta GA USA Univ Pecs Med Sch Dept Med Imaging Pecs Hungary Univ Pecs Med Sch Dept Internal Med 1 Pecs Hungary Med Univ Vienna Ctr Med Phys & Biomed Engn Appl Quantum Comp Grp Wahringer Gurtel 18-20 A-1090 Vienna Austria

BackgroundArtificial Intelligence (AI) approaches in clinical science require extensive data preprocessing (DP) steps prior to building AI models. Establishing DP pipelines is a non-trivial task, mainly driven by purely mathematical rules and done by data scientists. Nevertheless, clinician presence shall be paramount at this step. The study proposes a data preprocessing approach driven by clinical domain knowledge, where clinician input, in form of explicit and non-explicit rules, directly impacts the algorithms' decision-making processes, thus, making the DP planning phase more inclusive for *** rule set table (RST) was introduced as interface which accepts clinician's input as formal rules (including four actions: exp-keep, exp-remove, pref-keep, pref-remove features or samples) in human-readable form and translates it to machine readable input for preprocessing algorithms. A collection of commonly used algorithms was incorporated for data preprocessing of various clinical cohorts in both single and multi-center scenarios. The impact of RST was evaluated by utilizing 100-fold Monte Carlo cross-validation scheme for prostate and glioma cohorts (single center) with 80 - 20% training-testing split. Furthermore, diffuse large B-cell lymphoma (DLBCL) cohort was evaluated by using Center 1 as training and Center 2 as testing cohort for clinical endpoint prediction. Both scenarios were investigated in manual and automated data preprocessing setups across all cohorts. The XGBoost algorithm was employed for classification tasks across all established models. Predictive performance was estimated by confusion matrix analysis in validation samples of all cohorts. The performance of RST across all actions as well as without RST were compared in both manual and automated settings for each respective *** increase of ML models with manual preprocessing combined with RST was up-to 18% balanced accuracy (BACC) compared to models without RST.

关键词： AI data preprocessing Cancer Imaging analysis

来源：评论

学校读者我要写书评

暂无评论

On construction of data preprocessing for real-life SoyLeaf dataset & disease identification using Deep Learning Models

引用

COMPUTATIONAL BIOLOGY AND CHEMISTRY 2025年 117卷 108417页

作者： Gudge, Sujata Tiwari, Aruna Ratnaparkhe, Milind Jha, Preeti Indian Inst Technol Indore Indore 453552 Madhya Pradesh India ICAR Indian Inst Soybean Res Indore 452001 Madhya Pradesh India Koneru Lakshmaiah Educ Fdn Hyderabad 500043 Telangana India

The vast volumes of data are needed to train Deep Learning Models from scratch to identify illnesses in soybean leaves. However, there is still a lack of sufficient high-quality samples. To overcome this problem, we have developed the real-life SoyLeaf dataset and used the pre-trained Deep Learning Models to identify leaf diseases. In this paper, we have initially developed the real-life SoyLeaf dataset collected from the ICAR-Indian Institute of Soybean Research (IISR) Center, Indore field. This SoyLeaf dataset contains 9786 high-quality soybean leaf images, including healthy and diseased leaves. Following this, we have adapted data preprocessing techniques to enhance the quality of images. In addition, we have utilized several Deep Learning Models, i.e., fourteen Keras Transfer Learning Models, to determine which model best fits the dataset on SoyLeaf diseases. The accuracies of the proposed fine-tuned models using the Adam optimizer are as follows: ResNet50V2 achieves 99.79%, ResNet101V2 achieves 99.89%, ResNet152V2 achieves 99.59%, InceptionV3 achieves 99.83%, InceptionResNetV2 achieves 99.79%, MobileNet achieves 99.82%, MobileNetV2 achieves 99.89%, DenseNet121 achieves 99.87%, and DenseNet169 achieves 99.87%. Similarly, the accuracies of the proposed fine-tuned models using the RMSprop optimizer are as follows: ResNet50V2 achieves 99.49%, ResNet101V2 achieves 99.45%, ResNet152V2 achieves 99.45%, InceptionV3 achieves 99.58%, InceptionResNetV2 achieves 99.88%, MobileNet achieves 99.73%, MobileNetV2 achieves 99.83%, DenseNet121 achieves 99.89%, and DenseNet169 achieves 99.77%. The experimental results of the proposed fine-tuned models show that only ResNet50V2, ResNet101V2, InceptionV3, InceptionResNetV2, MobileNet, MobileNetV2, DenseNet121, and DenseNet169 have performed better in terms of training, validation, and testing accuracies than other state-of-the-art models.

关键词： SoyLeaf diseases data preprocessing Transfer learning Deep Learning Models

来源：评论

学校读者我要写书评

暂无评论

TockyPrep: data preprocessing methods for flow cytometric fluorescent timer analysis

引用

BMC BIOINFORMATICS 2025年第1期26卷 1-16页

作者： Ono, Masahiro Imperial Coll London Dept Life Sci Imperial Coll Rd London SW7 2AZ England

Background:Fluorescent Timer proteins, which display time-dependent changes in their emission spectra, are invaluable for analyzing the temporal dynamics of cellular events at the single-cell level. We previously developed the Timer-of-cell-kinetics-and-activity (Tocky) tools, utilizing a specific Timer protein, Fast-FT, to monitor temporal changes in cellular activities. Despite their potential, the analysis of Timer fluorescence in flow cytometry is frequently compromised by variability in instrument settings and the absence of standardized preprocessing methods. The development and implementation of effective data preprocessing methods remain to be ***:In this study, we introduce the R package that automates the data preprocessing of Timer fluorescence data from flow cytometry experiments for quantitative analysis at single-cell level. Our aim is to standardize Timer data analysis to enhance reproducibility and accuracy across different experimental setups. The package includes a trigonometric transformation method to elucidate the dynamics of Fluorescent Timer proteins. We have identified the normalization of immature and mature Timer fluorescence data as essential for robust analysis, clarifying how this normalization affects the analysis of Timer maturation. These preprocessing methods are all encapsulated within the TockyPrep ***:TockyPrep is available for distribution via GitHub at https://***/MonoTockyLab/TockyPrep, providing tools for data preprocessing and basic visualization of Timer fluorescence data. This toolkit is expected to enhance the utility of experimental systems utilizing Fluorescent Timer proteins, including the Tocky tools.

关键词： Fluorescent Timer Protein Tocky Flow Cytometry data preprocessing Nr4a3-Tocky

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：