Incremental Support Vector Data Description Research Articles

Research on online anomaly detection methods for bearing degradation

In industrial applications, rolling bearings operate under conditions of high precision and high speed, and their physical and mechanical characteristics change with the increase in operating time. Traditional diagnostic methods struggle to adapt well to the changing characteristics of bearings for online anomaly detection. Therefore, this research proposes an online anomaly detection method for rolling bearings based on time-density-weighted incremental support vector data description (TISVDD). A classification strategy is proposed to prevent samples misclassification in the updating process. The Detection Boundary is established based on SVDD decision boundary to enhance the recognition of abnormal samples in the process of model updating. A dual-screening mechanism update strategy for support vectors is proposed. It involves establishing a preliminary screening mechanism based on the Elimination Boundary. On this basis, an in-depth screening mechanism based on time density weight is introduced by considering spatiotemporal characteristics of samples, enhancing the real-time performance of online anomaly detection for bearings. Building upon the fused dual-boundary SVDD, a TISVDD framework for online anomaly detection is proposed, enabling the detection model to dynamically update in response to data changes over time. To validate the effectiveness of the proposed method, experiments were conducted using the XJTU-SY bearing dataset and real-time datasets collected on an online hardware platform. The results demonstrate the effectiveness and superiority of the method in practical applications.

Kernel Function Selection for FISVDD Applied to Outliers Detection Application for Energy Time Series Datasets

Forecasting energy consumption is crucial for maintaining stability in the energy supply-demand balance. This process requires accurate data, necessitating deep preprocessing steps, with outlier detection being a pivotal task. While most existing methods are tailored for batch learning, smart grid data collection operates continuously, demanding online algorithms for effective data cleaning. This paper proposes an analytical study emphasizing the significance of selecting an appropriate kernel function for data description methods and its impact on forecasting efficiency. The Fast Incremental Support Vector Data Description (FISVDD) algorithm is chosen as a kernel-based approach for outlier detection in time series datasets. The evaluation involves different kernel functions on two datasets: Electrical Grid Stability Simulated Data and Individual Household Electric Power Consumption Dataset. Criteria such as the objective function and data distribution are considered to determine the most suitable kernel function. FISVDD's performance with the right kernel function is compared with other outlier detection methods, and the results are fed into a Gated Recurrent Unit model to visualize the impact on forecasting. Mean Squared Error evaluation reveals that FISVDD with the appropriate kernel function yields the best results, underscoring the strength of data description methods with the appropriate kernel function.

Novel Online Network Intrusion Detection System for Industrial IoT Based on OI-SVDD and AS-ELM

The Industrial Internet of Things (IIoT) should be equipped with computational resources to detect network intrusions, types of attacks, and update their models automatically in real time. The most challenging aspect of machine learning (ML)-based network intrusion detection system (NIDS) design to secure IIoT is the continuous need for up-to-date definitions of attack data records. Moreover, the approaches employed by cyber attackers are in a dynamic state with changing trends and techniques. Hence, conventional signature-based NIDS are not suitable since they cannot update obsolete detection models. Anomaly-based NIDS that contains an online learning technique is adopted in our proposed method. Since IIoTs face resource-constrained problems, such as low memory, low computing capacity, and limited energy supply, it is very challenging to implement the exiting general-purpose anomaly-based NIDS. This article proposes a lightweight NIDS based on an online incremental support vector data description (OI-SVDD) anomaly detection system on the IIoT devices and an adaptive sequential extreme learning machine (AS-ELM) on the multiaccess edge computing (MEC) server. Additionally, we utilize the MEC server, which provides computational resources to execute the AS-ELM model at the network’s edge. To avoid data saturation in the proposed model, we apply data filtering using the rate of convergence (ROC). We evaluated the proposed NIDS through experiments using two data sets, such as UNSW-NB15 (public data set) and our self-generated data set. Our results show that the proposed OI-SVDD and AS-ELM perform effectively and detect network intrusion in a realistic IIoT environment.

Active Incremental Learning for Health State Assessment of Dynamic Systems With Unknown Scenarios

This article is concerned with the data-driven health state assessment task with unknown scenarios. Unknown scenarios are inaccessible in the stage of model training but they can appear unexpectedly during the running stage. A new problem called the within-class distribution mismatch is raised by assuming that unknown scenarios still belong to known classes. To tackle this challenging problem, a novel active incremental learning scheme with a classifier and an out-of-distribution (OOD) detector is proposed. A <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> -fast incremental support vector data description ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> -FISVDD) model is put forward as the OOD detector to recognize distribution mismatch samples online for label annotation. Specifically, it integrates the clustering algorithm to build local support vector sets, based on which an active query strategy is developed. An incremental learning mechanism is also designed to reduce the labeling cost. Then, the new labeled data can simultaneously refine the classifier and OOD detector. Two cases, including a bearing benchmark dataset and the operation data of a practical deep-sea manned submersible, are studied to demonstrate the effectiveness of the proposed scheme.

Equipment Health Assessment Based on Improved Incremental Support Vector Data Description

With the rapid development of Internet-of-Things and big data, health assessment of equipment is receiving more attention in recent years. It is critical to bridge the gap between real-time production data and health status evaluation, which helps maintenance team understand the health status of equipment exactly, and then make rational maintenance plans. For this purpose, this paper proposes a framework to realize real-time equipment health assessment with health status quantitatively characterized by health degree (HD). The proposed framework begins with removing redundant features using a principal component analysis (PCA) method. Then, to represent the optimal operation status, a support vector data description (SVDD) algorithm is employed for extracting normal observations in the offline part. Thereafter, HD is introduced based on the Euclidean distance between current observation and the normal sample set. In order to achieve online updating of the normal sample set, and promote accuracy and computational efficiency of the offline part, an improved incremental SVDD algorithm based on adaptive threshold ${N}$ (NISVDD) is proposed. A case study is used to demonstrate the effectiveness of the proposed framework and model using a benchmark dataset of rolling bearing. Results suggest that the proposed framework is effective, and PCA shows good potential to extract features and keep most of the original information. The proposed NISVDD model is able to trace the dynamics of equipment health status for whole run-to-failure process, and outperforms other models in both accuracy and computational efficiency.

Acoustical damage detection of wind turbine blade using the improved incremental support vector data description

The blade is a crucial part of wind turbine for generating electricity and prone to damage due to harsh external environment. Accurate damage detection of wind turbine blade (WTB) is still a prominent challenge. This paper presents an acoustical detection method for damage identification of the WTB based on pattern recognition. In the proposed method, sound pulse extraction of the WTB is first investigated through physical method in combination with the filter and sliding window. Subsequently, the wavelet packet energy ratios of acoustic signal are introduced to characterize the discrepancy between intact and cracked sound pulses, and the support vector data description (SVDD) model is built for WTB damage detection. Besides, an improved incremental learning method is presented and employed to adaptively update the SVDD model, which aims at simplifying calculation procedure. Finally, the performance of proposed method is evaluated using experimental data collected from the WTBs with both intact and damaged conditions in commercial wind farms. It is demonstrated that proposed method has improvement in prediction accuracy compared to previous incremental SVDD models and performs the best on training time.

One-Class Classifier Based Fault Detection in Distribution Systems With Varying Penetration Levels of Distributed Energy Resources

The integration of Distributed Energy Resources (DERs) into distribution systems greatly increases the system complexity and introduces two-way power flow. Conventional protection schemes are based upon local measurements and simple linear system models, and are thus not capable of handling the new complexity and power flow patterns in systems with high DER penetration. In this paper, we propose a data-driven protection framework to address the challenges introduced by DERs. Firstly, considering the limited available data under fault conditions, we adopt the Support Vector Data Description (SVDD) method, a commonly used one-class classifier, for distribution system fault detection, which only requires the normal data for its training process. Secondly, incremental learning is incorporated into the proposed SVDD-based protection framework to accommodate variations of the integration level of DERs in distribution systems over time. In particular, the artificial uniform-hyperspherical data generation model is incorporated into the incremental SVDD to boost the training speed. Finally, we validate the proposed method under the IEEE 123-node test feeder. Simulation results demonstrate that our proposed SVDD-based fault detection framework significantly improves the robustness and resilience against DERs in comparison with conventional protection systems. Meanwhile, the proposed online updating model outperforms the existing incremental SVDD models in terms of successful training speed.

Fast Incremental SVDD Learning Algorithm with the Gaussian Kernel

Support vector data description (SVDD) is a machine learning technique that is used for single-class classification and outlier detection. The idea of SVDD is to find a set of support vectors that defines a boundary around data. When dealing with online or large data, existing batch SVDD methods have to be rerun in each iteration. We propose an incremental learning algorithm for SVDD that uses the Gaussian kernel. This algorithm builds on the observation that all support vectors on the boundary have the same distance to the center of sphere in a higher-dimensional feature space as mapped by the Gaussian kernel function. Each iteration involves only the existing support vectors and the new data point. Moreover, the algorithm is based solely on matrix manipulations; the support vectors and their corresponding Lagrange multiplier αi’s are automatically selected and determined in each iteration. It can be seen that the complexity of our algorithm in each iteration is only O(k2), where k is the number of support vectors. Experimental results on some real data sets indicate that FISVDD demonstrates significant gains in efficiency with almost no loss in either outlier detection accuracy or objective function value.

Single-Class Data Descriptors for Mapping Panax notoginseng through P-Learning

Machine learning-based remote-sensing techniques have been widely used for the production of specific land cover maps at a fine scale. P-learning is a collection of machine learning techniques for training the class descriptors on the positive samples only. Panax notoginseng is a rare medicinal plant, which also has been a highly regarded traditional Chinese medicine resource in China for hundreds of years. Until now, Panax notoginseng has scarcely been observed and monitored from space. Remote sensing of natural resources provides us new insights into the resource inventory of Chinese materia medica resources, particularly of Panax notoginseng. Generally, land-cover mapping involves focusing on a number of landscape classes. However, sometimes a subset or one of the classes will be the only part of interest. In term of this study, the Panax notoginseng field is the right unit class. Such a situation makes single-class data descriptors (SCDDs) especially significant for specific land-cover interpretation. In this paper, we delineated the application such that a stack of SCDDs were trained for remote-sensing mapping of Panax notoginseng fields through P-learning. We employed and compared SCDDs, i.e., the simple Gaussian target distribution, the robust Gaussian target distribution, the minimum covariance determinant Gaussian, the mixture of Gaussian, the auto-encoder neural network, the k-means clustering, the self-organizing map, the minimum spanning tree, the k-nearest neighbor, the incremental support vector data description, the Parzen density estimator, and the principal component analysis; as well as three ensemble classifiers, i.e., the mean, median, and voting combiners. Experiments demonstrate that most SCDDs could achieve promising classification performance. Furthermore, this work utilized a set of the elaborate samples manually collected at a pixel-level by experts, which was intended to be a benchmark dataset for the future work. The measuring performance of SCDDs gives us challenging insights to define the selection criteria and scoring proof for choosing a fine SCDD in mapping a specific landscape class. With the increment of remotely sensed satellite data of the study area, the spatial distribution of Panax notoginseng could be continuously derived in the local area on the basis of SCDDs.

Classification of maize seeds of different years based on hyperspectral imaging and model updating

Seed classification and identification exhibit potential for detecting seed purity and increasing crop yield. In this study, hyperspectral imaging was employed to develop classification methods for maize seeds. A total of 2000 seeds, including four varieties of maize seeds of different years, were evaluated. Hyperspectral reflectance images were acquired between 400nm and 1000nm. Classification models based on the mean spectral features of seeds were developed using least squares support vector machine (LSSVM). Model updating using incremental support vector data description was also applied to update the LSSVM model online and ensure accurate identification of maize seeds of different years. The classification accuracy of the LSSVM model combined with model updating reached 94.4% and was 10.3% higher than that of other non-updated models. This study showed that combined hyperspectral imaging and model updating could be an effective method for classification of seeds of different years.

Online fault diagnosis method based on Incremental Support Vector Data Description and Extreme Learning Machine with incremental output structure

Online fault diagnosis system should be able to detect faults, recognize fault types and update the discriminating ability and knowledge of itself automatically in real time. But the class number in fault diagnosis is not constant and it is in a dynamic state with new members enrolled. The traditional recognition algorithms are not able to update diagnosis system efficiently when the class number of failure modes is increasing. To solve the problem, an online fault diagnosis method based on Incremental Support Vector Data Description (ISVDD) and Extreme Learning Machine with incremental output structure (IOELM) is proposed. ISVDD is used to find a new failure mode quickly in the continuous condition monitoring of the equipments. The fixed structure of Extreme Learning Machine is changed into an elastic structure whose output nodes could be added incrementally to recognize the new fault mode efficiently. Recognition experiments on the diesel engine under eleven different conditions show that the online fault diagnosis method based on ISVDD and IOELM works well, and the method is also feasible in fault diagnosis of other mechanical equipments.

Incremental Study Based on Support Vector Used to Anomaly Detection for Hyperspectral Imagery

While detecting anomalies in hyperspectral imagery with support vector data description (SVDD), large numbers of operation was run because of the high dimension character of dataset and the complexity of background and the high miss rate was discovered because of the interfered background by interior anomalies. This paper used incremental support vector data description (ISVDD) method that samples are divided many sub-sample, and incremental study is designed to simplify the computation. On every sub-sample study, optimization is needed according of the support vectors obtained from above sub-sample and current sub-sample data. By the experiment on the HYMAP data, computation complexity of the algorithm decrease obviously and the computation speed increase highly under the similar detection effect compared with SVDD algorithm.

Incremental Study Based on Support Vector Used to Anomaly Detection for Hyperspectral Imagery

While detecting anomalies in hyperspectral imagery with support vector data description (SVDD), large numbers of operation was run because of the high dimension character of dataset and the complexity of background and the high miss rate was discovered because of the interfered background by interior anomalies. This paper used incremental support vector data description (ISVDD) method that samples are divided many sub-sample, and incremental study is designed to simplify the computation. On every sub-sample study, optimization is needed according of the support vectors obtained from above sub-sample and current sub-sample data. By the experiment on the HYMAP data, computation complexity of the algorithm decrease obviously and the computation speed increase highly under the similar detection effect compared with SVDD algorithm.