Search

Flow instability is a class of abnormal operation in subsea oil wells. Applying machine learning models and improving detection and classification performance is decisive to reduce operational costs and downtime. In this paper we evaluate a pipeline of methods in order to increase correct classification rates. Our strategy is defined to avoid the similarity bias and approaches the binary problem in two distinct ways: A) using normal labels as negative and B) using both normal labels and all other kind of defects as negative, leveraging all data in the available dataset. The workflow includes feature extraction, hyperparameter tuning, feature selection with sequential algorithms, hybrid ranking wrapper and also with genetic algorithm. We show that hyperparameter tuning produces minor improvements and due to problem complexity a robust feature selection algorithm is required to deliver higher results.

In offshore petroleum exploration, subsea systems are susceptible to a variety of undesirable events or faults, in which oil wells operation is considered abnormal. Proper detection and classification of such events is crucial in order to reduce downtime, maintenance costs, and even damage to installations. Flow instability is a type of event inherently related to hydrocarbon multiphase flow and root cause of equipment stress and failure. This work investigates applying binary machine learning classifiers on real world captured data for the task of flow instability fault detection. Four different evaluation scenarios were considered. The mostly common scenarios used by the machine learning research community showed that even simple algorithms can reach high classification performance. The remaining scenarios, however, try to avoid the similarity bias problem and showed more realistic results.

A major challenge for remote patient monitoring (RPM) systems is processing data collected from a large number of healthcare devices and developing suitable algorithms and approaches to react accordingly to a wide spectrum of situations of interest, which must be properly detected. Flow-based programming, whose operation is based on state changes, is a promising approach to overcome this challenge, facilitating the personalization of data processing according to patients' health conditions. However, this approach is not suited to the detection of complex contextual situations, thereby hindering its adoption in RPM systems for data processing. In this regard, we propose an approach that combines situation awareness and flow-based programming to widen the capability of RPM systems to handle many-sided scenarios, which requires the monitoring of complex patient healthcare conditions. An evaluation of the proposed approach was conducted to demonstrate the flexibility of the solution for processing heterogeneous health information.

This study presents a virtual reality system for maintenance training of motors used in locomotives responsible for transportation of ore. Considering the cost and impact on the logistics chain of a mining company, preventive maintenance is required to reduce fault events and increase motor life. Technical staff training in a real locomotive has been difficult since the motor is long and heavy and hence not available for training. Thus, a virtual reality system offers a low-cost and efficient solution for maintenance training. This work consists in developing the 3D model for the virtual reality engine Unity 3D and proposing an XML module for motor part selection and the game design. We evaluated the simulator with 59 users, obtaining satisfactory results regarding comfort, immersion, usability, and user experience. Tests performed by technical staff validated the potential use in the training routine.

Clustering is a recurrent task in machine learning. The application of traditional heuristics techniques in large sets of data is not easy. They tend to have at least quadratic complexity with respect to the number of points, yielding prohibitive run times or low quality solutions. The most common approach to tackle this problem is to use weaker, more randomized algorithms with lower complexities to solve the clustering problem. This work proposes a novel approach for performing this task, allowing traditional, stronger algorithms to work on a sample of the data, chosen in such a way that the overall clustering is considered good. Preliminary experimental results indicate that the proposed approach is competitive to classical algorithms in large datasets with the advantage of automatically adapting to many different datasets.

Machine learning classification algorithms are highly dependent of a dataset composed of high-level features. In this paper, a deep learning approach is combined with traditional machine learning classifiers in order to circumvent the need of a specialist for extracting relevant features from one dimensional frequency-domain vibration signals. Our approach relies on a convolutional architecture trained with a triplet loss function for extracting relevant features directly from the raw data. A previously hand-crafted feature set, created by a specialist over the course of many years of research, is compared with the newly extracted feature set. Six conventional classifiers models (K-Nearest Neighbors, Support Vector Machine, Decision Tree, Random Forest, Quadratic Discriminant Analysis and Naive Bayes) are trained in both features set separately and compared in terms of macro F-measure. Results shows statistical evidence towards to the acceptance that the extracted feature set is as good as or better than the hand-crafted feature set, for classification purposes.

The usage of VR (Virtual Reality) has grown exponentially in the last decade. With various benefits, such as profitability, versatility and practicality, and applications that cover various fields of study, this new form of media has gained its ground in the market. The VR devices, however, have some special requirements for the 3D models it utilizes. The purpose of this paper is to present a workflow to create, with a reduced number of polygons, these 3D models in a way that enables their utilization in virtual reality systems, while also improving the performance of the game engine. There is also a comparison between the models of railway wagons maintenance centers developed with and without the workflow’s assistance by evaluating the number of polygons, appearance and performance in the game engine.

Electrical submersible pumps are devices frequently used in off-shore oil exploration. Vibration signals analysis and expert systems technology are used for detecting faults on these motor pumps. Fault diagnosis classifiers may need to be updated or expanded. This paper proposes a domain specific language for enabling non-programmer engineers to create and adjust rule-based fault diagnosis classifiers of electrical submersible pumps.

The main motivation to develop this work is to create a diagnosis system able to facilitate the work of human experts responsible for detecting faults before acquisition of submersible petroleum motor pump systems. A new approach for multiclass learning by reduction to multiple, binary classifiers, in a one-versus-one scheme, is presented as an alternative artificial intelligence solution to diagnose faults. Such an idea is based on the hypothesis that each pair of process conditions has different optimal feature sets to improve the classification performance. Thus, features are selected from datasets containing only two classes. Then, classifiers are trained with the selected features. The combination uses the average confidence of each classifier pair prediction to calculate the ensemble answer. Experimental results show that the proposed approach improves classification performance in a statistically significant way, when compared with correlated work. A secondary contribution is the analysis of the most difficult fault to be identified, namely rubbing.

This work investigates the use of soft computing techniques in a process of building a SVM classifier ensemble for a complex industrial machine fault diagnosis system. The process has three stages. The first stage uses a genetic algorithm to find a set of feature subsets for building an ensemble of acurated and diversified classifiers. The second stage uses a particle swarm optimization algorithm for tuning the SVM hyperparameters applied on each feature subset. The third stage uses the clonal selection algorithm or the simulated annealing algorithm for selecting the classifiers that compose the final ensemble. Experiments were performed on data extracted from vibration signals of oil rigs motor pumps. The results show that the proposed method is more efficient than others well-established methods applied to this problem.

This paper presents an extension of a comparative study of classifier architectures for automatic fault diagnosis, with a special emphasis on the Extreme Learning Machine (ELM), with and without kernel mapping. Besides the explanation of the ELM model, an attempt is made to find theoretical hints of the excellent generalization capabilities of this model, based on the findings of Cover about dichotomies and the equivalence of Mean Squared Error minimization in the high-dimensional feature spaces induced by kernels, and spaces defined by a finite sample set. The field of application is a practical problem in the context of offshore petroleum exploration where sophisticated submersible motor pumps are extensively tested before being deployed. The work juxtaposes the performance of ELM to an existing statistically sound comparison of state of the art classifier methods for a hand-crafted feature model tailored specially to the spectra of the vibrational signals of the pump. The results suggest the remarkably good generalization capability of ELM, exhibiting the highest scores for the chosen F-measure performance criterion.

Every year, energy consumption grows world widely. Therefore, power companies need to investigate models to better forecast and plan the energy use. One approach to address this problem is the estimation of energy consumption in the customer level. Energy consumption forecasting problem is a time series regression task. It consists of predicting the energy consumption for the next month given a finite history of a customer. Machine learning techniques have shown promising results in a variety of problems including time series and regression problems. Part of these promising results are attributed to deep neural networks. Although investigated in other domains, deep architectures have not been used to address the energy consumption prediction problem. In this work, we propose a system to predict monthly energy consumption using deep learning techniques. Three deep learning models were studied: Deep Fully Connected, Convolutional and Long Short-Term Memory Neural Networks. Due to the sensitivity of these models to the input range, normalization techniques were also investigated. The proposed system was validated with real data of almost a million customers (resulting in over 9 million samples). Results showed that our system can predict monthly energy consumption with an absolute error of 31.83 kWh and a relative error of 17.29%.

In this paper we present a sampling approach to run the k-means algorithm in large data sets. We propose a genetic algorithm to guide sampling based on evaluating the fitness of each individual of the population through the k-means clustering algorithm. Although we want a partition with the lowest SSE, our algorithm tries to find the sample with the highest SSE. After finding a good sample the remaining points of the entire data set are clustered using the nearest centroid and, after that, the SSE of the final solution is calculated. Our proposal is applied on a set of public domain data sets and the results are compared against two other methods: the k-means running in a uniform random sample of the data set, and the k-means in the complete data set. The results showed that our algorithm has a good trade off between quality and computational cost, especially for large data sets and higher number of clusters.

In this paper, an artificial intelligence solution to diagnose faults before acquisition of submersible petroleum motor pump systems is presented. Proper fault identification is time consuming and demands highly trained human experts. The diagnosis system is intended to facilitate the work of the human component of this important process by replicating the decision of highly trained experts through a classifier. To perform the automatic diagnosis, firstly intermediate features are extracted as the vibration spectra. Subsequently, high level features are extracted and fed into a classifier that outputs the final diagnose. To validate our proposal and to select the best classifier (among K-Nearest-Neighbour, Random Forest, Support Vector Machine and Decision Tree) for this problem, we performed a comparative study using real data acquired in tests accomplished before acquisition of submersible motor pumps. Our dataset comprises thousands of entries of accelerometer sensors (vertically distributed along the particular system components) data labelled by an human expert to one of the considered scenarios (normal pump, faulty sensor, faulty pump with rubbing, misalignment or unbalance). Results have showed that the evaluated classifiers have equivalent performance for the given problem, and that the standardization procedure can improve the performance of some classifiers. The performance of the classifiers is sufficient to facilitate the work performed by humans and consequently reduce the time spent in the pump fault diagnosis process.

A hardware and software system was devised for solving a problem related to energy losses in public lighting services. A computational intelligence framework based on heterogeneous feature models, feature extraction and selection is used for classifying lamp types and wattages. This paper shows how radiometric sensors and lamp image data are used for the classification task. It is also presented an experimental comparison on the public lighting problem domain between three feature selection algorithms.

In pattern recognition systems, ensemble techniques claim a potential performance improvement compared to single classifier approaches. Decision templates (DT) were proposed as a simple and effective method for combining continuous valued outputs of an ensemble of classifiers. In this paper, the concept of decision template single-label multi-class classifier combination is extended to the multi-label case. The different classifiers needed for a combination are obtained from the continuous re-estimation used in the Recursive Dependent Binary Relevance multi-label classifier. Each base classifier used in this work, delivers besides the class label, a continuous output for the class that can be used to assemble the DTs.

This work addresses the application of a population based evolutionary algorithm called shuffled complex evolution (SCE) in the core of multidimensional knapsack problem (MKP). The core of the MKP is a set of items which are hard to decide if they are or not selected in good solutions. This concept is used to reduce the original size of MKP instances. The performance of the SCE applied to the reduced MKP is verified through computational experiments using well-known instances from literature. The approach proved to be effective in finding near optimal solutions demanding a small amount of processing time.

As the first main contribution, this work proposes a feature selection algorithm to be used as base driver for comparisons in fast feature selection experiments. This heuristic algorithm tries to eliminate the redundant and irrelevant features of the datasets by creating a univariate ranking, in decreasing order with respect to their individual performance, followed by a sequential selection to establish the final set. Secondly, it presents examples where feature selection surpasses the predictive power of classifier ensembles based on feature selection. The proposed algorithm is compared to two ensemble methods, one fast feature selection algorithm, one pure ranking method and one classifier algorithm without feature selection, achieving a better performance in 17 of a total of 20 microarray gene datasets.

This work presents a novel hybrid approach for feature selection using a combination of ranking and wrapper methods. Its main goal is to select features quickly, without significant loss of classification performance. Experiments comparing this approach with Sequential Forward Feature (SFS) selection showed its viability using Support Vector Machine and K-Nearest Neighbor classifiers in specific scenarios. As a test bed, vibrational signals were employed which need a previous feature extraction stage to create a classification system. In two experiments, 74 and 130 features were extracted from these databases. The proposed approach performed at least ten times faster than SFS, with 0.32% loss of accuracy in the worst case, requiring 26% to 57.5% less features to achieve its highest accuracy.

The methodology of multi-label classification is experimentally evaluated in the context of a chemical process where the occurrence of multiple faults is a plausible scenario. As a benchmark, the Tennessee Eastman simulator is used. Modifications to the source code of this system were made in order to permit the simultaneous existence of different faulty machine states. In this work, the method of dependent binary relevance is compared to the binary relevance in terms of the subset accuracy performance criterion, since in a complexly coupled chemical process the dependence of certain fault classes should reveal.

The Extreme Learning Machine is an attractive artificial neural network architecture due to its low computational cost during the training process. In this work this classifier architecture is evaluated in the context of automatic fault diagnosis. As a benchmark, the data provided by the Tennessee Eastman simulator is used. The results are compared to the Support Vector Machine, K-Nearest Neighbor classifiers and methods based on feature extraction techniques, like e.g. Principal Component Analysis, Partial Least Squares, Independent Component Analysis. The test results suggest that the Extreme Learning Machine is an attractive alternative classification method of process conditions.

This work presents a performance analysis of the Extreme Learning Machine (ELM) compared to the Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN) classifiers for automatic diagnosis of machine conditions. Tests were performed using 5,314 real examples extracted from electrical submersible pumps. The vibration signal extraction was executed in laboratory and the samples were labeled by experts. Two feature extraction models were employed, statistical features from the time and frequency domains and amplitude peaks of harmonics and subharmonics of the shaft rotation frequency. Sequential feature selection was applied to improve classifier performance and to reduce dataset dimensionality. Experimental results suggest that the ELM may be used as a classification algorithm in automatic diagnosis systems. In certain scenarios, the ELM can outperform SVM regarding the quality of results and training speed.

This paper introduces a system based on light sensors that can, associated with a computational methodology, extract information about public lighting points. The objective is to give electricity companies exact information about the actual luminaries and bulbs installed on the lighting poles, thus minimizing commercial losses. A brief description of the equipment is presented, as well as the theoretical background on the technique applied to classify lamp types and wattages. The system performance is initially evaluated from the first experimental results, obtained from laboratory and field tests.

Brazilian energy distribution companies need to frequently update their database of the public lighting network. Since changes on this network are frequently not reported, it is important to monitor where they are made. An electronic device is under development for making this task easier. One component of this device uses image processing and recognition techniques to identify the luminaire models of the public lighting spot. This paper presents the comparative analysis of techniques for image description and classification we have performed in order to discriminate different models of luminaires.

This paper introduces a computational methodology developed to extract information about public lighting points. The objective is to give electricity companies exact information about the actual luminaries and bulbs installed on the lighting poles, thus minimizing commercial losses. The developed electronic system, that incorporates hardware and software elements, is discussed, as well as a theoretical background concerning spectrum signature detection and the computational intelligence employed to classify lamp classes and wattages. First experimental results obtained from the devised system are presented.

We discuss appropriate feature models for the automatic diagnosis of faults in two different application scenarios in a comparative study. The first test case is the Case Western Reserve University Bearing Data, the second is a submersible pump used in offshore oil exploration. Additionally we provide a visual comparison of the discriminative capabilities of the employed feature models using Principal Component Analysis and the Sammon plot to show the machine condition patterns.

We apply computational intelligence methods to the domain of fault diagnosis of rotating machinery, specifically submersible motor pumps used in offshore oil exploration. We propose distinct feature models to assemble a global feature pool from which the most discriminative information is filtered by feature selection. Statistically robust performance estimation for representative classifier models are used. The feature models are based on statistical parameters from the time and frequency domain and wavelet packet analysis. Feature selection is done by sequential techniques, with and without floating, applying wrapper and filter approaches. Performance estimation is based on the estimated accuracy and the area under the receiver operating characteristic curve (AUC-ROC). Experimental results are shown for 1834 vibration patterns, manually labeled by experts in the field of fault diagnosis. As representative classifiers we use the K-Nearest-Neighbor and Support Vector Machine.

We present a system for the detection and diagnosis of faults of a high performance electric submersible pump used in deep water oil exploration. During the installation phase 36 accelerometers acquire vibrational patterns under various load conditions. The machine condition is labeled with the help of human experts. The training set is submitted to an automatic model-free learning system based on Bayesian belief networks and compared to a reference Support Vector Machine classifier. Experiments are presented for three different condition classes, using sophisticated statistical evaluation methodologies to measure the classifier performance.

Non-technical energy losses mostly arise from illegal use of energy and force energy distribution companies to inspect large batches of clients in order to make decisions on actions for reducing these losses. Since an exhaustive inspection is impractical due to the high inspection cost and the very large number of clients, a carefully designed sampling procedure is needed. A useful strategy is offered by stratified sampling based on a division of the clients into homogeneous subgroups (strata). In this work we formulate the stratification task as a non-linear restricted optimization problem, in which the variance of overall energy loss due to the fraudulent activities is minimized. Solving this problem analytically is difficult and an exhaustive algorithm is intractable even for small problem instances. Therefore, we propose a Genetic Algorithm for finding practical solutions for the problem. Numerical experiments and a comparison with Simulated Annealing algorithm and a proportional allocation scheme are presented.

We perform an empirical performance analysis of the Multilayer Perceptron applied to the fault diagnosis of motor pumps installed on oil rigs. The conventional Multilayer Perceptron architecture is compared to a recently developed enhancement of this general purpose regression/classification paradigm, using an intermediate opaque layer which maps the original patterns to a reproducing kernel Hilbert space prior to learning the usual functional mapping of the network. State of the art statistical tools are used to corroborate our hypotheses that the kernel enhanced version improves the classification performance.

The objective of this work is the model-free diagnosis of faults of motor pumps installed on oil rigs by sophisticated kernel classifier ensembles. Signal processing of vibrational patterns delivers the features. Different kernel-based classifiers are combined in ensembles to optimize accuracy and increase robustness. A comparative study of various classification paradigms, all performing implicit nonlinear pattern mapping by kernels is done. We employ support vector machines, kernel nearest neighbor, Bayesian Quadratic Gaussian classifiers with kernels, and linear machines with kernels.

This paper presents the results achieved by fault classifier ensembles based on a model-free supervised learning approach for diagnosing faults on oil rigs motor pumps. The main goal is to compare two feature-based ensemble construction methods, and present a third variation from one of them. The use of ensembles instead of single classifier systems has been widely applied in classification problems lately. The diversification of classifiers performed by the methods presented in this work is obtained by varying the feature set each classifier uses, and also at one point, alternating the intrinsic parameters for the training algorithm. We show results obtained with the established genetic algorithm GEFS and our recently developed approach called BSFS, which has a lower computational cost. We rely on a database of real data, with 2000 acquisitions of vibration signals extracted from operational motor pumps. Our results compare the outcomes from the two methods mentioned, and present a modification in one of them that improved the accuracy, reinforcing the motivation for the usage of that method.

In this work, image segmentation is addressed as a crucial step within a specific motion analysis methodology. The issued methodology is used in biomechanics in order to obtain kinematics measurements of human and animal movement, namely, human and rat gait. The motion analysis system involves recording the movement, segmenting the images and, identification and tracking of the points of interest along the image sequences in order to estimate the kinematics parameters that will ultimately lead us to meaningful kinematics measurements. The quality of the segmentation revealed itself to be determinant regarding the accuracy of the biomechanics behavior characterization.

We present a collection of pattern recognition techniques applied to fault detection and diagnosis of motor pumps. Vibrational patterns are the basis for describing the condition of the process. We rely on the data-driven approach to the learning of the fault classes, i.e. supervised learning in pattern recognition. Our work is motivated by the diversity of the studied defects, the availability of real data from operational oil rigs, and the use of statistical pattern recognition techniques usually not explored sufficiently in similar works. We show the results of automatic methods to define, select and combine features that describe the process and to classify the faults on the provided examples. The support vector machine is chosen as the classification architecture.

We report about fault diagnosis experiments to improve the maintenance quality of motor pumps installed on oil rigs. We rely on the data-driven approach to the learning of the fault classes, i.e. supervised learning in pattern recognition. Features are extracted from the vibration signals to detect and diagnose misalignment and mechanical looseness problems. We show the results of automatic pattern recognition methods to define and select features that describe the faults of the provided examples. The support vector machine is chosen as the classification architecture.

The classification problem is recurrent in the con- text of supervised learning. A classification problem is a class of computational task in which labels must be assigned to object instances using information ac- quired from labeled instances of the same type of ob- jects. When these objects contain time sensitive data, special classification methods could be used to take ad- vantage of the inherent extra information. As far as this paper is concerned, the time sensitive data are se- quences of values that represent the measured energy consumption of residential clients in a given month. Traditional classifiers do not take temporal features into account, interpreting them as a series of unrelated static information. The proposed method is to develop methods of classification to be applied in a real time- series problem that somehow consider the time series as being the same value being repeatedly measured. Two new approaches are suggested to deal with this prob- lem: the first is a Hybrid classifier that uses clustering, DTW (Dynamic Time Warp) and Euclidean distance to label a given instance. The second is a Weighted Curve Comparison Algorithm that creates consumption profiles and compares them with the unknown instance to classify it.

This paper presents vibration analysis techniques for fault detection in rotating machines. Rolling-element bearing defects inside a motor pump are the object of study. A dynamic model of the faults usually found in this context is presented. Initially a graphic simulation is used to produce the signals. Signal processing techniques, like frequency filters, Hilbert transform and spectral analysis are then used to extract features that will later be used as a base to classify the states of the studied process. After that real data from a centrifugal pump is submitted to the developed methods.

The online HMMs (Hidden Markov Model) adaptation has been introduced by this work for the patient ECG signal adaptation problem. Two adaptive methods were implemented, namely the incremental version of the expectation- maximization (EM) and segmental k-means algorithms. The algorithms were implemented in an ECG segmentation system which classificatory is based on HMM. The performance criteria adopted are waveform detection, segmentation precision, and ischemia detection. For the tests, were used the QT and ST-T databases. The experiments have shown that the system adaptation for each individual improves the system reliability and increases the system performance. Furthermore, our results compare favorably with other works in the literature.

In this work, we propose an ECG analysis system to ischemia detection. This system is based on an original markovian approach for online beat detection and segmentation, providing a precise localization of all beat waves and particularly of the PQ and ST segments. Our approach addresses a large panel of topics never studied before in others HMM related works: multichannel beat detection and segmentation, waveform models and unsupervised patient adaptation. Thanks to the use of some heuristic rules defined by cardiologists, our system performs a reliable ischemic episode detection, showing to be a helpful tool to ambulatory ECG analysis. The performance was evaluated on the two-channel European ST-T database, following its ST episode definitions. The experimentation was performed over 48 files extracted from 90. Our best average statistic results are 83% sensitivity and 85% positive predictivity. Performance compares favorably to others reported in the literature.

In this article, we explore the use of a unique type of wavelets for ECG beat detection and classification. Once the different beats are segmented, classification is performed using at the input of a neural network different wavelet scales. This improves the noise resistance and allows a better representation of the different morphologies. The results, evaluated on the MIT/BIH database, are excellent (97.69% on the normal and PVC classes) thanks to the use of a regularization technique.