Development of a fault detection algorithm for an alternate aerobic/anoxic cycle nitrogen removal process

In data-based method for fault detection, principal component analysis (PCA) has been used successfully for fault detection in system with highly correlated variables. The aim of this paper is to combine the exponentially weighted moving average (EWMA) control scheme with PCA model in order to improve fault detection performance. In fact, PCA is used to provide a modeling framework for the develop fault detection algorithm. Because of the ability of EWMA control scheme for detecting small changes, this technique is appropriate to improve the detection of a small fault in PCA model. The performance of the PCA-based EWMA fault detection algorithm is illustrated and compared to conventional fault detection methods using simulated continuously stirred tank reactor (CSTR) data. The results show the effectiveness of the developed algorithm.

This paper deals with the problem of fault detection and diagnosis in sensors considering erratic, drift, hard-over, spike, and stuck faults. The data set containing samples of the above mentioned fault signals was acquired as follows: normal data signals were obtained from a temperature-to-voltage converter by using an Arduino Uno microcontroller board and MATLAB. Then, faults were simulated in normal data to get 100 samples of each fault, in which one sample is composed of 1000 data elements. A support vector machine (SVM) was used for data classification in a one-versus-rest manner. The statistical time-domain features, extracted from a sample, were used as a single observation for training and testing SVM. The number of features varied from 5 to 10 to examine the effect on accuracy of SVM. Three different kernel functions used to train SVM include linear, polynomial, and radial-basis function kernels. The fault occurrence event in fault samples was chosen randomly in some cases to replicate a practical scenario in industrial systems. The results show that an increase in the number of features from 5 to 10 hardly increases the total accuracy of the classifier. However, using ten features gives the highest accuracy for fault classification in an SVM. An increase in the number of training samples from 40 to 60 caused an overfitting problem. The k-fold cross-validation technique was adopted to overcome this issue. The increase in number of data elements per sample to 2500 increases the efficiency of the classifier. However, an increase in the number of training samples to 400 reduces the capability of SVM to classify stuck fault. The receiver operating characteristics curve comparison shows the efficiency of SVM over a neural network.

Fault detection (FD) algorithms based on supervisory control and data acquisition (SCADA) data have been widely used in the operation and maintenance (O&M) of wind turbines (WTs). However, the performance of FD models will degrade due to the time-varying operating conditions (TVOC). Incremental learning (IL) methods can be employed to update the models online to adapt to TVOC, but the error accumulation caused by false negatives will lead to a continuous decrease in the FD rate (FDR). A novel IL method considering false negatives is proposed to improve the performance of WT FD models. First, a data buffer is built to cache some new normal data used in model updating. Second, a processing strategy for false negatives is proposed to block some high-risk data from being added to the buffer, thereby weakening the error accumulation. Third, when the buffer is full, update the FD model using the data in it. A real-world SCADA dataset with gearbox faults and four different FD algorithms with various model updating strategies are used in the experiments. The results demonstrate that the proposed method can lower the false alarm rate (FAR) of all four FD algorithms. After processing false negatives, for example, for the FD model based on Gaussian kernel regression, its FDR increased by about 11%, and for the FD model based on multivariate state estimation technique, its FDR increased by about 14%. The results of hyperparameter experiments and early stopping experiments show that the proposed method has good potential for practical applications.

FAST ONLINE TRAINING OF L1 SUPPORT VECTOR MACHINES By Gabriella Melki A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at Virginia Commonwealth University. Virginia Commonwealth University, 2016. Directors: Vojislav Kecman, Alberto Cano Department of Computer Science This thesis proposes a novel experimental environment (non-linear stochastic gradient descent, NL-SGD), as well as a novel online learning algorithm (OL SVM), for solving a classic nonlinear Soft Margin L1 Support Vector Machine (SVM) problem using a Stochastic Gradient Descent (SGD) algorithm. The NL-SGD implementation has a unique method of random sampling and alpha calculations. The developed code produces a competitive accuracy and speed in comparison with the solutions of the Direct L2 SVM obtained by software for Minimal Norm SVM (MN-SVM) and Non-Negative Iterative Single Data Algorithm (NN-ISDA). The latter two algorithms have shown excellent performances on large datasets; which is why we chose to compare NL-SGD and OL SVM to them. All experiments have been done under strict double (nested) cross-validation, and the results are reported in terms of accuracy and CPU times. OL SVM has been implemented within MATLAB and is compared to the classic Sequential Minimal Optimization (SMO) algorithm implemented within MATLAB’s solver, fitcsvm. The experiments with OL SVM have been done using k-fold cross-validation and the results reported in % error and % speedup of CPU Time. viii

In this paper, the problem of accelerometer fault detection is studied for the rotary steerable drilling tool system (RSDTS) which is modeled by a nonlinear stochastic system. Owing to the strong vibration in the drilling process, RSDTSs are inevitably influenced by strong measurement noise, which brings in substantial difficulties in the fault detection. Firstly, a modified particle filtering algorithm is proposed to improve the accuracy of state estimation. Next, by means of the moving average method, the residual which approximatively obeys the χ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> distribution is designed when the estimation error is small enough. According to the property of the χ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> distribution, the window length of the moving average method is determined. Moreover, a fault detection algorithm is presented, and false alarm rate and missed detection rate are quantitatively analyzed for nonlinear stochastic systems. Finally, the effectiveness of the developed fault detection method is verified by simulation and experiment on the prototype of the RSDTS.

There is a growing body of knowledge indicating long-lasting residual electroencephalography (EEG) abnormalities in concussed athletes that may persist up to 10-year postinjury. Most often, these abnormalities are initially overlooked using traditional concussion assessment tools. Accordingly, premature return to sport participation may lead to recurrent episodes of concussion, increasing the risk of recurrent concussions with more severe consequences. Sixty-one athletes at high risk for concussion (i.e., collegiate rugby and football players) were recruited and underwent EEG baseline assessment. Thirty of these athletes suffered from concussion and were retested at day 30 postinjury. A number of task-related EEG recordings were conducted. A novel classification algorithm, the support vector machine (SVM), was applied as a classifier to identify residual functional abnormalities in athletes suffering from concussion using a multichannel EEG data set. The total accuracy of the classifier using the 10 features was 77.1%. The classifier has a high sensitivity of 96.7% (linear SVM), 80.0% (nonlinear SVM), and a relatively lower but acceptable selectivity of 69.1% (linear SVM) and 75.0% (nonlinear SVM). The major findings of this report are as follows: 1) discriminative features were observed at theta, alpha, and beta frequency bands, 2) the minimal redundancy relevance method was identified as being superior to the univariate t -test method in selecting features for the model calculation, 3) the EEG features selected for the classification model are linked to temporal and occipital areas, and 4) postural parameters influence EEG data set and can be used as discriminative features for the classification model. Overall, this report provides sufficient evidence that 10 EEG features selected for final analysis and SVM may be potentially used in clinical practice for automatic classification of athletes with residual brain functional abnormalities following a concussion episode.

To ensure high reliability of the Grid-Connected Photovoltaic (GCPV) systems, promptly faults detection, diagnosis and automatic process monitoring are essential tools to keep the PV and the grid network under optimal functioning. Regardless of fault types, incipient faults are usually more difficult to detect and accurately isolate. As an alternative and effective method, the principal components analysis (PCA) is proposed to extract and select more relevant features and support vector machines (SVM) technique is applied to quickly detect the faults that occur in a GCPV system. The T2 and squared weighted errors (SWE) statistics, generally used as fault detection indices, are appropriately extracted and selected within the PCA framework. Both of these features are fed to a SVM classifier handling the incipient fault detection. This task is carried out on a simulated GCPV operating under maximum power point trackers (MPPT) and matching realistic outdoor to demonstrate the effectiveness and robustness of the proposed technique.

Sensor nodes are tiny low-cost devices prone to various faults. So, it is imperative to detect those faults. This paper presents a sensor measurement fault detection algorithm based on Pearson’s correlation coefficient and the Support Vector Machine(SVM) algorithm. As environmental phenomena are spatially and temporally correlated but faults are somewhat uncorrelated, Pearson’s correlation coefficient is used to measure correlation. Then SVM was used to classify faulty readings from normal readings. After classification, faulty readings are discarded. Here each sensor nodes periodically collects environmental features and sends them to their associated cluster heads. Each cluster head analyze collected data using the classification algorithm to detect whether any fault is present or not. Network simulator NS-2.35 and Matlab are used for evaluation of our proposed method. The fault detection algorithm was evaluated using performance metrics, namely, Accuracy, Precision, Sensitivity, Specificity, Recall, \(F_1\) Score, Geometric Mean(G_mean), Receiver Operating Characteristics (ROC), and Area Under Curve(AUC). Performance evaluation shows, the proposed method performs well for high fault percentages.

Cancer survival classification using integrated data sets and intermediate information.

Large-scale image retrieval using transductive support vector machines

Applications of non-linear kernel support vector machines (SVMs) to large data sets is seriously hampered by its excessive training time. We propose a modification, called the approximate extreme points support vector machine (AESVM), that is aimed at overcoming this burden. Our approach relies on conducting the SVM optimization over a carefully selected subset, called the representative set, of the training data set. We present analytical results that indicate the similarity of AESVM and SVM solutions. A linear time algorithm based on convex hulls and extreme points is used to compute the representative set in kernel space. Extensive computational experiments on nine data sets compared AESVM to LIBSVM (Chang and Lin, 2011), CVM (Tsang et al., 2005), BVM (Tsang et al., 2007), LASVM (Bordes et al., 2005), SVMperf (Joachims and Yu, 2009), and the random features method (Rahimi and Recht, 2007). Our AESVM implementation was found to train much faster than the other methods, while its classification accuracy was similar to that of LIBSVM in all cases. In particular, for a seizure detection data set, AESVM training was almost 500 times faster than LIBSVM and LASVM and 20 times faster than CVM and BVM. Additionally, AESVM also gave competitively fast classification times.

Industrial big data has created a challenge for data measurement, detection, and processing. This paper shows that support vector machine (SVM) is extremely useful in detecting fault information in modern complex industrial processes. With a pilot plant of Continuous Stirred Tank Heater (CSTH) process, the SVM method with radial basis function (RBF) kernels is tested on the CSTH database and compared with an improved Partial Least Squares (IPLS) and Principal Component Analysis (PCA). The performance of SVM is validated using k-fold cross-validation where the classifier based on SVM outperforms those based on PCA and IPLS. These comparisons show that SVM has remarkable detection performance and satisfying elapsed time. From an industrial point of view, the vitality of SVM algorithm in actual industrial process is discussed.

peer reviewed

Fault detection and diagnosis has an effective role for the safe operation and long life of systems. Condition monitoring is an appropriate way of the maintenance techniques which is applicable in the fault diagnosis of rotating machinery faults. We considered the Support Vector Machine (SVM) method for classifying the condition of centrifugal pump into two types of faults through six features: flow, temperature, suction pressure, discharge pressure, velocity, and vibration. The SVM method is based on statistical learning theory (SLT) and powerful for the problem with small sampling, nonlinear and high dimension. (L.V. Ganyun et al 2005). The SVM classifying is implemented with 4 kernel functions and the results of them are compared. We use an Artificial Neural Network (ANN) as the second classifying method to have comparison among the performance of two methods. After applying the two methods to our data set we make the data set noisy and again we try our SVMs and ANN to compare their robustness in noisy conditions and the results obtained from two methods confirmed the superiority of SVM with some specific kernel functions.

Research on turbopump fault detection is significance in engine health monitoring. Support Vector Machine (SVM) is a novel machine learning method, and we can use it in turbopump fault detection to solve the problems such as small sample and nonlinear problems. In this paper, we established a kind of adaptive two-class C-Support Vector Machine (C-SVM) algorithm for fault detection based on original C-SVM algorithm, and described the main parts of the algorithm such as fault feature extraction, kernel function choosing and classifier real-time updating in detail. At last we compared the algorithm with original C-SVM algorithm by the tests of false alarm, missing alarm and computing speed. The test results showed that the adaptive two-class C-SVM algorithm has lower false alarm rate, lower missing alarm rate and lower computing speed.