• Energy Research

Noise in power-quality (PQ) signals has been the biggest hurdle in wavelet-based detection and time localization of PQ events. The well-known threshold-based denoising techniques, used in the signal-processing area, do not perform well with practical PQ waveform data. This paper proposes a simple yet effective denoising technique using inter and intrascale dependencies of wavelet coefficients to denoise PQ waveform data for enhanced detection and time localization of PQ disturbances. Utilizing the fact that the wavelet coefficients are not only correlated with its local neighborhood within the subband but also across the subband, the proposed method exploits the local structure of wavelet coefficients as well as high correlation of adjacent wavelet scales. The effectiveness of the proposed approach is tested and demonstrated with both simulated and measured power-line disturbance data, and the results show that the proposed scheme significantly outperforms existing methods used to denoise PQ data.

This paper proposes a new and simple approach based on orthogonal polynomial approximation (OPA) to detect, localize, and investigate the feasibility of classification of various types of power quality disturbances. The key idea in this approach is to approximate a given disturbance signal in the least square sense, such that the uncorrelated part (disturbance) of the signal is not present in the approximated version of the signal. It is, therefore, possible to detect and localize power quality (PQ) disturbances by analyzing the difference of the original and approximated signals. This is achieved by choosing the degree of the polynomial using the criterion of minimum error-variance. The effectiveness of the proposed approach is tested and demonstrated to detect and localize PQ disturbances with simulated and actual power line disturbance data.

Unwanted process shutdown is an important and challenging problem in petrochemical Industry. Incipient fault detection and diagnosis of a fault while the system is still operating not only avoids abnormal event progression, but also reduce productivity loss and hazards. In this paper, simulated version of Tennessee Eastman (TE) process, which is replica of real time process, is studied. There are 21 types of identified faults out of which two faults, `reactor cooling water valve sticking', and `condenser cooling water valve sticking', have been simulated using MATLAB. Dynamics of important process variables under faulty operation, obtained using simulated data, are presented and analyzed. A brief discussion on the need for a multivariate dimensionality reduction technique is also presented in this paper.

Recently, energy distribution of wavelet coefficient at different resolution level has been found to be very effective discriminatory feature for classification of power quality (PQ) disturbances. In practice, signals captured by monitoring devices are often corrupted by noise. The presence of noise will change the energy distribution pattern and may result in increased false classification rate. The robustness of the energy features, extracted for classification in the presence of noise and its effect on classification accuracy, which has been rarely discussed, need to be addressed. Recognizing such importance and necessity, this paper discusses the effect of noise on classification accuracy and proposes a low complexity robust denoising scheme in wavelet domain, to extract the required energy features for automatic classification of PQ disturbances. The "noise variance preserving property of Daubechies wavelet across the time-frequency scales, is used to estimate the noise energy at different resolution levels. The proposed approach is demonstrated for various PQ disturbances.

Abstract Newer generation sources and loads are posing new challenges to the conventional power system protection schemes. Adaptive and intelligent protection methodology, based on advanced measurement techniques and intelligent fault diagnosis such as machine learning (ML), is found to be useful to meet these challenges. A large number of research works are reported on ML-based power system fault diagnosis. However, ML techniques are evolving at a very fast pace, and an inclusive, as well as state-of-the-art review on ML-based power system fault diagnosis, is not available in the literature. Given this need and growing trend towards ML, the study presented in this paper aims to provide a comprehensive review of ML-based power system fault diagnosis. At first, efforts have been made to enlist the issues present in conventional fault diagnosis which led to the popularity of ML techniques. Also, a baseline framework and workflow for ML-based fault diagnosis are presented. Next, various unsupervised and supervised learning techniques have been discussed separately which have been used by several researchers for fault diagnosis. The discussion throughout is supported with tabulated facts for fault detection, classification and localization works with techniques used, different simulation tools used, and their application system. The advantages and disadvantages of all the techniques of fault diagnosis have also been discussed which will help the readers in the selection of techniques for their research. A brief review of reinforcement learning and transfer learning is also given as they are gaining popularity in power system-related studies and have the potential to be used for fault diagnosis. Finally, the research trends, some key issues, and directions for future research have been highlighted.

Abstract The excellent time-frequency localization property of the wavelet transform has made it a very promising tool for detection and analysis of the power quality disturbances. Many researchers have shown the adverse effect of noise on wavelet-based power quality monitoring and demonstrated that the performance of the wavelet transform in detecting the power quality disturbance would be greatly degraded due to the difficulty of distinguishing the noise and the disturbances. Practically, the power quality signals are often mixed with electromagnetic noise. This article proposes a denoising scheme of wavelet transform coefficients in noisy environment to avoid the false alarm rate and to increase the detection capability of wavelet transform-based power quality monitoring schemes. Contrary to the threshold-based techniques used so far in the power area for denoising power quality data, the technique used in this article exploits the local structure of wavelet coefficients. The effectiveness of the propo...

Selection of the most suitable drill bit type is an important task for drillers when planning for new oil and gas wells. With the advancement of intelligent predictive models, the automated selection of drill bit type is possible using earlier drilled offset wells’ data. However, real-field well data samples naturally involve an unequal distribution of data points that results in the formation of a complex imbalance multi-class classification problem during drill bit selection. In this analysis, Ensemble methods, namely Adaboost and Random Forest, have been combined with the data re-sampling techniques to provide a new approach for handling the complex drill bit selection process. Additionally, four popular machine learning techniques namely, K-nearest neighbors, naïve Bayes, multilayer perceptron, and support vector machine, are also evaluated to understand the performance degrading effects of imbalanced drilling data obtained from Norwegian wells. The comparison of results shows that the random forest with bootstrap class weighting technique has given the most impressive performance for bit type selection with testing accuracy ranges from 92% to 99%, and G-mean (0.84–0.97) in critical to normal experimental scenarios. This study provides an approach to automate the drill bit selection process over any field, which will minimize human error, time, and drilling cost.

A wavelet-transform (WT)-based power-quality (PQ) monitoring system captures voltage and current waveforms, when magnitudes of WT coefficients exceed the set threshold values across the scales. A lot of literatures has proposed several methods based on WT to detect and classify PQ disturbances. But a problem in the practical implementation of the wavelet-based triggering method is the presence of noise, riding on the signal. The presence of noise not only degrades the detection capability of wavelet-based PQ monitoring systems but also hinders the recovery of important information from the captured waveform for time localization and classification of the disturbances. Therefore, to enhance the performance of WT-based monitoring systems and to improve the classification accuracy of WT-based classifiers, two standard statistical hypothesis test-based denoising procedures have been proposed in this paper. Extensive tests conducted on the data obtained from simulations of a practical distribution system confirm the effectiveness of the proposed approaches in denoising of the PQ waveforms.

The oil and gas industry plays a vital role in meeting the ever-growing energy demand of the human race needed for its sustainable existence. Newer unconventional wells are drilled for the extraction of hydrocarbons that requires advanced innovations to encounter the challenges associated with the drilling operations. The type of drill bits utilized in any drilling operation has an economical influence on the overall drilling operation. The selection of suitable drill bits is a challenging task for driller while planning for new wells. Usually, when it comes to deciding the drill bit type, generally, the data of previously drilled wells present in similar geological formation are analyzed manually, making it subjective, erroneous, and time consuming. Therefore, the main objective of this study was to propose an automatic data-driven bit type selection method for drilling the target formation based on the Optimum Penetration Rate (ROP). Response Surface Methodology (RSM) and Artificial Bee Colony (ABC) have been utilized to develop a new data-driven modeling approach for the selection of optimum bit type. Data from three nearby Norwegian wells have been utilized for the testing of the proposed approach. RSM has been implemented to generate the objective function for ROP due to its strong data-fitting characteristic, while ABC has been utilized to locate the global optimal value of ROP. The proposed model has been generated with a 95% confidence level and compared with the existing model of Artificial Neural Network and Genetic Algorithm. The proposed approach can also be applied over any other geological field to automate the drill bit selection, which can minimize human error and drilling cost. The United Nations Development Programme also promotes innovations that are economical for industrial sectors and human sustainability.