Voltage Sag Data Research Articles

Prediction models of voltage sag characteristics based on measured data

An effective method of assisting utilities and users to avoid the risk of voltage sags is the accurate prediction of voltage sags. This study analyzes the predictability and proposes the prediction models for three characteristics of voltage sags based on measured data. First, this study defines the time series of voltage sag and analyzes the predictability of the residual voltage, duration, and time of occurrence. Second, this study proposes a fuzzy logic prediction model of the residual voltage to address the fuzziness of the residual voltage. Third, this study proposes an evidence theory prediction model of the duration according to the correlation between the residual voltage and duration. Further, this study proposes a chaotic signal prediction model of the occurrence time by reconstructing the phase space of the occurrence time. Finally, the measured voltage sag data from 10 sites are used to analyze and validate the proposed prediction models.

Real-time Monitoring Technology of Voltage Sag Disturbance in Distribution Network Based on TCN-Attention Neural Network and Flink Flow Computing

In the face of the challenges brought by the complexity of power grid, diversification of disturbance factors, isolation of monitoring points and other issues to the cause identification of voltage sag disturbance, this paper proposes a real-time monitoring technology for voltage sag disturbance in distribution network based on TCN-Attention neural network and Flink flow calculation, which has important practical significance for controlling voltage sag and reducing economic losses. This method uses Temporal Convolutional Network (TCN) to extract the cross time nonlinear characteristics of voltage sag time series data, which effectively solves the problems of long-term dependence on time series and low training output efficiency of existing time series models. In order to further improve the recognition accuracy of the model, Attention mechanism is introduced to mine the duration relationship in voltage sag data. At the same time, the method also constructs a parallel real-time monitoring platform based on Flink streaming computing framework, embeds the TCN-Attention voltage sag cause identification model generated by training, so as to realize real-time identification and monitoring analysis of voltage sag disturbances at each monitoring point of the distribution network. In this paper, various voltage sags are simulated on IEEE 14 bus system using PSCAD software, and the proposed method is verified and tested. The deep learning fusion model has high recognition accuracy for the cause of voltage sag, and the flow computing platform has excellent performance in time delay and throughput indicators, and can realize the parallel real-time monitoring and analysis of voltage sag causes in distribution network.

Voltage Sag Recognition Based on CM-eCNN Model For Few-Shot Learning

Voltage sag is a serious problem in power systems. To reduce voltage sag, it is necessary to identify and classify voltage sag sources. However, due to insufficient voltage sag data, our paper designs a CM-eCNN model through few-shot learning. Firstly, this paper introduces the CM-eCNN model and studies how to improve the extraction efficiency of the constellation model to feature map through the attention mechanism. Then, it introduces the training method of few-shot learning. Finally, our model training with few-shot learning has better performance than other models, which proves effectiveness of the model.

A Novel Approach of Voltage Sag Data Analysis Stochastically: Study, Representation, and Detection of Region of Vulnerability

A voltage sag is a major power quality problem at any load location, mainly due to short circuit faults. Its effects are especially clear in the industrial sector because they lead to direct financial losses. The first and most important step to getting rid of or at least reducing voltage sag is to find the places in the network where it can happen. So, this article discusses a creative way to find the weak spots in a network where the voltage is more likely to drop. This innovative strategy is based on correlation and a stochastic normal probability distribution. The voltage sag is evaluated using an analytical method. The method simulates the number, type, and location of faults randomly throughout the system. After gathering voltage dip data, it is analyzed using normal probability distribution and correlation concepts. The correlation concept provides the relationship between the sag frequency and other parameters, and the sag occurrence area was indicated by the normal probability distribution. After that, the region of vulnerability (ROV) is developed using the ROV flow chart. This paper uses an IEEE 30-bus RTS system as a case study to demonstrate the usefulness of the suggested strategy using MATLAB software.

Rancang Bangun Alat Deteksi Sag Tegangan Menggunakan Sistem Data Logger

-This research discusses the design and manufacture of a voltage sag detection device using a data logger system. Voltage sag detection device to detect the presence of voltage sag which is one of the power quality problems in industrial electrical systems. Voltage sag can cause disturbances in electronic components that are sensitive to changes in voltage values. In this device, the voltage sag is detected using a voltage sensor by reading the RMS voltage from the PLN 220V AC source, then it is converted into a digital signal. The digital signal is sent to the ESP32 module to be processed and outputs a voltage value which is then displayed on the LCD Display. The voltage value data collection is carried out every <10ms and the data taken is the voltage value <90% of the rms voltage. The data taken is then saved to a micro SD card and displayed on the LCD Display. Several tests were carried out to ensure the performance of the voltage sag detector as designed. By switching the 110 / 220VAC transformer connected to the voltage sensor for 10 seconds, 12 seconds and 17 seconds. Based on these tests, the voltage sag detector can read and store the voltage sag data when the voltage read is below 90% of the rms voltage.

Load Loss Rate Evaluation Method of Sensitive Users Under Voltage Sag Based on Monitoring Data

Voltage sag can result in extensive losses of sensitive users. However, such losses have yet to be quantified based on power quality monitoring data. The load loss rate (LLR) is a practical index for evaluating the consequences of voltage sag, providing vital information for mitigation decisions. This study proposes a data-driven LLR evaluation method of sensitive users under voltage sag, which can be used as a basis for configuring mitigation devices. First, the LLR calculation index is proposed, based on the sensitive load response characteristics and the active power trajectory under voltage sags. Second, a feature indices system is established for the active power trajectory, based on the basic attributes and load response characteristics of sensitive load, to characterize the change of active power under voltage sag. Furthermore, a voltage sag data augmentation model is constructed to generate sample data, considering the characteristic calculation model of voltage sag and the probability density distribution of characteristics. Finally, LLR evaluation model is proposed, based on random forest regression for sensitive users under voltage sags. The effectiveness and accuracy of the proposed method are verified through a case study.

Clustering Analysis of Voltage Sag Events Based on Waveform Matching

Voltage sags are a serious problem within power supplies, which pose threats to both residential electricity and industrial manufacturing. Since any one sag may be recorded by multiple monitoring devices from different substations, the issue of redundant information in data arises. In this regard, a novel method for voltage sag events based on projection technology, shape dynamic time warping (shapeDTW), and spectral clustering is proposed. The main contributions of this paper may be summarized as follows: (1) We present a new method for extracting the voltage anomaly waveform, which is a fast projection segmentation algorithm (FPSA). The voltage sag waveform is only a part of the voltage anomaly waveform, so the voltage anomaly waveform contains more information. (2) ShapeDTW and spectral clustering are used to match and cluster voltage anomaly waveforms, so as to achieve the normalization of voltage sag events. (3) In practical engineering, the proposed method in the paper can be used to obtain the impact of voltage sags, reduce computational complexity, and ease the workload of the operation and maintenance engineers. Experiments were conducted using voltage sag data from voltage sag events recorded by the 10 kV monitoring points in Beijing, China. The results showed the effectiveness and reliability of our proposed methods.

Classification Method of Voltage Sag Sources Based on Sequential Trajectory Feature Learning Algorithm

The classification of voltage sag sources is essential for the establishment of controlling scheme and reasonable division of responsibilities in voltage sag-associated accidents. Existing methods for classifying voltage sag sources usually ignore the interpretability of the classification model, and are only dedicated to improving the accuracy of voltage classification, which cannot provide a reliable classification basis for users and power enterprises. Therefore, this article proposes an effective and interpretable voltage sag sources classification method based on sequential trajectory feature learning and Random Forest algorithm. Firstly, to fully consider the interpretable sequential trajectory features of voltage sag signals so as to improve the quality and interpretability of calculation process, the fused lasso generalized eigenvector (FLAG) algorithm is adopted to quickly search for interpretable shapelets sub-sequences from the labeled voltage sag data. After that, the labeled data and samples to-be-classified are subjected to shapelet transformation through the shapelet sub-sequences to obtain the sequential trajectory features. Finally, the random forest is trained on the labeled sequential trajectory feature data to achieve supervised sample classification, which inherits the interpretability of shapelet. To test the feasibility and validity of the proposed voltage sag sources classification method, the simulation cases based on the simulated voltage sag signals were studied. The simulation results show that the proposed method has significant advantages in terms of accuracy and interpretability of voltage sag sources classification.

Data-driven prediction method for characteristics of voltage sag based on fuzzy time series

To inform the power utility and users, and help them reduce the huge financial losses due to voltage sag, it is important to obtain information on voltage sag events in advance. This paper proposes a method for predicting voltage sag characteristics based on fuzzy time series. First, we propose a homologous aggregation method to eliminate redundant data representing the same disturbance event and obtain the time series of voltage sag (TSOVS), which can describe the trend of the voltage sag data. Second, this paper introduces a fuzzification method for the time series of voltage sag based on the fuzzy c-means algorithm (FCMA), which transforms the time series of voltage sag into a fuzzy time series composed of interval symbols, to characterize the mapping relationship between the disturbance and voltage sag event. Furthermore, a hidden Markov model (HMM) of voltage sag is constructed to reveal the transformation relationship among elements in the fuzzy time series, considering the causal relationship between the disturbance and voltage sag event. Finally, the occurrence time and residual voltage of the voltage sag in the future were predicted based on this transformation relation. The measured voltage sags in a province in central China were used to verify the accuracy of the proposed method, prediction results with an accuracy of up to 90%.

Rancang Bangun Alat Deteksi Sag Tegangan Menggunakan Sistem Data Logger

This research discusses the design and manufacture of a voltage sag detection device using a data logger system. Voltage sag detection device to detect the presence of voltage sag which is one of the power quality problems in industrial electrical systems. Voltage sag can cause disturbances in electronic components that are sensitive to changes in voltage values. In this device, the voltage sag is detected using a voltage sensor by reading the RMS voltage from the PLN 220V AC source, then it is converted into a digital signal. The digital signal is sent to the ESP32 module to be processed and outputs a voltage value which is then displayed on the LCD Display. The voltage value data collection is carried out every <10ms and the data taken is the voltage value <90% of the rms voltage. The data taken is then saved to a micro SD card and displayed on the LCD Display. Several tests were carried out to ensure the performance of the voltage sag detector as designed. By switching the 110 / 220VAC transformer connected to the voltage sensor for 10 seconds, 12 seconds and 17 seconds. Based on these tests, the voltage sag detector can read and store the voltage sag data when the voltage read is below 90% of the rms voltage.

A novel approach for voltage sag representation in a chemical industry: A case study

AbstractVoltage sag is a major Power Quality issue which may occur at every possible load point. In the industrial sector, its impact is particularly prominent at it is directly related to financial losses. To eliminate or, at least, mitigate voltage sag, the first, and the most crucial, step is to identify the portions of the network which are vulnerable to voltage sag. Therefore, a novel approach is presented in this article to identify the areas within the network, which are more prone to voltage sag. An impedance matrix‐based model followed by Monte Carlo Simulation (MCS) technique is used for the assessment of the voltage sag. This MCS technique is implemented to simulate the number, type, and locations of faults randomly within the system. Then, voltage sag data are collected, and using Bivariate Frequency Distribution, a graphical representation of voltage sag prone buses is developed, which named Voltage Sag Map. The effectiveness of the proposed approach is validated by adopting an industrial distribution system as a case study in this article.

Detection and Characterization of Time-variant Non-stationary Voltage Sag Waveforms Using Segmented Hilbert Huang Transform

The automation of distribution systems will need a fast and efficient signal-processing tool in order to extract useful information from the power quality (PQ) monitors, smart meters, or Internet of Things, which are essential for real-time monitoring, detection, and faster decision-making capability of smart grids. This article proposes the application of the Hilbert–Huang transform (HHT) method for decomposition of the PQ data into its individual frequency components in order to separate the nonstationary voltage sag waveform containing the fundamental frequency component. After the decomposition of the data using the HHT method, an innovative fundamental frequency-based algorithm is proposed for the accurate detection of voltage sag starting and ending times, which are useful for accurate calculation of point-on-wave and voltage sag duration. The proposed algorithm is further extended for automatic detection of the transition segments to accurately calculate the time-varying and single event characteristics such as voltage sag magnitude and phase-angle jump. The composite method is referred to as the segmented Hilbert–Huang transform (SHHT). The limitations of the existing methods of voltage sag detection using the rms method, the fast Fourier transform method, and the waveform envelope method are also outlined in this article. Finally, the results of the detection and characterization of time-variant nonstationary voltage sag data using the proposed SHHT method is presented and discussed.

Voltage Sag Data Sampling Method for Papermaking Process Based on Compressed Sensing Technology

Important production equipment and sensitive load in the papermaking process are susceptible to the voltage sag event, and production may be interrupted in severe cases. Sampling and recording this event are conducive to judge and analyze the cause of production interruption. To effectively reduce the number of samples and storage costs of the voltage sag data, this paper studies and discusses a data sampling method based on compressed sensing technology, designs an experimental device for verifying the method, proposes relevant indicators and test steps for measuring the performance of the method. The results show that the voltage sag data per unit sampling time is much lower than the minimum value specified by the traditional data sampling method and the original signal can be accurately recovered. Moreover, combined with the actual situation of papermaking, the method is featured by good anti-noise robustness and strong scalability.

Online Recognition Method for Voltage Sags Based on a Deep Belief Network

Voltage sag is a serious power quality phenomenon that threatens industrial manufacturing and residential electricity. A large-scale monitoring system has been established and continually improved to detect and record voltage sag events. However, the inefficient process of data sampling cannot provide valuable information early enough for governance of the system. Therefore, a novel online recognition method for voltage sags is proposed. The main contributions of this paper include: 1) The causes and waveform characters of voltage sags were analyzed; 2) according to the characters of different sag waveforms, 10 voltage sag characteristic parameters were proposed and proven to be effective; 3) a deep belief network (DBN) model was built using these parameters to complete automatic recognition of the sag event types. Experiments were conducted using voltage sag data from one month recorded by the 10 kV monitoring points in Suqian, Jiangsu Province, China. The results showed good performance of the proposed method: Recognition accuracy was 96.92%. The test results from the proposed method were compared to the results from support vector machine (SVM) recognition methods. The proposed method was shown to outperform SVM.

Classification and regression analysis using support vector machine for classifying and locating faults in a distribution system

Various fault location methods have been developed in the past to identify the faulty phase, fault type, faulty section, and distance. However, this identification is commonly conducted in a separate manner. An effective fault location should be able to identify all of these at the same time. Therefore, in this work, a method using a support vector machine (SVM) to identify the fault type, faulty section, and distance considering the faulty phase is proposed. The proposed method uses voltage sag magnitude of the distribution system as the main feature for the SVM to identify faults. The fault type is classified using a directed acyclic graph SVM. The possible faulty sections are identified by estimating the fault resistance using support vector regression and matching the voltage sag data in the database with the actual voltage sag data. The most possible faulty sections are ranked using ranking analysis. The fault distance for the possible faulty sections is then identified using support vector regression analysis and its overfitting or underfitting issues are addressed by the proper selection of a regularization parameter. The feasibility of the proposed method was tested on an actual Malaysian distribution system. The results of faulty phase, fault type, faulty section, and fault distance are analyzed. The performance of the proposed method is compared with various other intelligent methods such as the artificial neural network, deep neural network, extreme learning machine, and kriging method. The test results indicate that the faulty phase and fault type yield 100 % accurate results. All the faulty sections are identified and the proposed method obtains reliable fault location.

Fault location using mathematical analysis and database approach

PurposeThe purpose of this paper is to identify faults in distribution systems which are unavoidable because of adverse weather conditions and unexpected accidents. Hence, quick fault location is vital for continuous power supply. However, most fault location methods depend on the stored database for locating fault. The database is created by simulation, which is time consuming. Therefore, in this work, a comprehensive fault location method to detect faulty section and fault distance from one-ended bus using limited simulated data is proposed.Design/methodology/approachThe work uses voltage sag data measured at a primary substation. Support vector machine estimates the data which are not simulated. The possible faulty section is determined using matching approach and fault distance using mathematical analysis.FindingsThis work proposed a ranking analysis for multiple possible faulty sections, and the fault distance is calculated using Euclidean distance approach.Practical implicationsThe research work uses Malaysian distribution system as it represents a practical distribution system with multiple branches and limited measurement at primary substation. The work requires only metering devices to identify fault which is cost effective. In addition, the distribution system is simulated using real-time PSCAD by which the capability of proposed method can be fully tested.Originality/valueThe paper presents a new method for fault analysis. It reduces simulation time and storage space of database. The work identifies faulty section and ranks the prior faulty section. It also identifies fault distance using a mathematical approach.

Detection of power quality disturbances using an adaptive process noise covariance Kalman filter

This paper proposes an adaptive process noise covariance Kalman filter (APNCKF) for detecting the power quality disturbances present in distorted power signals. Based on the Kalman filter (KF), the new algorithm updates the process noise covariance matrix of the KF by maximizing the evidence density function. The evolution of the estimated process noise exhibits remarkable singularities when the voltage sag occurs or the phase jumps, and the increase in the process noise allows the KF to jump from one stationary regime to another dynamic regime. Thus, the proposed method can determine the start time and the end time of voltage sag and phase jump by using the estimated process noise. Moreover, the APNCKF is used to detect power quality disturbances like voltage sag with phase jumps in presence of additive white Gaussian noise. To test the effectiveness of the algorithm, several time-varying power system signals are simulated with fundamental and harmonic components distorted with voltage sag, phase jump, transient impulse and harmonics. Also, two experiments with real voltage sag data from IEEE 1159.2 Working Group are conducted. Experimental results show that the proposed algorithm can exactly determine the start time and the end time of disturbances by use of the evolution of the estimated process noise covariance matrix.

Support vector classification and regression for fault location in distribution system using voltage sag profile

Fault location identification is an important task to provide reliable service to the customer. Most existing artificial intelligence methods such as neural network, fuzzy logic, and support vector machine (SVM) focus on identifying the fault type, section, and distance separately. Furthermore, studies on fault type identification are focused on overhead transmission systems and not on underground distribution systems. In this paper, a fault location method in the distribution system is proposed using SVM, addressing the limitations of existing methods. Support vector classification (SVC) and regression analysis are performed to locate the fault. The method uses the voltage sag data during a fault measured at the primary substation. The type of fault is identified using SVC. The fault resistance and the voltage sag for the estimated fault resistance are identified using support vector regression (SVR) analysis. The possible faulty sections are identified from the estimated voltage sag data and ranked using the Euclidean distance approach. The proposed method identifies the fault distance using SVR analysis. The performance of the proposed method is analyzed using Malaysian distribution system of 40 buses. Test results show that the proposed method gives reliable fault location.

Premium Power Valuation Method Based on Customer Perception of Utility for High-Technology Manufacturing Customers

Many devices and processes installed at the high-technology manufacturing customer side are very sensitive to voltage sags and short-time interruptions. To meet the customers' requirements for power quality, premium power is needed with a large amount of investment. The investment of premium power depends on the premium power value (PPV) perceived by the customers. One valuation method of PPV based on customers' perceived power utility (PPU) and willingness to pay (WTP) is proposed in this study. The quantitative method of the power utility and the PPU has been studied in detail. Taking customers' loss aversion psychology to economic losses into account, the quantitative assessment methods of the customer WTP and PPV have also been proposed. As a case study, one field investigation to five high-technology manufacturing customers located in one High Technology Park in Western China has been done. With the help of the field-recorded voltage sag data and customer economic losses data over the last 10 years, the proposed method was applied to valuate these five customers. The valuated results compared with investigated results proved that the proposed method is correct, rational, and feasible.

Faulted Segment Identification in Multi-phase Distribution Systems Using Sequence Component Modeling

This article discusses the potential for identifying faulted segment(s) on multi-phase distribution primaries using sequence component modeling and standard three-phase short-circuit solvers. Identifying faulted segments in a timely manner speeds up fault isolation and restoration processes, contributing to greater feeder reliability. The methodology utilizes fundamental frequency current magnitudes from the feeder head in conjunction with an approximate sequence model of the multi-phase feeder including single-phase and double-phase line segments augmented with dummy lines. Under certain fault conditions, multiple faulted segments may be yielded for which a new faulted segment reduction method based on voltage sag data is proposed. The Saskpower network simulated in MATLAB/Simulink (The MathWorks, Natick, Massachusetts, USA) environment was used to evaluate the faulted segment identification and reduction methods. The evaluation studies show quite promising results in uniquely identifying the actual faulted segment with reduced measurement and modeling requirements.