To ensure the power system operates safely and reliably, it is essential to monitor and evaluate the health condition of power equipment on-line or off-line. This study reviews the research status in condition monitoring and diagnosis of power equipment, including transformer, gas insulated switchgear, cable, external insulation, generator, and power capacitor in recent years. Although much progress has been made in technologies of condition monitoring and fault diagnosis such as test accuracy, fast and accurate fault localisation and recognition of fault types, there are still many deficiencies which needs further research work, including the reliability of signal collection from sensors, the accuracy of data treatment and analysis, anti-interference performance of test equipment, appropriate models used for condition evaluation. The prospective of condition monitoring and diagnosis technologies of power equipment are also presented in this study. It is proposed that the application of big data, internet of things and cloud computing should be expected and given special attention in the near future.

Condition monitoring and diagnosis of power equipment: review and prospective

This paper presents a state-of-the-art review on machine learning (ML) based intelligent diagnostics that have been applied for partial discharge (PD) detection, localization, and pattern recognition. ML techniques, particularly those developed in the last five years, are examined and classified as conventional ML or deep learning (DL). Important features of each method, such as types of input signal, sampling rate, core methodology, and accuracy, are summarized and compared in detail. Advantages and disadvantages of different ML algorithms are discussed. Moreover, technical roadblocks preventing intelligent PD diagnostics from being applied to industry are identified, such as insufficient/imbalanced dataset, data inconsistency, and difficulties in cost-effective real-time deployment. Finally, potential solutions are proposed, and future research directions are suggested.

Condition Monitoring Based on Partial Discharge Diagnostics Using Machine Learning Methods: A Comprehensive State-of-the-Art Review

This paper examines the potential in extracting the instantaneous location of maritime moving targets using a passive multistatic radar with Global Navigation Satellite Systems (GNSS) as illuminators of opportunity and a single receiver. This paper presents a theoretical framework for the localization of a moving target from a set of bistatic range measurements. The algorithm and its predicted accuracy are presented. The localization is achieved by what is essentially a multilateration technique, which can be applied while the transmitting platform is also in motion. The algorithms and the accuracy predictions, as a function of the number of transmitters, have been experimentally confirmed via a dedicated experimental campaign, where two different maritime targets were detected by up to 12 GNSS satellites belonging to different satellite constellations (GPS, GLONASS, and Galileo) simultaneously. To the best of the author’s knowledge, these are the first results of their kind and on this scale not only for GNSS-based passive radar but also for multistatic radar in general.

/pdf/maritime-moving-target-localization-using-passive-gnss-based-20ej5xvw22.pdf

Maritime Moving Target Localization Using Passive GNSS-Based Multistatic Radar

The paper, prepared by CIGRE WG D1.03 (TF 09), presents the guidelines for risk assessment procedure on defects in GIS based on PD diagnostics. The procedure, described in detail in CIGRE Technical Brochure 525, starts with sensitive PD measurement to detect the critical defects and follows with identification of the type of the defect and its location inside the GIS. This information taken together with other essential data from laboratory measurements, manufacturer's experience, design aspects and trend analysis of the PD activity, are the base for the estimation of the criticality of the defects. Finally, the risk assessment is performed based on the estimated dielectric failure probability and failure consequences that can be different in case of on-site testing or in service activity.

Risk Assessment on Defects in GIS based on PD Diagnostics

Partial discharge (PD) is regarded as a precursor to plant failure and therefore, an effective indication of plant condition. Locating the source of PD before failure is key to efficient maintenance and improving reliability of power systems. This paper presents a low cost, autonomous partial discharge radiolocation mechanism to improve PD localization precision. The proposed radio frequency-based technique uses the wavelet packet transform (WPT) and machine learning ensemble methods to locate PDs. More specifically, the received signals are decomposed by the WPT and analyzed in order to identify localized PD signal patterns in the presence of noise. The regression tree algorithm, bootstrap aggregating method, and regression random forest are used to develop PD localization models based on the WPT-based PD features. The proposed PD localization scheme has been found to successfully locate PD with negligible error. Additionally, the principle of the PD location scheme has been validated using a separate test dataset. Numerical results demonstrate that the WPT-random forest PD localization scheme produced superior performance as a result of its robustness against noise.

/pdf/improving-rf-based-partial-discharge-localization-via-1o5s4yawyg.pdf

Improving RF-Based Partial Discharge Localization via Machine Learning Ensemble Method

Ultra-high-frequency (UHF) sensing technique has been introduced to detect and localize partial discharge (PD) sources in air-insulated substation (AIS). This paper presents a probability-based algorithm to localize multiple PD sources which may occur simultaneously in different power equipment. Assuming that the time difference of arrival (TDOA) between all pairs of antennas in a array are normally distributed, the probability density function (PDF) of PD source coordinates can be obtained by substituting the linearized form of time difference equations into PDFs of TDOAs. When large number of PD signals are recorded, the joint PDF (JPDF) can be calculated from the product of PDF of each TDOA. Then the PD coordinates to be solved are regarded as with highest probability, and can be solved by taking the derivative of JPDF. In the case of multiple PD sources, mixed UHF signals are separated by clustering the TDOA vectors with K Means clustering method. PD experiments are performed to test the presented algorithm, and the localization accuracy of proposed algorithm is compared with other typical methods such as Newton-Raphson, Particle Swarm Optimization and plane intersection method. The results indicate that the probability-based localization algorithm reasonably integrates the TDOAs of continuous signal sequence, which can effectively reduce the influence of TDOA estimation errors and improve the localization accuracy.

Localization of multiple partial discharge sources in air-insulated substation using probability-based algorithm

Ultra-high-frequency (UHF) antenna array system is applied to detect and locate partial discharge (PD) sources in air-insulated substation (AIS). Time-difference localization algorithm is generally adopted in such system in which the antenna array is of large scale. In this paper, the phased array theory from radar field is introduced to estimate the direction-of-arrival (DOA) of PD source to reduce the array size. The DOA estimation results are calculated by two-sided correlation transformation (TCT) focusing algorithm and multiple signal classification (MUSIC) algorithm. Based on that, the optimum arrangement of antenna array is investigated considering both high directional accuracy and small array size. The DOA estimation performances of three array geometries namely uniform linear array (ULA), uniform circular array (UCA) and uniform Y-shaped array are compared through Cramer-Rao lower bound (CRLB), numerical simulation and experimental testing. Moreover, the number of elements and element spacing are optimized. The results indicate that 4-elements UCA with 0.2m element spacing is the optimum array arrangement for PD location in AIS.

UHF antenna array arrangement optimization for partial discharge direction finding in air-insulted substation based on phased array theory

Partial discharge (PD) measurement is an effective tool for insulation condition assessment of high-voltage power equipment. The occurrence of multiple PD sources causes great difficulty on pattern recognition and failure risk assessment. This paper presents a self-adaptive PD separation algorithm based on optimized feature extraction of cumulative energy (CE) function. The CE functions in time domain (TCE) and frequency domain (FCE) are calculated from PD waveforms and their FFT spectrums, respectively. By using an oblique line to cross the CE curves, width features are extracted from the intersection points between them. Through the mathematical morphology gradient (MMG) operation, sharpness features are extracted to characterize the rise steepness of CE. It is found that the separation capability of width and sharpness are dependent on the pre-selected oblique line and the structure element length (SEL) in MMG, respectively. In order to obtain satisfactory PD separation results for various experimental conditions, a density-function based parameter is proposed to evaluate the separation capability, and the oblique line and SEL are optimized with the goal of maximizing the evaluation parameter. A clustering algorithm is adopted to discover different clusters in feature space and separate PD signals. The separation algorithm is examined with mixed PD current pulses and ultra-high-frequency (UHF) signals acquired from experiments in laboratory and on-site equipment. The results indicate that the self-adaptive separation method is immune to the change of experimental conditions, and is effective for separating mixed PD signals.

Self-adaptive separation of multiple partial discharge sources based on optimized feature extraction of cumulative energy function

The invention discloses a partial discharge direction finding method based on an ultrahigh-frequency phased array principle in a substation. The method includes following steps: (1) acquiring and recording partial discharge signals in the substation by employing the ultrahigh-frequency phased array and a digital detection device with high sampling rate; (2) focusing the partial discharge signals to form narrowband signals at a reference frequency point by employing a broadband focusing algorithm; (3) performing direction-of-arrival estimation on the narrowband signals by employing a narrowband direction finding algorithm; and (4) obtaining the point with maximum azimuth density with combination of multiple direction-finding results, namely the azimuth of the partial discharge signals. According to the method, the size of the ultrahigh-frequency phased array is small so that tour inspection is facilitated, and the spatial resolution is high; and according to the method, for the partial discharge signals in the substation, the partial discharge signals are converted to the narrowband signals by employing the focusing algorithm, and the azimuth of the partial discharge signals is obtained by employing the direction finding algorithm so that the direction-finding precision is high, and the ability of suppressing noise interference is great.

Partial discharge direction finding method based on ultrahigh-frequency phased array principle in substation

The invention discloses an optimal arrangement method of an UHF phased array used for substation partial discharge direction finding. The optimal arrangement method comprises the steps of: 1) making a number of array elements and array element spacing of the phased array unchanged, deriving Cramer-Rao low bound expressions in different geometric shapes, and optimizing a geometric shape of the UHF phased array; 2) making array element spacing of the phased array after geometric shape optimization unchanged, drawing a relation curve between direction finding errors and the number of array elements in numerical simulation, and optimizing the number of array elements of the UHF phased array by combining with a construction cost of an acquisition system; 3) and acquiring partial discharge signals in a substation for the phased array after geometric shape and array element number optimization, analyzing frequency spectra of the signals, and optimizing array element spacing of the UHF phased array. The optimal arrangement method adopts the UHF phased array which is small in size, has the advantages of high anti-interference ability and high spatial resolution, and is convenient for partial discharge patrol inspection.

Qing Liu

Papers

Localization of multiple partial discharge sources in air-insulated substation using probability-based algorithm

UHF antenna array arrangement optimization for partial discharge direction finding in air-insulted substation based on phased array theory

Self-adaptive separation of multiple partial discharge sources based on optimized feature extraction of cumulative energy function

Partial discharge direction finding method based on ultrahigh-frequency phased array principle in substation

Optimal arrangement method of UHF phased array used for substation partial discharge direction finding