Data Distribution Difference Research Articles

Addressing Heterogeneity in Federated Learning with Client Selection via Submodular Optimization

Federated learning (FL) has been proposed as a privacy-preserving distributed learning paradigm, which differs from traditional distributed learning in two main aspects: the systems heterogeneity, meaning that clients participating in training have significant differences in systems performance including CPU frequency, dataset size, and transmission power, and the statistical heterogeneity, indicating that the data distribution among clients exhibits Non-Independent Identical Distribution. Therefore, the random selection of clients will significantly reduce the training efficiency of FL. In this article, we propose a client selection mechanism considering both systems and statistical heterogeneity, which aims to improve the time-to-accuracy performance by trading off the impact of systems performance differences and data distribution differences among the clients on training efficiency. First, client selection is formulated as a combinatorial optimization problem that jointly optimizes systems and statistical performance. Then, we generalize it to a submodular maximization problem with knapsack constraint, and propose the Iterative Greedy with Partial Enumeration (IGPE) algorithm to greedily select the suitable clients. Then, the approximation ratio of IGPE is analyzed theoretically. Extensive experiments verify that the time-to-accuracy performance of the IGPE algorithm outperforms other compared algorithms in a variety of heterogeneous environments.

Heterogeneous Domain Adaptation Framework for Logging Lithofacies Identification

Summary Reservoir lithofacies type is an important indicator of reservoir quality and oiliness, and understanding lithofacies type can help geologists and engineers make informed decisions about exploration and development activities. The use of well-log data to establish machine learning models for lithofacies identification has gained popularity; however, the assumption that data are independent identical distribution followed by these models is often unrealistic. Additionally, there is a possible incompatibility between the training and test data in terms of feature space dimensions. We propose the heterogeneous domain adaptation framework for logging lithofacies identification (HDAFLI) to address these problems. The framework comprises three main contributions: (i) The denoising autoencoder feature mapping (DAFM) module is adopted to resolve the incompatibility issue in feature space between training and test data. The connection between training and test data can be effectively established to improve the performance and generalization ability. (ii) The transferability and discriminative joint probability distribution adaptive (TDJPDA) module addresses the issue of data distribution differences. It improves the transferability of training and test data by minimizing the maximum mean difference (MMD) of the joint probabilities of the source and target domains and enhances their discriminative ability by maximizing the joint probability MMD of different lithofacies categories. (iii) Bayesian optimization is used to optimize hyperparameters in the light gradient boosting machine (LightGBM) model for high computational efficiency in determining the best accuracy. We selected well-logging data from eight wells in the Pearl River Mouth Basin of the South China Sea to design four tasks and compared HDAFLI with various baseline machine learning algorithms and baseline domain adaptive algorithms. The results show that HDAFLI has the highest average accuracy among the four tasks. It is 19.76% and 8.94% higher than the best-performing baseline machine learning algorithm and baseline domain adaptive method among the comparison algorithms, respectively. For HDAFLI, we also conducted ablation experiments, time cost and convergence performance analysis, parameter sensitivity experiments, and feature visualization experiments. The results of ablation experiments show that the three modules of HDAFLI all play an active role, working together to achieve the best results. In addition, HDAFLI has a reasonable time cost, can become stable after several iterations, and has good convergence performance. The results of parameter sensitivity experiments confirm that the accuracy of HDAFLI does not change significantly with changes in hyperparameters, which is robust. The results of feature visualization experiments show that the data of the training set and the test set are concentrated together to a certain extent, which indicates that HDAFLI has completed the task of data distribution alignment very well. The findings of this study can help for a better understanding of how to address the challenge of reservoir lithofacies identification through a heterogeneous domain adaptation framework. By solving the problem of feature space incompatibility and data distribution difference between training data and test data, the application of HDAFLI provides geologists and engineers with more accurate lithofacies classification tools. This study has practical application value for reservoir quality assessment, oiliness prediction, and exploration and development decision-making.

Flood Susceptibility Assessment with Random Sampling Strategy in Ensemble Learning (RF and XGBoost)

Due to the complex interaction of urban and mountainous floods, assessing flood susceptibility in mountainous urban areas presents a challenging task in environmental research and risk analysis. Data-driven machine learning methods can evaluate flood susceptibility in mountainous urban areas lacking essential hydrological data, utilizing remote sensing data and limited historical inundation records. In this study, two ensemble learning algorithms, Random Forest (RF) and XGBoost, were adopted to assess the flood susceptibility of Kunming, a typical mountainous urban area prone to severe flood disasters. A flood inventory was created using flood observations from 2018 to 2022. The spatial database included 10 explanatory factors, encompassing climatic, geomorphic, and anthropogenic factors. Artificial Neural Network (ANN) and Support Vector Machine (SVM) were selected for model comparison. To minimize the influence of expert opinions on model training, this study employed a strategy of uniformly random sampling in historically non-flooded areas for negative sample selection. The results demonstrated that (1) ensemble learning algorithms offer higher accuracy than other machine learning methods, with RF achieving the highest accuracy, evidenced by an area under the curve (AUC) of 0.87, followed by XGBoost at 0.84, surpassing both ANN (0.83) and SVM (0.82); (2) the interpretability of ensemble learning highlighted the differences in the potential distribution of the training data’s positive and negative samples. Feature importance in ensemble learning can be utilized to minimize human bias in the collection of flooded-site samples, more targeted flood susceptibility maps of the study area’s road network were obtained; and (3) ensemble learning algorithms exhibited greater stability and robustness in datasets with varied negative samples, as evidenced by their performance in F1-Score, Kappa, and AUC metrics. This paper further substantiates the superiority of ensemble learning in flood susceptibility assessment tasks from the perspectives of accuracy, interpretability, and robustness, enhances the understanding of the impact of negative samples on such assessments, and optimizes the specific process for urban flood susceptibility assessment using data-driven methods.

Gearbox fault diagnosis method based on lightweight channel attention mechanism and transfer learning

In practical engineering, the working conditions of gearbox are complex and variable. In varying working conditions, the performance of intelligent fault diagnosis model is degraded because of limited valid samples and large data distribution differences of gearbox signals. Based on these issues, this research proposes a gearbox fault diagnosis method integrated with lightweight channel attention mechanism, and further realizes the cross-component transfer learning. First, time–frequency distribution of original signals is obtained by wavelet transform. It could intuitively reflect local characteristics of signals. Secondly, based on a local cross-channel interaction strategy, a lightweight efficient channel attention mechanism (LECA) is designed. The kernel size of 1D convolution is affected by channel number and coefficients. Multi-scale feature input is used to retain more detailed features of different dimensions. A lightweight convolutional neural network is constructed. Finally, a transfer learning method is applied to freeze lower structures of the network and fine-tune higher structures of the model using small samples. Through experimental verification, the proposed model could effectively utilize samples. The application of transfer learning could realize accurate and fast fault classification of small samples, and achieve good gearbox fault diagnosis effect under varying working conditions and cross-component conditions.

A New Framework Based on Supervised Joint Distribution Adaptation for Bearing Fault Diagnosis across Diverse Working Conditions

To address the degradation of diagnostic performance due to data distribution differences and the scarcity of labeled fault data, this study has focused on transfer learning-based cross-domain fault diagnosis, which attracts considerable attention. However, deep transfer learning-based methods often present a challenge due to their time-consuming and costly nature, particularly in tuning hyperparameters. For this issue, on the basis of classical features-based transfer learning method, this study introduces a new framework for bearing fault diagnosis based on supervised joint distribution adaptation and feature refinement. It first utilizes ensemble empirical mode decomposition to process raw signals, and statistical features extraction is implemented. Then, a new feature refinement module is designed to refine domain adaptation features from high-dimensional feature set by evaluating the fault distinguishability and working-condition invariance of feature data. Next, it proposes a supervised joint distribution adaptation method to conduct improved joint distribution alignment that preserves neighborhood relationships within a manifold subspace. Finally, an adaptive classifier is trained to predict fault labels of feature data across varying working conditions. To prove the cross-domain fault diagnosis performance and superiority of the proposed methods, two bearing datasets are applied for experiments, and the experimental results verify that the model built by the proposed framework can achieve desirable diagnosis performance under different working conditions and that it apparently outperforms comparative models.

IC‐GraF: An Improved Clustering with Graph‐Embedding‐Based Features for Software Defect Prediction

Software defect prediction (SDP) has been a prominent area of research in software engineering. Previous SDP methods often struggled in industrial applications, primarily due to the need for sufficient historical data. Thus, clustering‐based unsupervised defect prediction (CUDP) and cross‐project defect prediction (CPDP) emerged to address this challenge. However, the former exhibited limitations in capturing semantic and structural features, while the latter encountered constraints due to differences in data distribution across projects. Therefore, we introduce a novel framework called improved clustering with graph‐embedding‐based features (IC‐GraF) for SDP without the reliance on historical data. First, a preprocessing operation is performed to extract program dependence graphs (PDGs) and mark distinct dependency relationships within them. Second, the improved deep graph infomax (IDGI) model, an extension of the DGI model specifically for SDP, is designed to generate graph‐level representations of PDGs. Finally, a heuristic‐based k‐means clustering algorithm is employed to classify the features generated by IDGI. To validate the efficacy of IC‐GraF, we conduct experiments based on 24 releases of the PROMISE dataset, using F‐measure and G‐measure as evaluation criteria. The findings indicate that IC‐GraF achieves 5.0%−42.7% higher F‐measure, 5%−39.4% higher G‐measure, and 2.5%−11.4% higher AUC over existing CUDP methods. Even when compared with eight supervised learning‐based SDP methods, IC‐GraF maintains a superior competitive edge.

Robust Surface Recognition With the Maximum Mean Discrepancy: Degrading Haptic-Auditory Signals Through Bandwidth and Noise.

Sliding a tool across a surface generates rich sensations that can be analyzed to recognize what is being touched. However, the optimal configuration for capturing these signals is yet unclear. To bridge this gap, we consider haptic-auditory data as a human explores surfaces with different steel tools, including accelerations of the tool and finger, force and torque applied to the surface, and contact sounds. Our classification pipeline uses the maximum mean discrepancy (MMD) to quantify differences in data distributions in a high-dimensional space for inference. With recordings from three hemispherical tool diameters and ten diverse surfaces, we conducted two degradation studies by decreasing sensing bandwidth and increasing added noise. We evaluate the haptic-auditory recognition performance achieved with the MMD to compare newly gathered data to each surface in our known library. The results indicate that acceleration signals alone have great potential for high-accuracy surface recognition and are robust against noise contamination. The optimal accelerometer bandwidth exceeds 1000 Hz, suggesting that useful vibrotactile information extends beyond human perception range. Finally, smaller tool tips generate contact vibrations with better noise robustness. The provided sensing guidelines may enable superhuman performance in portable surface recognition, which could benefit quality control, material documentation, and robotics.

Adaptive multi-source domain collaborative fine-tuning for transfer learning.

Fine-tuning is an important technique in transfer learning that has achieved significant success in tasks that lack training data. However, as it is difficult to extract effective features for single-source domain fine-tuning when the data distribution difference between the source and the target domain is large, we propose a transfer learning framework based on multi-source domain called adaptive multi-source domain collaborative fine-tuning (AMCF) to address this issue. AMCF utilizes multiple source domain models for collaborative fine-tuning, thereby improving the feature extraction capability of model in the target task. Specifically, AMCF employs an adaptive multi-source domain layer selection strategy to customize appropriate layer fine-tuning schemes for the target task among multiple source domain models, aiming to extract more efficient features. Furthermore, a novel multi-source domain collaborative loss function is designed to facilitate the precise extraction of target data features by each source domain model. Simultaneously, it works towards minimizing the output difference among various source domain models, thereby enhancing the adaptability of the source domain model to the target data. In order to validate the effectiveness of AMCF, it is applied to seven public visual classification datasets commonly used in transfer learning, and compared with the most widely used single-source domain fine-tuning methods. Experimental results demonstrate that, in comparison with the existing fine-tuning methods, our method not only enhances the accuracy of feature extraction in the model but also provides precise layer fine-tuning schemes for the target task, thereby significantly improving the fine-tuning performance.

UMCGL: Universal Multi-View Consensus Graph Learning With Consistency and Diversity.

Existing multi-view graph learning methods often rely on consistent information for similar nodes within and across views, however they may lack adaptability when facing diversity challenges from noise, varied views, and complex data distributions. These challenges can be mainly categorized into: 1) View-specific diversity within intra-view from noise and incomplete information; 2) Cross-view diversity within inter-view caused by various latent semantics; 3) Cross-group diversity within inter-group due to data distribution differences. To this end, we propose a universal multi-view consensus graph learning framework that considers both original and generative graphs to balance consistency and diversity. Specifically, the proposed framework can be divided into the following four modules: i) Multi-channel graph module to extract principal node information, ensuring view-specific and cross-view consistency while mitigating view-specific and cross-view diversity within original graphs; ii) Generative module to produce cleaner and more realistic graphs, enriching graph structure while maintaining view-specific consistency and suppressing view-specific diversity; iii) Contrastive module to collaborate on generative semantics to facilitate cross-view consistency and reducing cross-view diversity within generative graphs; iv) Consensus graph module to consolidate learning a consensual graph, pursuing cross-group consistency and cross-group diversity. Extensive experimental results on real-world datasets demonstrate its effectiveness and superiority.

A zero-sample intelligent fault diagnosis method for bearings based on category relationship model

Collecting data containing all categories is crucial for fault diagnosis methods based on machine learning, which can be difficult in real industrial scenarios, particularly under new working condition. Therefore, a zero-sample intelligent fault diagnosis method is proposed for the bearings working in new condition. Specifically, a health-fault category relationship model among the different categories is developed using improved sparsity-constrained generative adversarial network. The fault data in new working condition can be generated by the model. Then, a fault diagnosis model in new working condition is established using transfer learning between existing working condition and new working condition. Finally, two bearing datasets are used to verify the proposed method. The results show that the model is effective for the bearings under new working condition without fault data, especially for the big difference of data distribution caused by the changing working conditions. It indicates that the proposed method has great potential in practical industrial applications.

Transferable Takagi-Sugeno-Kang Fuzzy Classifier With Multi-Views for EEG-Based Driving Fatigue Recognition in Intelligent Transportation

The safety monitoring system of intelligent transportation provides driving fatigue warning and risk control. Electroencephalogram (EEG) signals can directly reflect the neuronal activity of the brain. The detection and early warning of driving fatigue using EEG signals has important practical significance. However, because of the non-stationarity and timeliness of EEG signals, the single feature detection method is significantly impacted by data distribution differences. In this paper, in the framework of multi-input multi-output (MIMO) Takagi-Sugeno-Kang (TSK) fuzzy system, transferable TSK fuzzy classifier with multi-views (T-TSK-MV) is developed for EEG-based driving fatigue recognition in intelligent transportation. First, in view-specific consequent parameter learning, the view-specific consequent regularizer is designed based on technologies of ridge regression, maximum mean discrepancy (MMD), and manifold regularization, which becomes the bridge to transfer the discriminative information from the related domain to the target domain. In addition, the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\ell_{2,1} $</tex-math> </inline-formula> -norm sparse constraint on consequent parameters is used to simplify fuzzy rules. Then multi-view learning is integrated into the consequent parameter learning, in which T-TSK-MV explores the view-shared consequent regularizer and adaptively assigns weights to each view. The <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\ell_{2,1} $</tex-math> </inline-formula> -norm sparse constraint on view-shared consequent regularizer can effectively exploit the local structure of multi-view data. Finally, the fuzzy classifier is constructed on view-specific regularizers and view weights. The experiment on real-word datasets shows that the proposed fuzzy classifier can significantly improve the driving fatigue recognition performance.

Unsupervised domain adaptation without source data for estimating occupancy and recognizing activities in smart buildings

Activities Recognition (AR) and Occupancy Estimation (OE) are topics of current interest. AR and OE help many smart building applications such as energy systems and provide good services for occupants. Prior research on AR and OE has typically focused on supervised machine learning methods. For a specific smart building domain, a model is trained using data collected from the current environment (domain). The created model will not generalize well when evaluated in a new related domain due to data distribution differences. Creating a model for each smart building environment is infeasible due to the lack of labeled data. Indeed, data collection is a tedious and time-consuming task. Unsupervised Domain Adaptation (UDA) is a good solution for the considered case. UDA solves the problem of the lack of labeled data in the target domain by allowing knowledge transfer across domains. Most of the previous research in UDA requires having access to source data while creating target models which leads to privacy problems. This work considers techniques that use only a trained source model instead of a huge amount of source data to make domain adaption. This research adapted and tested UDA methods called Source HypOthesis Transfer (SHOT), Higher-Order Moment Matching (HoMM), and Source data Free Domain Adaptation (SFDA) on smart building data. SHOT is a deep learning method that learns a feature encoding module for the target model to align the data representation of the target environment with the data representation of the source environment, and it freezes the hypothesis (classifier) of the source model. Data alignment is done using information maximization and self-supervised pseudo-labeling. HoMM is also a deep learning method, however, it freezes the feature encoding module, and it learns a classifier to perform data alignment. HoMM also performs pseudo-labeling to target samples to enhance data alignment. SFDA is a deep domain adaptation method that optimizes two losses to train the target model without using any source-labeled data. The target model updates its initialized weights from the source model by minimizing a first loss that uses pseudo labels of target samples using the pre-trained source model, and by minimizing a second loss that uses pseudo labels of target samples generated by the trainable target model. To prove the efficiency of SHOT, HoMM, and SFDA, this research tests them on AR and OE datasets for a different number of activities and levels of occupancy. The impressive obtained results, with scores up to 90% for OE and up to 97% for AR, show that the considered approaches can be used to transfer knowledge across different related domains.

Research on the Prediction Method of Clock Tester Calibration Data Based on Radial Basis Function Neural Network

A radial basis function (RBF) neural network-based calibration data prediction model for clock testers is proposed to address the issues of fixed calibration cycles, low efficiency, and waste of electrical energy. This provides a new method for clock tester traceability calibration. First, analyze the mechanism of clock tester calibration parameters and the influencing factors of prediction targets. Based on the learning rules of an RBF neural network, determine the data types of training and testing sets. Second, normalize the training and testing data to avoid the adverse effects of data characteristics and distribution differences on the prediction model. Finally, based on different prediction objectives, time-driven and data-driven calibration data prediction models are constructed using RBF neural networks. Through simulation analysis, it is shown that an RBF neural network is superior to a BP neural network in predicting clock tester calibration data, and time-driven prediction accuracy is superior to data-driven prediction accuracy. Moreover, the prediction error and mean square error of both prediction models are on the order of 10−9, meeting the prediction accuracy requirements.

A Deep Learning Framework for Anesthesia Depth Prediction from Drug Infusion History.

In the target-controlled infusion (TCI) of propofol and remifentanil intravenous anesthesia, accurate prediction of the depth of anesthesia (DOA) is very challenging. Patients with different physiological characteristics have inconsistent pharmacodynamic responses during different stages of anesthesia. For example, in TCI, older adults transition smoothly from the induction period to the maintenance period, while younger adults are more prone to anesthetic awareness, resulting in different DOA data distributions among patients. To address these problems, a deep learning framework that incorporates domain adaptation and knowledge distillation and uses propofol and remifentanil doses at historical moments to continuously predict the bispectral index (BIS) is proposed in this paper. Specifically, a modified adaptive recurrent neural network (AdaRNN) is adopted to address data distribution differences among patients. Moreover, a knowledge distillation pipeline is developed to train the prediction network by enabling it to learn intermediate feature representations of the teacher network. The experimental results show that our method exhibits better performance than existing approaches during all anesthetic phases in the TCI of propofol and remifentanil intravenous anesthesia. In particular, our method outperforms some state-of-the-art methods in terms of root mean square error and mean absolute error by 1 and 0.8, respectively, in the internal dataset as well as in the publicly available dataset.

A Long Time Span-Specific Emitter Identification Method Based on Unsupervised Domain Adaptation

Specific emitter identification (SEI) is a professional technology to recognize different emitters by measuring the unique features of received signals. It has been widely used in both civilian and military fields. Recently, many SEI methods based on deep learning have been proposed, most of which assume that the training set and testing set have the same data distribution. However, in reality, the testing set is generally used later than the training set and lacks labels. The long time span may change the signal transmission environment and fingerprint features. These changes result in considerable differences in data distribution between the training and testing sets, thereby affecting the recognition and prediction abilities of the model. Therefore, the existing works cannot achieve satisfactory results for a long time span SEI. To address this challenge and obtain stable fingerprints, we transform the long time span SEI problem into a domain adaptive problem and propose an unsupervised domain adaptive method called LTS-SEI. Noteworthily, LTS-SEI uses a multilayer convolutional feature extractor to learn feature knowledge and confronts a domain discriminator to generate domain-invariant shallow fingerprints. The classifier of LTS-SEI applies feature matching to source domain samples and target domain samples to achieve the domain alignment of deep fingerprints. The classifier further reduces the intraclass diversity of deep features to alleviate the misclassification problem of edge samples in the target domain. To confirm the effectiveness and reliability of LTS-SEI, we collect multiple sets of real satellite navigation signals using two antennas with 13 m- and 40 m-large apertures, respectively, and construct two available datasets. Numerous experiments demonstrate that LTS–SEI can considerably increase the recognition accuracy of the long time span SEI and is superior to the other existing methods.

Class incremental learning of remote sensing images based on class similarity distillation.

When a well-trained model learns a new class, the data distribution differences between the new and old classes inevitably cause catastrophic forgetting in order to perform better in the new class. This behavior differs from human learning. In this article, we propose a class incremental object detection method for remote sensing images to address the problem of catastrophic forgetting caused by distribution differences among different classes. First, we introduce a class similarity distillation (CSD) loss based on the similarity between new and old class prototypes, ensuring the model's plasticity to learn new classes and stability to detect old classes. Second, to better extract class similarity features, we propose a global similarity distillation (GSD) loss that maximizes the mutual information between the new class feature and old class features. Additionally, we present a region proposal network (RPN)-based method that assigns positive and negative labels to prevent mislearning issues. Experiments demonstrate that our method is more accurate for class incremental learning on public DOTA and DIOR datasets and significantly improves training efficiency compared to state-of-the-art class incremental object detection methods.

FedBrain: A robust multi-site brain network analysis framework based on federated learning for brain disease diagnosis

In recent years, deep learning models have shown their advantages in neuroimage analysis, such as brain disease diagnosis. Unfortunately, it is usually difficult to acquire numerous brain networks at a single centralized site to effectively train a high-quality deep learning model. To address this issue, federated learning (FL) has gained popularity in brain disease diagnosis, which allows deep learning models to be trained without centralizing data. However, most FL-based works might still face two following challenges. Firstly, the high-dimensional features of brain networks are often far larger than sample size, which might lead to poor performance due to the curse of dimensionality. Secondly, differences in data distributions across different sites can impact the communication efficiency and performance of FL models. To overcome these challenges, we design a novel FL framework for diagnosing brain disorders, named FedBrain. Firstly, FedBrain proposes data augmentation based on L1 regularization to select significant features shared by all clients. The domain alignment loss based on the maximum mean discrepancy criterion is introduced to minimize differences in the marginal and conditional distributions between local clients. Furthermore, FedBrain proposes a personalized predictor based on mixture of experts to adapt to different clients, using a global and private predictor as two experts. Eventually, FedBrain integrates the above modules with differential privacy and homomorphic encryption into a unified FL framework. Experimental results on the Autism Brain Imaging Data Exchange (ABIDE) dataset demonstrate its effectiveness and robustness, which shows that FedBrain can reduce the communication burden of FL and achieve the highest average accuracy of 79% against other counterparts.

An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning

Due to significant differences in data distribution under different working conditions during Pichia pastoris biochemical reaction process, traditional soft-sensor model suffer from the model failure and deterioration, this paper propose a soft-sensor modeling method combing long short-term memory network (LSTM) and balanced distribution adaptation method (BDA). Firstly, the source domain data is used to establish an accurate source domain LSTM prediction model, and the structure and parameters of the first layer of LSTM are fixed to migrate to the target domain prediction model. Then use the balanced distribution adaptation method to shrink the distribution differences between different domains of data. Finally, data that has been modeled with balanced and adaptive distribution assist the real-time data to train the remaining layer of the network, and the accurate target domain prediction model is finally obtained. The simulation results show that the mentioned method has the preponderance of timely prediction and high prediction accuracy, which validates the effectiveness and practicality of the method. This method solves the problem of soft-sensor modeling under unknown modes of multiple operating conditions in Pichia pastoris biochemical reaction process, achieving the prediction of key parameters under different operating conditions, which can be widely applied, and also providing a new method for soft-sensor modeling of other non system systems.

Transfer condition monitoring across wind turbines using feature alignment and parameter fine-tuning

Transfer condition monitoring across wind turbines still exists two issues. (1) The model of single turbine trained is directly used for other ones in the wind farm to result in low monitoring accuracy because of individual features of each turbine; (2) if the model is trained separately for each turbine, it is not only time-consuming and inefficient, but also is difficult to transfer across turbines. For solving the two issues, a transfer condition monitoring method across wind turbines using feature alignment and parameter fine-tuning is proposed. First, the representative normal turbine in a wind farm is screened by calculating the correlation between the monitoring parameters of wind turbines; second, the input features are constructed by using the massive normal supervisory control and data acquisition data of the representative turbine. The intelligent condition monitoring model of wind turbines is designed by using convolutional neural networks for spatial features and long-term and short-term memory networks for time features. Finally, feature alignment is performed using the normal data of the representative turbine and a large amount of historical normal data from other turbines to reduce data distribution differences. The aligned data is inputted into pre-training monitoring models for parameter fine-tuning. The personalized features of each turbine are incorporated to achieve high-precision transfer condition monitoring across turbines. The proposed method is verified by the real-world turbine data from a wind farm cooperated with us. The results show that the proposed transfer condition monitoring method using feature alignment and parameter fine-tuning can accurately monitor and identify the health states of wind turbines.

An efficient intrusion detection method using federated transfer learning and support vector machine with privacy-preserving

In recent decades, network security for organizations and individuals has become more and more important, and intrusion detection systems play a key role in protecting network security. To improve intrusion detection effect, different machine learning techniques have been widely applied and achieved exciting results. However, the premise that these methods achieve reliable results is that there are enough available and well-labeled training data, training and test data being from the same distribution. In real life, the limited label data generated by a single organization is not enough to train a reliable learning model, and the distribution of data collected by different organizations is difficult to be the same. In addition, various organizations protect their privacy and data security through data islands. Therefore, this paper proposes an efficient intrusion detection method using transfer learning and support vector machine with privacy-preserving (FETLSVMP). FETLSVMP performs aggregation of data distributed in various organizations through federated learning, then utilizes transfer learning and support vector machines build personalized models for each organization. Specifically, FETLSVMP first builds a transfer support vector machine model to solve the problem of data distribution differences among various organizations; then, under the mechanism of federated learning, the model is used for learning without sharing training data on each organization to protect data privacy; finally, the intrusion detection model is obtained with protecting the privacy of data. Experiments are carried out on NSL-KDD, KDD CUP99 and ISCX2012, the experimental results verify that the proposed method can achieve better results of detection and robust performance, especially for small samples and emerging intrusion behaviors, and have the ability to protect data privacy.