Search

On-line partial discharge (PD) detection and location has proven to be a valuable tool for assessing the condition of medium-voltage (MV) power cables in service. A few advances are discussed in this paper. It is of importance to know all factors affecting PD activity in order to judge its severity. Cable temperature, more specifically its variation due to load cycling, is one key parameter. From the variation of the signal propagation velocity along the cable it is shown that the temperature can be monitored within one degree centigrade accuracy. The PD activity was found to respond upon temperature cycling caused by load variation. Moreover, data gathered from years of field experience has led to more insight in the predictability of upcoming faults based on (trends in) PD activity. It is possible now to identify the time between the moment partial discharges appear (or reach a certain level) and a final breakdown for different types of cable insulation. Occasionally, cable faults occur without being preceded by PD activity. For distribution network operators, a quick location of a MV cable fault, either a direct complete fault or a self-healing (or intermittent) fault, will contribute to reduce the outages both in duration and frequency. The earlier applied PD location technique is extended to capture also faults based on first arrival of the associated transients. This fault pinpointing is possible for any MV power cable type within 1% of the cable length, independent of the network grounding and whether a fault is a permanent one or a self-healing one. This paper presents the various techniques and results from test cases and a real captured field fault.

Summary High frequency signal components from partial discharge in an underground power cable propagate according to available modes provided by the cable structure. Which mode or combination of modes is detected depends on type and installation of sensors. Without the ground return impedance, the earth screen voltage is assumed to be equal to the ground potential and all current will flow through phase conductor(s) and earth screen. However the soil may contribute to the transmission line propagation modes and affect the detected partial discharge signals. This paper investigates the effect of the ground return path on partial discharge detection at cable ends. Existing theory and modeling (modal analysis, ground impedance) are customized for the analysis. Transmission line parameters measurements approve the modelings. Time domain signal propagation simulation shows that the ground mode has a minor contribution to the signal propagation. The implication for partial discharge detection is discussed. Copyright © 2016 John Wiley & Sons, Ltd.

An on-line partial discharge (PD) monitoring and location system for medium voltage cable circuits was developed previously. This paper explores ways to extend its range of application. The extension includes PD location method by time domain reflectometry (TDR) when reflections are not easily identifiable. The functionality of the PD monitoring equipment can also be widened by sensing other quantities related to the condition of the cable insulation using the same device. Dynamic cable temperature monitoring can be achieved by recording variation in the high frequency signal propagation velocity along the cable. Water ingress in paper-insulated lead-covered (PILC) cable decreases the cable’s characteristic impedance while it increases the permittivity of the insulation. It can be observed by changes in the reflection pattern from the cable or by a lowered propagation velocity. Instead of recording reflection patterns in time domain, a frequency domain approach based on an impedance scan is investigated to be applied for cables in service.

Models in high frequency range for underground power cable connections are essential for the interpretation of partial discharge (PD) signals arising e.g. diagnostic techniques. This paper focuses on modeling of power cable joints. A lumped parameter odel and a cascaded transmission line model are proposed based on scattering parameters (S -parameters) measurement on a 10 kV oil-filled PILC-PILC straight cable joint in the frequency range of 300 kHz-800 MHz. It is shown that the lumped model is suitable for up to 10 MHz while the transmission line model can cover the whole frequency range. The cascaded transmission line model is applied to simulate the reflection on a 150 kV single core XLPE straight joint. Comparison between measurement and simulation indicates that the model parameters (characteristic impedance and propagation coefficient) can be matched to predict the joint’s propagation characteristics.

Matched filtering can be applied to recorded signals for maximum detection sensitivity during online partial discharge (PD) measurements in medium-voltage cable circuits. PD waveform predictions are combined with a noise spectrum estimation to construct matched filters. The PD waveform is specific for each cable circuit under test and the predictions are based on online system identification measurements and a standard cable model. For some cable circuits, for instance with a ring-main-unit along the cable, the predicted waveforms only approximate the real waveforms due to signal distortion not accounted for in the model. This results in suboptimal detection sensitivity. This paper describes an automated procedure for updating the matched filters based on detected PD signals. The algorithm analyzes detected pulses to judge whether the current set of matched filters perform optimally. If not, the current set of matched filters is updated, increasing the PD detection sensitivity and potentially also the location accuracy. At the same time, recognition of recurring disturbances is improved and better rejection criteria can be defined. The method is tested on signals measured on live cable connections.

Online partial-discharge (PD) monitoring of medium-voltage power cables is traditionally performed on a cable between two consecutive ring main units (RMUs). PD location is achieved using a sensor at each cable end. Monitoring multiple consecutive cables, with RMUs in between, using a single monitoring system is more efficient. In addition, substations and RMUs without possibilities for sensor installation can be circumvented by installing the sensor at the next RMU. This paper studies how RMUs and substations along the cable under test affect online monitoring, including their influence on detection sensitivity, location accuracy, and charge estimate accuracy. Single-sided PD measurement, including PD location, with an RMU or substation at the far end is also considered. Models for RMUs and substations are proposed and verified by measurements. The performance of online PD monitoring is studied for a number of network configurations.

Time domain reflectometry (TDR) applied to live cables is hampered by the absence of a clear reflection from the far end. Instead, location of partial discharge (PD) can be extracted from small structures in recorded patterns arising from any impedance variation along the signal propagation channel. This paper explores two approaches for a single-sided PD location system. The first method compares similar signal sequences occurring in PD pattern and pattern obtained from an injected signal. The second method exploits the change of impedance at a far end Ring Main Unit (RMU) by introducing ferrite material, whose effect can be switched on and off. Experiment on medium voltage (MV) cable with about 560 m length shows that both methods can locate the PD within 2% uncertainty.

Models in high frequency range for underground power cabling system are the basis for the analysis of many condition diagnostic techniques, for instance, partial discharge (PD) monitoring. This paper focuses on modeling of power cable joints. A cascaded transmission line model is proposed for 10kV oil-filled PILC-PILC straight power cable joint in thefrequency range of 300 kHz-800 MHz. It can be applied to interpret PD signal transmission and reflection behavior in power cable connections. The model is verified by S-parameter measurement.

A power cable model for partial discharge (PD)pulse propagation requires knowledge of all factors influencingthe transmission line parameters of the cable (characteristicimpedance, attenuation coefficient and propagation velocity). These frequency dependent quantities are related, amongst others, to the dielectric properties of the semiconducting layers. This paper discusses methods to estimate the characteristicimpedance, attenuation coefficient and propagation velocity using the cable geometryand the material properties of conductor, conductor shield, XLPE and semi-conducting layers. Measurementson cable samples show that the cable characteristics can be approximated withsufficient accuracy for PD signal propagation modeling, e.g. estimating the PD magnitude at their origins.

Partial discharge (PD) location in online diagnostics on medium voltage cables is achieved using a sensor at each cable end. Monitoring consecutive cables with a single monitoring system would, however, be more efficient. Moreover, substations and ring main units (RMU) along a cable connection without possibilities for sensor installation can be circumvented by installing the sensor at the next RMU. This paper studies the influence of RMUs and substations along the cable under test on online PD monitoring, including their influence on detection sensitivity, location accuracy and charge estimate accuracy. Models for RMUs and substation are proposed and verified by measurements. The performance of online PD monitoring is studied for a number of network configurations.

For specific circuits the accuracy of online partial discharge (PD) detection and location (PD-OL, [1]) systems is impaired by complications in the cable circuit configuration, such as ring-main-units (RMUs) along the cable under test. For maximum sensitivity matched filtering is applied for extracting PD pulses from the recorded signal. The matched filters are based on a series of prior online system identification measurements and a standard cable model. Especially, with an RMU along the cable, the matched filters only approximate the real waveforms due to signal distortion. This results in sub-optimal detection sensitivity. This paper describes automated procedure for updating the matched filter bank based on measured (large) PD signals. The algorithm analyzes detected pulses to judge whether the current templates perform optimally. If not, a new template is created increasing the PD detection sensitivity and the location accuracy, and also improving the recognition of recurring disturbances. The method is tested on signals measured on a live cable connection.

Online partial discharge (PD) detection and location systems for medium-voltage (MV) cables have been developed over the past years and are currently introduced by utilities on an increasingly large scale. This paper addresses a few of the new challenges related to the integration of these systems in various cable networks. The effect of ring-main-units (RMUs) and substations on the propagation of PD signals is studied to determine feasible options for PD monitoring on two consecutive cables connected through an RMU or a substation. Different algorithms for PD pulse arrival time estimation are analyzed for robustness and PD origin location accuracy. An algorithm based on the signal energy proved to be most suitable for complex circuits where the PD shape is not only distorted by the propagation through the cable, but also by complex impedances in the RMUs along the cable under test and each cable end. Matched filters for PD detection can be improved using measured PD signals. This results in improved detection sensitivity and location accuracy.

Online partial discharge (PD) monitoring systems are traditionally installed at a single mediumvoltage (MV) cable connection between two ring-mainunits (RMUs). It is more efficient to monitor two or more consecutive cables using a single monitoring system. Moreover, practical experience with the PD-OL system [1], shows that for substations, with many parallel MV cables, and RMUs installing the inductive sensor may be hampered or even impossible. in this paper the influence of RMUs and substations on the propagation of PDs is studied. An RMU or substation can be modeled as a combination of complex impedances representing switchgear, transformer and MV cables. A PD pulse from a cable encounters a load impedance that does not match the cable's characteristic impedance, resulting in partial reflection and partial transmission transmission to other cables. Models for RMUs and substations are proposed and verified by measurements. Feasible options for online PD monitoring through RMUs or substations are determined. .

On-line partial discharge detection and location equipment for medium voltage cables have become available recently. Efficient implementation can be realised by monitoring cable connections containing several cable sections with ring main units or even a complete substation in between. This paper discusses the consequences related to diagnoses of complex cable connections by a single partial discharge detection system. The feasibility of monitoring several power cable sections at once is illustrated with a practical example obtained with recently installed systems in the Dutch grid. It is shown that detection and location of partial discharges is indeed possible for signals travelling over long connections including ring main units.

Online partial discharge (PD) detection and location in cable systems is a valuable tool for the estimation of the condition of the system. Different types of cables are encountered in medium-voltage (MV) connections and appropriate models must be constructed to predict PD signal propagation. For three-core power cables with a single common earth screen a multiconductor transmission line (MTL) model is required. This paper presents an MTL model of a such a cable, including interpretation of the decoupled solution as convenient propagation modes. A complicating factor in the estimation of the transmission line parameters of these propagation modes is the presence of semiconducting layers. These layers have a significant influence on the transmission line parameters. Unfortunately, the dielectric properties of these layers are usually unknown for the frequency range of interest for PD diagnostics. It is shown that nonetheless the characteristic impedance and propagation velocity can be estimated using information available in most cable specifications. the estimated values are validated using pulse response measurements on a cable sample. ©2009 IEEE.

A power cable model for partial discharge (PD) pulse propagation requires knowledge of all factors influencing the transmission line parameters of the cable (characteristic impedance, attenuation coefficient and propagation velocity). These frequency dependent quantities depend, amongst others, on the dielectric properties of the semi-conducting layers. For existing power cable connections, but often also for newly manufactured cables, these quantities, which vary between different cable types, are not known with acceptable accuracy. This paper discusses methods to estimate the characteristic impedance and propagation velocity using the cable geometry and the dielectric properties of XLPE, but without using the dielectric properties of the semiconducting layers. Measurements on cable samples show that these approximations are valid with sufficient accuracy. The attenuation of PD signals can only be accurately predicted with detailed knowledge of the dielectric properties of the semi-conducting screens.

Accurate location of partial discharges in power cable systems, based on arrival times, is critical for the identification and assessment of defects. This paper evaluates different time-of-arrival algorithms in order to determine which method yields most accurate location under different circumstances. These methods are based on threshold, Akaikepsilas information criterion, energy criterion, Gaborpsilas signal epoch and phase in frequency domain. Several criteria are defined by which the algorithms are evaluated. These criteria include the sensitivity to noise, pulse shape and effect of load impedance. The sensitivity of the methods upon varying these quantities is evaluated analytically and by means of simulations. From the results the energy criterion method and the phase method show the best performance.