Flow shop failure prediction problem based on Grey-Markov model

Predicting telemetry data is vital for the proper operation of orbiting spacecraft. The Grey–Markov model with sliding window (GMSW) combines Grey model (GM (1, 1)) and Markov chain forecast model, which allows it to describe the fluctuation of telemetry data. However, the Grey–Markov model with sliding window does not provide better predictions of telemetry series with the pseudo-periodic phenomenon. To overcome this drawback, we improved the GMSW model by applying particle swarm optimization (PSO) algorithm a sliding window for better prediction of spacecraft telemetry data (denoted as PGMSW model). In order to produce more accurate predictions, background-value optimization is specially carried out using the particle swarm optimization technique in conventional GM (1, 1). For verifying PGMSW, it is utilized in the prediction of the cyclic fluctuation of telemetry series data and exponential variations therein. The simulation results indicate that the PGMSW model provides accurate solutions for prediction problems similar to the pseudo-periodic telemetry series.

A failure prediction Bayesian Networks model (FPBN) is brought forward by expanding the traditional Bayesian network (BN) methodology and tailored to deal with failure prediction problems. In the FPBN model, the classification of network nodes: failure detection nodes, failure cause nodes and failure mode nodes, and the algorithm to initialize orientations of network edges are illustrated. Moreover, with the conditional probability distributions, the implementation of the FPBN model in failure mode prediction, failure cause diagnosis and failure detection efficiency analysis are also described in details. At last, the simulation case study of a helicopter convertor is carried out and the comparison results show that, the FPBN model could perform effectively under uncertainty.

A disk failure prediction method based on LSTM network due to its individual specificity

Equipment utilized in intelligent underground coal mines necessitates clear distinction from general equipment due to its unique exposure to the specialized subterranean environment. This exposure results in distinct failure patterns, usage, and maintenance protocols. Historically, issues such as incomplete failure data records have impeded accurate failure prediction, leading to a reliance on experience-based maintenance practices for underground mining equipment. To enhance the accuracy of fault prediction for coal mine machinery, this paper proposes a two-stage fault prediction method based on Grey–Markov method. Leveraging daily maintenance data of intelligent dispatching center, this study employs a Grey model to generate initial failure rate forecasts, followed by the integration of the Markov model to refine prediction accuracy. Based on the daily maintenance data from the intelligent dispatching center, the anchor digging integrated machine and conveyor are selected as the case study equipment. The results show that the prediction accuracy of Grey–Markov model is higher than that of the control group.

We consider a parametric framework for predicting the future number of failures based on censored and right truncated field data. This type of data regularly occurs in industrial field data with reporting delays. This article presents methods to account for the truncation in this prediction problem. Our simulations show the influence of truncation on the coverage of the resulting prediction interval. The coverage properties of the introduced methods are evaluated. The adjusted prediction intervals show improved coverage properties compared with the unmodified prediction interval based on the truncated data.

In the petroleum industry, multivariate time series data is commonly used to monitor the performance of their assets, in which wells artificial lift systems are among the key assets that bring oil up to the surface. Failures frequently occur among these artificial lift systems, and they can greatly increase the operational expense due to loss of production and cost of repairs (also known as workovers). Predicting these failures before they occur can dramatically improve operational performance, such as by adjusting operating parameters to forestall failures or by scheduling maintenance to reduce unplanned repairs and to minimize downtime. Artificial lift failure prediction problem poses interesting challenges to data mining algorithms, because of the many real-world data issues, such as noise, missing data, delay of failure event logs, and large variability among normally functioning well artificial lift units. This paper presents the Smart Engineering Apprentice (SEA) framework that incorporates robust feature extraction algorithm, clustering and semi-supervised learning techniques, to enable learning of failure/normal patterns from noisy and poorly labeled multivariate time series, while achieving a high recall and precision for failures for real-world dataset.

Prognostics and Health Management (PHM) is an emerging engineering discipline which is concerned with the analysis and prediction of equipment health and performance. One of the key challenges in PHM is to accurately predict impending failures in the equipment. In recent years, solutions for failure prediction have evolved from building complex physical models to the use of machine learning algorithms that leverage the data generated by the equipment. However, failure prediction problems pose a set of unique challenges that make direct application of traditional classification and prediction algorithms impractical. These challenges include the highly imbalanced training data, the extremely high cost of collecting more failure samples, and the complexity of the failure patterns. Traditional oversampling techniques will not be able to capture such complexity and accordingly result in overfitting the training data. This paper addresses these challenges by proposing a novel algorithm for failure prediction using Generative Adversarial Networks (GAN-FP). GAN-FP first utilizes two GAN networks to simultaneously generate training samples and build an inference network that can be used to predict failures for new samples. GAN-FP first adopts an infoGAN to generate realistic failure and non-failure samples, and initialize the weights of the first few layers of the inference network. The inference network is then tuned by optimizing a weighted loss objective using only real failure and non-failure samples. The inference network is further tuned using a second GAN whose purpose is to guarantee the consistency between the generated samples and corresponding labels. GAN-FP can be used for other imbalanced classification problems as well. Empirical evaluation on several benchmark datasets demonstrates that GAN-FP significantly outperforms existing approaches, including under-sampling, SMOTE, ADASYN, weighted loss, and infoGAN augmented training.

Thus far, we’fve mostly discussed EVT in the context of distributions with abstract relationships to visual data. In order to make the transition from theory to practice, we will assume that scores are available as samples drawn from some distribution reflecting the output of a measurable recognition function. The distance or similarity score produced by a recognition function (e.g., a distance calculation between two vectors or a machine learning-based model) is a primary artifact of the pattern recognition process itself, and can tell us much about specific instances of visual perception. Most typically in automatic visual recognition, little, if anything, is currently done with these scores beyond checking the sign to assign class membership, or perhaps a quick comparison against an ad hoc threshold to accept or reject a sample. At a fundamental level, we can ask what exactly a recognition score is and why it is important for decision making. Further, we can model distributions of scores to determine if they were generated by a matching or non-matching process. Once we understand this basic model, we can then extend it to other modes such as score normalization and calibration (see Chapters 4 and 5). In this chapter, we will examine the problem of failure prediction as a case study embodying these concepts.

Purpose– This paper aims to review the existing literature systematically so as to contribute towards a better understanding of methodological problems of the classical statistical techniques, artificially intelligent expert systems and theoretical approaches to solve the corporate failure syndrome.Design/methodology/approach– This paper presented a systematic review of 83 articles reporting 137 prediction failure models published within 1966-2012 in scholarly reviewed journals in four main disciplines, namely, accounting, finance, banking and economics. The authors performed the systematic literature review with five main sources, namely, Science Direct, Google Scholar, Wiley Interscience, Metalib, Web of Science and Business Source Complete of the Social Sciences. The review modified the approaches used by Aziz and Dar (2006), Ravi and Ravi (2007) and Balcaen and Ooghe (2006).Findings– The results indicate significant body of prior literature on prediction of corporate failure, but a theoretically sound, highly accurate, simple and widely used corporate failure prediction model for stakeholders has yet to be developed.Originality/value– This paper contributes towards a systematic understanding of the methodological problems associated with the statistical, artificially intelligent expert systems and theoretical approaches to solve the corporate failure prediction problems faced by firms in 11 countries.

Medicine will experience many changes in the coming years because the so-called "medicine of the future" will be increasingly proactive, featuring four basic elements: predictive, personalized, preventive, and participatory. Drivers for these changes include the digitization of data in medicine and the availability of computational tools that deal with massive volumes of data. Thus, the need to apply machine-learning methods to medicine has increased dramatically in recent years while facing challenges related to an unprecedented large number of clinically relevant features and highly specific diagnostic tests. Advances regarding data-storage technology and the progress concerning genome studies have enabled collecting vast amounts of patient clinical details, thus permitting the extraction of valuable information. In consequence, big-data analytics is becoming a mandatory technology to be used in the clinical domain.Machine learning and big-data analytics can be used in the field of cardiology, for example, for the prediction of individual risk factors for cardiovascular disease, for clinical decision support, and for practicing precision medicine using genomic information. Several projects employ machine-learning techniques to address the problem of classification and prediction of heart failure (HF) subtypes and unbiased clustering analysis using dense phenomapping to identify phenotypically distinct HF categories. In this chapter, these ideas are further presented, and a computerized model allowing the distinction between two major HF phenotypes on the basis of ventricular-volume data analysis is discussed in detail.

Adaptive stochastic-filter-based failure prediction model for complex repairable systems under uncertainty conditions

The computer numerically controlled (CNC) system is the key functional component of CNC machine tool control systems, and the servo drive system is an important part of CNC systems. The complex working environment will lead to frequent failure of servo drive systems. Taking effective health management measures is the key to ensure the normal operation of CNC machine tools. In this paper, the comprehensive effect of fault prediction and fault diagnosis is considered for the first time, and a health management system for machine tool servo drive systems is proposed and applied to operation and maintenance management. According to the data collected by the system and related indicators, the technology can predict the state trend of equipment operation, identify the hidden fault characteristics in the data, and further diagnose the fault types. A health management system mainly includes fault prediction and fault diagnosis. The core of fault prediction is the gated recurrent unit (GRU). The attention mechanism is introduced into a GRU neural network, which can solve the long-term dependence problem and improve the model performance. At the same time, the Nadam optimizer is used to update the model parameters, which improves the convergence speed and generalization ability of the model and makes it suitable for solving the prediction problem of large-scale data. The core of fault diagnosis is the self-organizing mapping (SOM) neural network, which performs cluster analysis on data with different characteristics, to realize fault diagnosis. In addition, feature standardization and principal component analysis (PCA) are introduced to balance the influence of different feature scales, enhance the feature of fault data, and achieve data dimensionality reduction. Compared with the other two algorithms and their improved versions, the superiority of the health management system with high-dimensional data and the enhancement effect of fault identification are verified. The relative relationship between fault prediction and diagnosis is further revealed, and the adjustment idea of the production plan is provided for decision makers. The rationality and effectiveness of the system in practical application are verified by a series of tests of fault data sets.

Dissolved gas analysis (DGA) of the insulation oil of power transformers is an investigative tool to monitor their health and to detect impending failures by recognizing anomalous patterns of DGA concentrations. We handle the failure prediction problem as a simple data-mining task on DGA samples, optionally exploiting the transformer's age, nominal power and voltage, and consider two approaches: 1) binary classification and 2) regression of the time to failure. We propose a simple logarithmic transform to preprocess DGA data in order to deal with long-tail distributions of concentrations. We have reviewed and evaluated 15 standard statistical machine-learning algorithms on that task, and reported quantitative results on a small but published set of power transformers and on proprietary data from thousands of network transformers of a utility company. Our results confirm that nonlinear decision functions, such as neural networks, support vector machines with Gaussian kernels, or local linear regression can theoretically provide slightly better performance than linear classifiers or regressors. Software and part of the data are available at http://www.mirowski.info/pub/dga.

For composite fault is difficult to diagnose, the characteristics of the large amount of data. This paper presents a method of The Prediction method of Composite Fault Based on data driven to establish intelligence unit Based on a collection of virtual individuals associated with the virtual failure associated collection and virtual behavior associated collection. Composite fault warning engine modeling is proposed, and give the warning value of composite fault finally. This method is fully assessing the future "dominant state" on the basis of the fully aware of current "hidden state". The impact of factors such as disturbance of hidden failures on composite fault prediction are fully considered, to some extent, the long-span composite failure prediction problem is solved, and the experiments show that the method effectively increases the accuracy of composite fault prediction.

An efficient and effective ensemble of support vector machines for anti-diabetic drug failure prediction