Small Sample Case Research Articles

Explainable fault diagnosis method for process flow based on data augmentation with system graph relationship

The main objective of process industrial systems is to ensure safe operation. Signal fluctuations resulting from equipment failures and variable abnormalities can progressively propagate over the entire system, potentially resulting in fault conditions. However, the sporadic occurrence of fault conditions leads to a scarcity of data samples for diagnosis models. A fault diagnosis method for process flow based on data augmentation with system graph relationships is proposed. Propagation paths of fault signals are generated through graph relationships of systems. Simultaneously, the data of multiple nodes in paths is transformed to generate augmentation samples. To ensure comprehensive learning of the information included in augmentation samples, a large number of samples are generated for pre-training based on real-world data from a system. The explainability analysis method is used to analyze the focus area of diagnostic networks on samples with and without augmentation. The experimental results demonstrate that the proposed method can substantially improve diagnostic accuracy and explainability, hence offering advantages for diagnosis tasks in small sample cases.

Reliability test for degradation data based on ranked set sampling

AbstractIn this article, we consider test for the two null hypotheses for and , two widely useful tests in reliability, based on ranked set sampling (RSS). We derive the likelihood ratio test as well as the associated exact and asymptotic results. Considering a fixed significance level and power of the test, we show that the proposed test statistic outperforms the existing test. In small sample cases, the proposed test leads to a much narrower confidence interval for the reliability function . Then, the test statistics obtained from simple random sampling and RSS schemes are compared through which, the efficiency of using RSS is demonstrated. For illustration, we apply the proposed test to a degradation data from the reliability literature. Upon using RSS, the cost of measurement gets reduced and efficiency gets improved, suggesting the importance and use of RSS data in reliability experiments and their design.

Deep discriminative sparse representation learning for machinery fault diagnosis

The high complexity of actual machinery vibration environments introduces various interferences into vibration signals, making it challenging to eliminate redundant information and extract discriminative features of mechanical faults. Shallow-structured sparse classification models enable intelligent fault diagnosis but struggle with deeper feature extraction, especially in small-sample cases. In this paper, we introduce a deep discriminative sparse representation learning (DDSRL) framework with a deep architecture for this issue. In DDSRL, the first-layer dictionary atoms are automatically learned from raw signals and borrowed for sparse coding to enhance sparsity and interclass discriminability. Then, the weight matrix is defined to screen high-energy atoms from the upper dictionary as new training samples for the next dictionary learning and sparse coding layers, thereby eliminating interclass redundant atoms and mining deeper discriminative representations of cyclostationary impulses or meshing harmonics from the shallow dictionary. Finally, DDSRL extracts feature sequences sensitive to fault-induced waveforms via the learned dictionaries and coding coefficients for intelligent fault identification. Experimental verifications on two datasets with high-speed bearing and compound gear-bearing faults demonstrate that DDSRL outperforms six popular sparse classification models, sparse autoencoder, convolutional neural network, and principal component analysis network. Furthermore, the layer-by-layer feature extraction process of DDSRL and feature visualization analysis provide more credible fault diagnosis results.

Profits Before Patients? Rethinking the Shareholder Primacy Model of Corporate Governance for Health Care

Concerns about the ballooning cost of health care and medicine are at the heart of perpetual debate about the corporate accountability of health care corporations. This essay critiques the shareholder primacy model of corporate governance for health care enterprise using the American pharmaceutical company Merck as a small sample case study that might be indicative of broader corporate governance issues generally, that deserve further research. In 2009 an Australian Federal Court class action was led by Merck product patient users against Merck’s Australian subsidiary, for allegedly failing to withdraw the drug Vioxx from the market. The patient group claimed that Merck insiders knew Vioxx increased the risk of heart attacks, yet persisted to manufacture and market the drug to secure profits for shareholders. At first instance, the trial judge awarded damages in favor of the original plaintiff claimant, finding that the drug had caused the claimant's alleged heart risk injury. However that decision was overturned on appeal, with further leave to appeal to the High Court of Australia being refused on 11 May 2012. Simultaneously, settlements between the company and other groups of claimants involving the same drug, have been and are continuing to be reached in other jurisdictions around the world in 2013.The case more generally evokes a familiar normative question of to whom a duty of care is and should be owed by health care corporate decision makers: shareholders, patients, or both? This paper briefly answers that question by critically analysing the market impact of the shareholder primacy model of corporate governance, on end user health care patients by transforming them from patients into consumers. The paper then constructs a potential solution in the form of an evolved shareholder primacy model of corporate governance which encompasses a value duty of virtue to shareholders that potentially meets the interests of both shareholders and patients.

Digital twin-assisted dual transfer: A novel information-model adaptation method for rolling bearing fault diagnosis

Rolling bearing fault diagnosis is of great importance to the safety management of mechanical equipment. The scarcity of labelled fault data makes it difficult to adequately perform the training process of intelligent diagnosis models, and this will result in these intelligent models not being effectively and widely used in practice. Although some recent studies have verified that the addition of dynamic model response to the training process will greatly improve the ability of the model with low cost and high efficiency, it is still stuck in poor effect caused by large information distribution difference between dynamic model response and real measured data. Focusing on this issue, a digital twin-assisted dual transfer (DTa-DT) method with information and model adaptation was proposed for rolling bearing fault diagnosis. Different from the traditional digital-analogue driven transfer methods, the proposed DTa-DT aims to simultaneously synthesize data information transfer and feature model transfer together with domain transfer error minimization. In particular, it should be noted that the DTa-DT architecture consists of a dual transfer learning process, including digital twin-driven information transfer (DTd-IT) and digital-analogue-driven model transfer (DAd-MT), where the information is collaborated with the model to improve the integrated transfer diagnosis effect under sampling. On one aspect, with the employment of bearing dynamic model responses, DTd-IT is innovatively designed to establish the transfer of dynamic information and measured information. The information distribution difference between these twin data and real measured data is effetely adjusted with the introduced actual inference components, where the twin data with low information distribution difference can be well fusion generated by the information transfer digital twin (ITDT) model. On the other aspect, considering the truth that there are still small sample cases of real measured data and information distribution differences will affect the quality of the twin data, a digital-analogue driven model transfer (DAd-MT) method is further proposed, where the deep branch transfer network (DBTN) model with improved convolutional neural network (CNN) is used to achieve an accurate fault diagnosis effect with the help of digital twin data. Experiments and wear analysis verified that the proposed DTa-DT can significantly reduce the distribution difference between the dynamic model response and the real measured data, thus achieving low-cost and efficient rolling bearing transfer diagnosis compared to other ten state-of-the-art deep learning models. It can be predicted that the proposed dual transfer architecture provides more opportunities for the practical application of intelligent fault diagnosis under small sample sizes.

TBTA-D2Net: a novel hyperspectral image classification method based on triple-branch ternary-attention mechanism and Dense2Net

ABSTRACTRecently, there has been a growing interest in the hyperspectral image (HSI) classification methods that employ deep learning techniques in small sample cases. To address issues with network degradation and enhance the extraction of discriminative HSI features, this article proposes a TBTA-D2Net network utilizing a triple-branch ternary-attention mechanism and Dense2Net. Furthermore, a new deep model optimizer named Adan is introduced to improve the training speed of the network model. This article takes spatial information as a two-dimensional vector, extracting spectral features as well as spatial-X and spatial-Y features separately in three branches. Each branch includes a Dense2Net bottleneck module and an attention module. Classification is achieved by fusing the features extracted from the three branches. Experimental results on four public datasets indicate that TBTA-D2Net can achieve competitive results over state-of-the-art methods. The code is available at https://github.com/TeresaTing/TBTA-D2Net.

Selection strategy of uniform expert evaluation scale in group decision making

The scoring habits of experts are embodied in expert evaluation scale (EES). We show that many inaccurate consensus and ranking results can occur in group decision making (GDM) problems without considering EES. To remove the impacts of EESs, we propose the definition of EES and the uniform EES (UEES) strategy. Then, we prove the influence of the selection of UEES strategy on two consistency measures and two aggregation operators. In real-world GDM problems, a decision maker (DM) usually hopes that the final scores of alternatives are subject to his/her desired distribution. Therefore, a novel ranking method by UEES strategy is proposed to obtain DM’s desired final scores. Finally, a numerical example is used to clarify the effectiveness of the proposed method, and a simulation experiment shows that the proposed method is more suitable for large sample cases but the accuracy is still at a high level under small sample cases.

O-151 Intrahepatic cholestasis of pregnancy and relevant neonatal outcomes in IVF versus spontaneous conception: a prediction nomogram-based study

Abstract Study question Does the incidence of intrahepatic cholestasis of pregnancy (ICP) differ in spontaneous conception(SC) versus in vitro fertilization (IVF) conception? Summary answer We observed that the ICP rate was higher in IVF than in SC, and first built a prediction nomogram to find the predictors of ICP. What is known already There are limited clinical studies comparing intrahepatic cholestasis of pregnancy(ICP) and neonatal outcomes in puerperae who spontaneously conceived and those who conceived via assisted reproductive technology. Several studies comparing the maternal, laboratory, and perinatal characteristics of in-vitro fertilization(IVF) and spontaneous pregnancies concerning ICP show that IVF treatment has an increased risk of ICP in singleton and multiple deliveries. However, given that it is difficult to collect big data, most research on IVF patients with ICP involve small sample size or case reports. Study design, size, duration This was a retrospective real-world data study that linked the information from puerperae and neonates. We included 4,467 puerperae who conceived via ART, and 28,336 puerperae who conceived spontaneously. Participants/materials, setting, methods Cochran–Mantel–Haenszel (CMH) analysis and a general linear model (GLM) were used to control bias. We compared the related serum-derived indicators and neonatal outcomes of ICP patients with in vitro fertilization (IVF) and SC. Multivariate logistic regression analysis, a forest plot, and nomogram were used to assess impact factors and risk prediction. Main results and the role of chance Logistic analysis adjusted for confounders revealed significant differences in the ICP rate of singleton delivery (4.24% vs. 3.41%, adjusted OR = 1.26 [95% confidence interval (CI) 1.03–1.53], P=0.025) and in groups with total bile acids(TBA) ≥40 and <100 µmol/L (14.77% vs. 10.39%, aOR=1.31[95% CI 1.06–1.63], P=0.023) between IVF and SC. When we divided newborns into singleton and twins delivery, the GLM revealed a higher rate with Apgar score <7 (13.44% vs. 3.87%; aOR=3.85 [95% CI: 2.07–7.17], P <0.001) and fetal distress for IVF in comparison with SC (19.32% vs. 5.55%; OR = 3.48 [95%CI: 2.39–6.95], P <0.001) in the singleton group. In multivariate logistic regression analysis, five factors were independent predictors of ICP: body mass index (BMI) (aOR=1.75 [95% CI 1.03–2.13], P=0.036), number of embryo transferred (ET) (single ET vs. double ET: aOR=4.82 [95% CI 3.83–6.05], P<0.001), E2 level on the ET day (aOR=2.79 [95% CI 1.79–4.05], P=0.011), fresh ET which compared with frozen ET (FET) (aOR=1.40 [95% CI 1.09–1.80], P=0.008), and severe ovarian hyperstimulation syndrome which compared with non-OHSS (aOR=3.97 [95% CI 1.79–8.80], P<0.001). These predictive factors in the logistic regression model were integrated into the nomogram (C-index=0.735 [95% CI, 0.702–0.764]); for each patient, higher total points indicated a higher risk of ICP. Limitations, reasons for caution The limitations of this study include its retrospective nature, and the balance in baseline characteristics between IVF and SC measured using statistical methods. Wider implications of the findings Our results provide evidence for the incidence of ICP between spontaneously conceived and those who conceived via IVF, and found predictors of ICP in IVF treatment. It could assist physicians in making clinical decisions avoiding risk factors during IVF, and taking preventive countermeasures for patients. Trial registration number not applicable

Fast inspection and accurate recognition of target objects for astronaut robots through deep learning

Target object detection based on deep learning and image processing technology is widely used in robot astronauts. However, existing detection methods have limitations in detection speed and accuracy due to the complex space environment (such as uneven illumination and particle radiation) and inadequate training samples. Inspired by the neural network structure and transfer learning, a deep learning detection method for small samples in complex environments is proposed. A depthwise separable convolution is added to a feature fusion network to reduce the number of parameters in image output feature mapping, and a linear bottleneck inverted residual structure is introduced into a backbone feature extraction network to reduce the computation and memory requirements during feature extraction. As a result, a backbone feature extraction–fusion network structure is established to solve the problem in detection speed. A squeeze-and-excitation (SE) attention module is introduced in front of the head, and an SE detector is constructed to improve the detection accuracy in a spatially complex environment by dynamically assigning image channel weights to highlight the target object features in blurred images. The learning efficiency and accuracy of the network model in the small sample case in this paper are addressed by incorporating the transfer learning idea and establishing the evaluation function of learning samples. Experimental results show that the proposed algorithm enables astronaut robots to detect object rapidly and accurately in complex environments. The average speed (frames per second) and accuracy of detection under 2200 training samples are 45.19 and 93.14%, respectively.

Approximate bias for first-order autoregressive model with uniform innovations. Small sample case

Approximate bias for first-order autoregressive model with uniform innovations. Small sample case

Reliability inference and remaining useful life prediction for the doubly accelerated degradation model based on Wiener process

<abstract><p>Degradation data are an important source of products' reliability information. Though stochastic degradation models have been widely used for fitting degradation data, there is a lack of efficient and accurate methods to get their confidence intervals, especially in small sample cases. In this paper, based on the Wiener process, a doubly accelerated degradation test model is proposed, in which both the drift and diffusion parameters are affected by the stress level. The point estimates of model parameters are derived by constructing a regression model. Furthermore, based on the point estimates of model parameters, the interval estimation procedures are developed for the proposed model by constructing generalized pivotal quantities. First, the generalized confidence intervals of model parameters are developed. Second, based on the generalized pivotal quantities of model parameters, using the substitution method the generalized confidence intervals for some interesting quantities, such as the degradation rate $ \mu_0 $, the diffusion parameter $ \sigma_0^2 $, the reliability function $ R(t_0) $ and the mean lifetime $ E(T) $, are obtained. In addition, the generalized prediction intervals for degradation amount $ X_0(t) $ and remaining useful life at the normal use stress level are also developed. Extensive simulations are conducted to investigate the performances of the proposed generalized confidence intervals in terms of coverage percentage and average interval length. Finally, a real data set is given to illustrate the proposed model.</p></abstract>

Unit root test in the presence of a single additive outlier small sample case

Unit root test in the presence of a single additive outlier small sample case

Bias-Corrected Maximum Likelihood Estimation and Bayesian Inference for the Process Performance Index Using Inverse Gaussian Distribution

In this study, the estimation methods of bias-corrected maximum likelihood (BCML), bootstrap BCML (B-BCML) and Bayesian using Jeffrey’s prior distribution were proposed for the inverse Gaussian distribution with small sample cases to obtain the ML and Bayes estimators of the model parameters and the process performance index based on the lower specification process performance index. Moreover, an approximate confidence interval and the highest posterior density interval of the process performance index were established via the delta and Bayesian inference methods, respectively. To overcome the computational difficulty of sampling from the posterior distribution in Bayesian inference, the Markov chain Monte Carlo approach was used to implement the proposed Bayesian inference procedures. Monte Carlo simulations were conducted to evaluate the performance of the proposed BCML, B-BCML and Bayesian estimation methods. An example of the active repair times for an airborne communication transceiver is used for illustration.

Specific Emitter Identification Model Based on Improved BYOL Self-Supervised Learning

Specific emitter identification (SEI) is extracting the features of the received radio signals and determining the emitter individuals that generate the signals. Although deep learning-based methods have been effectively applied for SEI, their performance declines dramatically with the smaller number of labeled training samples and in the presence of significant noise. To address this issue, we propose an improved Bootstrap Your Own Late (BYOL) self-supervised learning scheme to fully exploit the unlabeled samples, which comprises the pretext task adopting contrastive learning conception and the downstream task. We designed three optimized data augmentation methods for communication signals in the former task to serve the contrastive concept. We built two neural networks, online and target networks, which interact and learn from each other. The proposed scheme demonstrates the generality of handling the small and sufficient sample cases across a wide range from 10 to 400, being labeled in each group. The experiment also shows promising accuracy and robustness where the recognition results increase at 3-8% from 3 to 7 signal-to-noise ratio (SNR). Our scheme can accurately identify the individual emitter in a complicated electromagnetic environment.

Local Matrix Feature-Based Kernel Joint Sparse Representation for Hyperspectral Image Classification

Hyperspectral image (HSI) classification is one of the hot research topics in the field of remote sensing. The performance of HSI classification greatly depends on the effectiveness of feature learning or feature design. Traditional vector-based spectral–spatial features have shown good performance in HSI classification. However, when the number of labeled samples is limited, the performance of these vector-based features is degraded. To fully mine the discriminative features in small-sample case, a novel local matrix feature (LMF) was designed to reflect both the correlation between spectral pixels and the spectral bands in a local spatial neighborhood. In particular, the LMF is a linear combination of a local covariance matrix feature and a local correntropy matrix feature, where the former describes the correlation between spectral pixels and the latter measures the similarity between spectral bands. Based on the constructed LMFs, a simple Log-Euclidean distance-based linear kernel is introduced to measure the similarity between them, and an LMF-based kernel joint sparse representation (LMFKJSR) model is proposed for HSI classification. Due to the superior performance of region covariance and correntropy descriptors, the proposed LMFKJSR shows better results than existing vector-feature-based and matrix-feature-based support vector machine (SVM) and JSR methods on three well-known HSI data sets in the case of limited labeled samples.

Small-sample continual learning classification method with vaccine to update memory cells based on the artificial immune system

In this paper, a novel continual learning classification method (SCLM) in small sample cases is proposed, which inspired by the immune system's continuous improvement of immunity through injecting vaccines. Data-driven classification method requires a large number of historical data to establish a pattern recognition model with good generalization performance. However, in practice, the data that can be used for training is usually small and unbalanced, which lead to poor classification accuracy. In addition, batch learning method cannot improve continually classification performance by learning test phase data. In view of the above problems, SCLM generates sample as vaccine by finding the group center of training samples, so that B cells mature and activate memory cells in the train phase. In the test phase, the recognition ability of SCLM is further improved by learning new samples and updating memory cells. In order to evaluate its performance under the condition of less training samples and its possible advantages, the experiments on well-known datasets in UCI repository and reciprocating compressor faults diagnose were performed. The results show that SCLM has better classification performance than other methods when the number of training samples is insufficient. At the same time, the method of generating data has significantly improved the classification performance of other methods.

A Fault-Line Selection Method for Small-Current Grounded System Based on Deep Transfer Learning

Usually, data-driven methods require many samples and need to train a specific model for each substation instance. As different substation instances have similar fault features, the number of samples required for model training can be significantly reduced if these features are transferred to the substation instances that lack samples. This paper proposes a fault-line selection (FLS) method based on deep transfer learning for small-current grounded systems to solve the problems of unstable training and low FLS accuracy of data-driven methods in small-sample cases. For this purpose, fine-turning and historical averaging techniques are proposed for use in transfer learning to extract similar fault features from other substation instances and transfer these features to target substation instances that lack samples to improve the accuracy and stability of the model. The results show that the proposed method obtains a much higher FLS accuracy than other methods in small-sample cases; it has a strong generalization ability, low misclassification rate, and excellent application value.

A novel remaining useful life prediction method based on multi-support vector regression fusion and adaptive weight updating

Remaining useful life prediction is of huge significance in preventing equipment malfunctions and reducing maintenance costs. Currently, machine learning algorithms have become hotspots in remaining useful life prediction due to their high flexibility and convenience. However, machine learnings require large amounts of data, and their prediction performance depends heavily on the selection of hyper-parameters. To overcome these shortcomings, a novel remaining useful life prediction method for small sample cases is proposed based on multi-support vector regression fusion. In the offline training phase, the fusion model is established, consisting of multiple support vector regression sub-models To obtain the optimal sub-model parameters, the Bayesian optimization algorithm is applied and an improved optimization target is formulated with various metrics describing regression and prediction performance. In the online prediction phase, an adaptive weight updating algorithm based on dynamic time warping is developed to measure the fitness of each sub-model and determine the corresponding weight value. The C-MAPSS engine dataset is used to test the performance of the proposed method, along with some existing machine learning methods as comparison. The proposed method only requires 30% of the training data sample to achieve high accuracy, with a root mean square error of 14.98, which is superior to other state-of-the-art methods. The results demonstrate the superiority of the proposed method.

Relevance vector machine with optimal hybrid kernel function for electric vehicles ownership forecasting: The case of China

Electric Vehicles (EVs) ownership forecasting is of great significance to deal with the opportunities and challenges brought by the rapid development of EVs. However, most of the existing EVs ownership forecasting methods are model-driven methods that based on artificial experience, leading to the low reliability of the forecasting results. An ideal choice is data-driven method that can automatically mine regression relationships from available data. In this paper, we propose a data-driven method based on Relevance Vector Machine (RVM), which is suitable for small sample cases such as EVs ownership forecasting and can automatically find the changing pattern of EVs ownership to make high precision forecasting results. In the proposed method, an optimal hybrid kernel function is proposed to further improve the forecasting effect of RVM. Besides, factors affecting EVs are analyzed comprehensively to determine the input variables of the proposed method. The simulation results based on the data from 2011 to 2019 show that the proposed method outperforms some existing mature methods and other data-driven methods. Then, the EVs ownership forecasting results of China up to 2030 are given based on the proposed method.

Estimating and backtesting risk under heavy tails

While the estimation of risk is an important question in the daily business of banking and insurance, many existing plug-in estimation procedures suffer from an unnecessary bias. This often leads to the underestimation of risk and negatively impacts backtesting results, especially in small sample cases. In this article we show that the link between estimation bias and backtesting can be traced back to the dual relationship between risk measures and the corresponding performance measures, and discuss this in reference to value-at-risk, expected shortfall and expectile value-at-risk.Motivated by the consistent underestimation of risk by plug-in procedures, we propose a new algorithm for bias correction and show how to apply it for generalized Pareto distributions to the i.i.d. setting and to a GARCH(1,1) time series. In particular, we show that the application of our algorithm leads to gain in efficiency when heavy tails or heteroscedasticity exists in the data.