Switching Detection/Classification Using Discrete Wavelet Transform and Self-Organizing Mapping Network

Electric railway power systems (ERPS) as one of the most critical and high-power end-user loads of utility grids are characterized by outlandish power quality (PQ) problems all over the world. The extension and evolution of different supply topologies for these systems has resulted in significant and various forms of distortions in network voltage and current in all ERPS, the connected power system, and adjacent consumers. During the last years, numerous studies have been offered to investigate various aspects of PQs in a specific supplying topology. Variation in the supply structure of the ERPS and different types of locomotives has propelled the observation of different PQ phenomena. This versatility and development have led to confront considerable types of two-way interactive interfaces as well as reliability and PQ problems in ERPS. In addition, the lack of standards explicitly dedicated to ERPS has added to the ambiguity and complexity of this issue. In this paper, an extensive review of PQ distortions and phenomena in different configurations of ERPS is proposed and a systematic classification is presented. More than 140 scientific papers and publications are studied and categorized which can provide a fast review and a perfect perspective on the status of PQ indexes for researchers and experts.

Power Quality (PQ) monitoring in a systematic and automated way is the important issue to prevent detrimental effects on power system. The development of new methods for the automatic classification of PQ disturbances is at present a major concern. This paper presents a novel approach of automatic detection and classification of single and hybrid PQ disturbances using Discrete Wavelet Transform (DWT) and Modular Probabilistic Neural Network (MPNN). The automatic classification of the PQ disturbances consists of three stages i) data generation, ii) feature extraction and iii) disturbance classification. The data is generated by synthetic models of single and hybrid PQ disturbance signals based on IEEE 1159 standard. DWT with multiresolution analysis was applied for feature extraction from the PQ waveforms. The entropy and energy features extracted from the detail and approximation coefficients were applied as the training and testing data to MPNN in order to accomplish the automatic classification process. The effectiveness of the proposed algorithm has been validated by using a typical real-time underground distribution network in Malaysia which was simulated in PSCAD/EMTDC power system software to generate PQ disturbances. The simulation results show that the classifier has an excellent performance in terms of accuracy and reliability even in the case of PQ signals under noisy condition.

This paper introduces a novel automatic hybrid classifier to detect and classify the Power Quality(PQ) problems in power systems using Wavelet Packet Transform (WPT) and Artificial Neural Networks (ANN). Various PQ events like Normal, Sag, Swell and Interruptions are obtained by modeling three phase distribution system using MATLAB simulink. For classification, the selection of suitable features from the disturbance signal is extremely important. Recent literature survey deals with various signal processing techniques like Fast Fourier Transform (FFT), Discrete Wavelet Transform (DWT) and Wavelet Packet Transform (WPT) were used for feature extraction. From the above techniques, the main issues addressed are selection of optimal feature subset and the design of ANN architecture model. To build a competent and robust classifier, it is essential to extract the utilizable feature vectors from the disturbance signal that can optimize data size as well as incorporate the main characteristics of the signal. This paper addresses these issues, and the distinctive feature vectors are obtained with reduced number of coefficients by using Energy Entropy based WPT. From these, the best discriminative feature vectors with reduced number of coefficients are obtained and are used as input to ANN, so that the burden over classification can be reduced. The classification performance is compared with FFT based ANN. The simulation results obtained have significant improvement over existing methods. Thus the proposed hybrid classifier provides the most excellent detection among PQ problems that arises in real time by improving classification accuracy in terms of both computation time and means square error.

Abstract- Based on Fast Fourier Transform (FFT), power quality indices (PQIs) definitions contained in the standards cannot be successfully applied for non-stationary signals analysis in unbalanced three-phase systems with non-sinusoidal conditions. Unlike FFT, Discrete Wavelet Transform (DWT) gives more accurate definitions preserving time-frequency information and reduces the computational effort. This paper proposes a comparative analysis of definitions employing these two instruments for unbalanced three-phase systems case in sinusoidal and nonsinusoidal situations using waveforms captured with data acquisition system which contains Hall voltage and current transducers. The experimental testing of harmonic distortion was made for a RL unbalanced three-phase load, a three-phase rectifier with diodes (TPRD) and a rectifier with near sinusoidal currents (RNSIC). I. I NTRODUCTION For the electric power quality monitoring, the IEEE working group has analysed the customers expectations concerning electric power delivery, the susceptibility limits of the electric and electronic equipments connected at the electric network and thus obtained the standards regarding the evaluation methods of electric power quality, the parameters that define this concept and the functioning classes of the measurement tools [1-3]. In the last years electric power quality became un important issue in the power systems domain because of the increased usage of nonlinear loads which are producing great harmonic distortion in power supply [4-12, 33-36]. Voltage and current quality can be determined based on the deviations between the real voltage/current and the ideal voltage/current waveforms [13]. These deviations are counted power quality disturbances and can be classified as stationary (with time unchangeable characteristics) or non-stationary (with time changeable characteristics) and each of them as sinusoidal or non-sinusoidal. The Fast Fourier Transform (FFT) approach in Power Quality Indices (PQIs) definition is replaced in this study with a time-frequency domain approach using Discrete Wavelet Transform (DWT). This method eliminates the FFT drawbacks as high computational effort, time related information loss, spectral leakage and provides representations for any non-stationary waveforms [14-17]. DWT is a useful tool in the waveforms analysis due to its accuracy and sensitivity at signal irregularities [5], [32]. The waveforms of voltage and current are obtained using a data acquisition system with linear/nonlinear loads composed of a Digital Signal Processor (DSP) board with high-speed 12 bit Analog to Digital Converters (ADC) coupled to a three-phase system using the Hall transducers. This paper presents a comparative analysis of DWT based measurements of power quality indices and the IEEE Standard 1459-2000 based values using unbalanced linear three-phase systems and nonlinear power loads. The power quality disturbances are calculated with a RL unbalanced three-phase load, a three-phase rectifier with diodes (TPRD) and a rectifier with near sinusoidal currents (RNSIC). II. T

Built-in event diagnosis and assessment modules are the key to the power quality (PQ) monitoring systems. Discrete wavelet transform (DWT) has been used recently by many researchers for detection and analysis of the of PQ disturbances. Despite the excellent time-frequency localization property, DWT based event detectors can not adapt itself to the dynamic changes in the load configuration because of its fixed filter coefficients and, hence, generates repeated detection and processing of the same steady state disturbance. An adaptive filter on the other hand adapts its filter coefficients continuously with changing monitored signal. Therefore, an adaptive filtering and DWT based hybrid system has been proposed in this paper for the detection and analysis of transient disturbances. Adaptive filter based monitoring avoids the detection of the same study event repeatedly. Extensive tests conducted on the data obtained from a practical distribution system confirm the effectiveness of the proposed approach in automatic detection and diagnosis of transient PQ violations.

Extending the Kohonen self-organizing map networks for clustering analysis

The worldwide increasing applications of nonlinear loads, mostly consisting of power electronics devices, have made the power quality problems an important concern than ever before for utilities and consumers. Therefore, detection and classification of Power Quality disturbances is highly desirable. Most power quality disturbances are non-stationary and transitory and their detection and classification have proved to be much needed. Hence there is a requirement of advanced tools and techniques for the analysis of Power Quality disturbances. In this work detection and classification of various types of transients and interruptions, caused due to severe fault and reclosing of circuit breaker, is done by using Discrete Wavelet Transform and two-layer feed forward neural network. This work presents new approach aimed at automating the analysis of power quality disturbances including transients, interruption and normal waveform. The disturbance current waveform can be obtained from the disturbance generation model. The Discrete Wavelet Transform is chosen for feature extraction. Feature extraction outputs are the coefficients (detailed and approximate) of Discrete Wavelet Transform represents the power quality disturbance signal at the different levels in time and frequency domain.

The transient caused by capacitor switching is one of the important power quality (PQ) problems. In particular, switching on the capacitor may result in an overvoltage to a sensitive load, serious disturbance to an adjustable-speed drive or a resonance in the system. This paper presents a new method to locate the positions of the transient sources with the help of the PQ monitoring system. The discrete wavelet transform is used to extract the features of transients caused by capacitor switching. The fuzzy-c-means is then used to determine the placement of PQ measurement facilities. The signal energies attained by wavelet coefficients serve as inputs to the hybrid principal component analysis neural network for locating the transient sources. The simulation results obtained from an 18-bus distribution system show the applicability of the proposed method

In this paper, it expounded in detail the principle of energy spectrum analysis based on Discrete Wavelet Transform and Multi-resolution Analysis. In the aspect of study on feature extraction method, with investigating the feature of impact factor in vibration signals and considering the non-placidity and nonlinear of vibration diagnosis signals, this paper imported wavelet analysis and fractal theory as the tools of faulty signal feature description. Experimental results proved the validity of this method. To some extent, this method provides a good approach of solving the problem that fault feature symptom is described comprehensively.

Power quality (PQ) problems include harmonics, voltage flicker, three-phase imbalance, voltage swell/sag, interruption, switching transient and so on. Identification of accurate location and occurrence time of PQ problem is important for understanding the underlying cause and to prevent future occurrence. This study presents a new method for placement of power quality measurement facilities to identify PQ problems. The power harmonics and capacitor switching transients are used as disturbance sources for illustration. The proposed method first utilises harmonic voltages and wavelet coefficients as PQ features. An enhanced genetic algorithm is proposed for solving the modified K-means clustering to place PQ measurement monitors. Simulation results for an 18-bus system show that the proposed approach is effective in comparison with the existing approaches.

This paper proposes a clustering ensemble method that introduces cascade structure into the self-organizing map (SOM) to solve the problem of the poor performance of a single clusterer. Cascaded SOM is an extension of classical SOM combined with the cascaded structure. The method combines the outputs of multiple SOM networks in a cascaded manner using them as an input to another SOM network. It also utilizes the characteristic of high-dimensional data insensitivity to changes in the values of a small number of dimensions to achieve the effect of ignoring part of the SOM network error output. Since the initial parameters of the SOM network and the sample training order are randomly generated, the model does not need to provide different training samples for each SOM network to generate a differentiated SOM clusterer. After testing on several classical datasets, the experimental results show that the model can effectively improve the accuracy of pattern recognition by 4%∼10%.

Chapter 38 - Power quality

In this paper a novel and simple approach for detection and classification of wide variety range of power quality(PQ) events based on discrete wavelet transform (DWT) and correlation coefficient is presented. For this purpose, two new indices is proposed and by comparing the values of the correlation coefficient between the value of these indices for pre-stored PQ events and for a recorded indistinct signal, type of PQ events will be detected. This algorithm has advantages of DWT and correlation coefficient which, it does not have disadvantage of neural network or neural network-fuzzy based algorithms such as; training, and high dimension input matrices nor it does not have disadvantage of Fourier transform based approach such as unsuitability for non-stationary signal as it does not track signal dynamics properly due to limitation of fixed window width. The effectiveness of this method has been tested using numerous PQ disturbance and simulation results confirm the competency and the ability of the proposed method in PQ disturbances detection and automatic diagnosis. Compared with the other methods, the simulation under different noises conditions, verify the effectiveness of noise immunity, and relatively better accuracy of the proposed method.

For an effective communication system whether indoor or outdoor, the most important concern is minimum noise. In this paper, an efficient noise reduction technique is presented using various wavelet transform techniques for indoor optical wireless communication system (IOWC). In IOWC system, Fluorescent Light Interference (FLI) is main source of noise. Here, in this paper three methods are used to reduce the effect of noise from a digital signal. These are Discrete Wavelet Transform (DWT), Stationary Wavelet transform (SWT) and Discrete Wavelet transform-Stationary Wavelet Transform (DWT-SWT). Through sub band coding in DWT the signal is decomposed into lower sub bands of high and low frequency respectively of unequal size; while in SWT the decomposed signal have sub bands of equal size. In DWT-SWT the high frequency components of both DWT and SWT are added. Using Pulse Position Modulation, the comparison between these three techniques is described here to enhance the overall performance of the IOWC system.

Power quality monitors are typically deployed for power quality standard compliance survey power quality troubleshooting triggered by customer complaint. However, the rich data collected by power quality monitors can provide additional value to asset owners and system operators for purposes further than power quality itself. This paper presents utility experiences in asset condition assessment and operation anomaly detection by using power quality data analytics. Transformer ferroresonance, renewable generator instability and breaker restrike are selected to exemplify the usefulness of power quality data in applications beyond passive power quality monitoring.