Search

This paper proposes a new modulated model predictive control method for the control of modular multilevel converters (MMCs) in voltage source converter-high voltage dc systems. The proposed method retains the advantages of the conventional finite control set-model predictive control methods by programing the nonlinear properties of the MMC into the design calculations while minimizing the ac line and circulating current ripples and steady-state error by generating modulated switching signals with a fixed switching frequency. In this paper, first, the predictive modeling of the MMC is provided. Next, the proposed control method is described. Then, the application of the proposed method to an MMC system is detailed. Finally, experimental results from an MMC system connected to a three-phase grid are provided to validate the theoretical outcomes.

Vibro-acoustic measurement on a transformer's on-load tap changer (OLTC) can provide indications on its mechanical condition. Recently, a joint vibro-acoustic and arcing measurement system has been proposed, which can correlate the vibro-acoustic signal to mechanical events of OLTC's operation. However, there are still considerable difficulties in extracting useful information from both vibro-acoustic signal and arcing signal in a synchronized manner without any distortions in the extracted signals. In this paper, Savitzky–Golay filter is introduced to process the signals acquired from a joint vibro-acoustic and arcing measurement system installed on in-service OLTCs. It proves that the Savitzky–Golay filter can process both vibro-acoustic and arcing signals induced by OLTC, extract essential information without any time delay from both types of signals, and retrieve voltage phase information from the arcing signal. The methodologies developed in this paper can improve the visibility of OLTC's mechanical operation for an effective online condition monitoring.

This paper 1) introduces a new approach for dependable and secure detection of loss-of-field (LOF) events based on estimating rotor signals, 2) implements digital algorithm of the corresponding LOF relay in an industrial hardware platform, and 3) evaluates performance of the LOF relay in a real-time-digital-simulator (RTDS) based control-hardware-in-the-loop (CHIL) environment. The main features of the proposed relay as compared with the conventional approaches include 1) capability to discriminate between system disturbances and LOF events, and 2) dependability and higher speed of the LOF detection mechanism. To evaluate the proposed approach performance for various types of LOF events, a detailed excitation system is modeled and implemented in the RTDS-based CHIL environment. This paper also verifies superior performance of the proposed LOF relay as compared with that of a conventional impedance-based LOF relay, using the RTDS-based CHIL platform.

The Nelson River HVDC system currently consists of two thyristor-based bipole schemes: Bipoles I and II. Pole 1 in Bipole I was found to have suffered from a serious thyristor leakage problem (TLP) after 20 years in service. In order to understand, evaluate, and resolve this problem, a comprehensive study was conducted in 2013 at Manitoba Hydro, which involved thyristor voltage monitoring, leakage current measurements, and simulations. The study revealed that the majority of thyristors in two of the three valve groups at both ends were leaky (i.e., exhibiting increased leakage currents) to various degrees. This paper provides an overview of the TLP and related investigation. The effects and symptoms/behaviors of the leaky thyristors are illustrated with the study results for one of the worst valves at each end. The failure mechanism of the leaky thyristors is also discussed. This paper should provide a clear understanding of the TLP from the user's perspective. A question, however, remains with regard to the root-cause of the TLP. Further research is required to answer this question.

Power-electronic sources, such as Type 3 wind turbine generators and static synchronous compensators (STATCOM), interface to the grid through, partial- or full-scale, and power converters that have inherently fast switching capability to control their output current during short circuits. The short-circuit current is a function of the specific converter control algorithm and differs significantly from the conventional rotating machine sources without converter interfaces. Therefore, if a transmission-line protection scheme is designed for conventional sources, (not taking into account these differences in short-circuit current characteristics), reliability can be at risk. Using real-life short-circuit currents on lines supplied by sources having a power converter interface, this paper illustrates the reliability risk to conventional line protection schemes, in particular, to those which use negative-sequence quantities for the detection of unbalanced faults. This paper discusses the protection schemes, adopted by BC Hydro–a large Canadian Electric Utility, for transmission lines interconnecting Type 3 wind turbine and STATCOM sources. Their application for ground faults is independent of the converter control algorithm as long as the source is interconnected to the grid via a transformer which is a source of zero-sequence current.

Locating a fault in a series-capacitor-compensated transmission line (SCCTL) is a challenging task due to the action of a metal-oxide varistor a nonlinear element present as a part of the protection system of the series capacitor. In this paper, an accurate fault-location algorithm that uses synchronized measurements from both ends of the SCCTL is presented which provides fault-location results without using the model of MOV or natural fault loop for single-phase-to-ground, and double-phase-to-ground faults. Another salient feature of the proposed technique is that the subroutines for locating faults in different sections transmission line yield almost identical fault-location results regardless of which section the transmission line is faulted. The proposed technique could also be extended to the double-circuit transmission lines. First, the proposed technique is introduced and its features are elaborated through detailed mathematical analysis. Thereafter, a 500- $\text{kV}$ system with an SCCTL is designed in PSCAD, while the fault-location algorithm is modeled in MATLAB. The proposed algorithm is tested through simulations covering various fault scenarios in an SCCTL. For performance evaluation, the comparative analysis of the proposed technique with a well-known existing technique is performed in this paper.

Series compensated transmission lines challenge legacy protection schemes. In this paper, a dynamic state estimation based protection (EBP) method is proposed to address these challenges. The method requires GPS synchronized measurements at both ends of the line and a high-fidelity model of the protection zone. The paper presents the dynamic model of the protection zone and its impact on the performance of the protection scheme. Numerical simulations show that the method can correctly identify faults, independently of position and type. The paper also compares the proposed method versus legacy protection functions such as distance protection and line differential protection. The comparison shows faster detection of internal faults, immunity to current inversion caused by series capacitors (SCs) and improved detection sensitivity for high-impedance faults.

For the fast transient study of power transformers, several types of models are available that can be categorized into three groups: white box models, black box models, and gray box models. The identification of their parameters and their usages make the models different. Since white and black box models cannot cover certain cases, such as the failure analysis of a transformer with unknown geometrical data, a new type of model has recently become important for industry and academic researchers. This model, called the gray box model, is intensely analyzed in the current paper, both theoretically and experimentally. The gray box model conceptually lies between the black and white box models. The outcomes of this work have been verified using a suitable test object. The parameter identification of the model is performed by artificial methods. Some useful new ideas, including applying the Weibull Distribution function in inductance estimation and employing an exponential function for the series resistance of windings, are developed in this paper. The number of unknown parameters to be identified in the gray box model is reduced by applying these ideas, which makes the model more simple and feasible.

An effective computational tool for the simulation and design of voltage dividers has been realized. It takes advantage of a finite-element procedure for the accurate computation of the divider stray capacitances. These latter are then the input of a circuit solver procedure based on the modified node potential technique, which provides the divider frequency response. Furthermore, a test divider has been set up for the purpose of the tool validation. A very good agreement has been found between measured and computed results. This paper highlights that a correct representation of the resistors is necessary in the finite-element method (FEM) model, during the computation of the stray parameters. This issue has been clarified for the first time in this paper. In addition, a comparison between the FEM and boundary-element method has been carried out before the implementation of the numerical code, stating the superior effectiveness of FEM. The computational tool has been finally tested on a real voltage divider. Good results have been found in terms of computational accuracy. Limited discrepancies between measured and computed results have been found up to 10 kHz, with 5.7% for the scale factor and less than 24 mrad for the phase error.

In conventional HVDC systems where the harmonic filters and shunt capacitors are installed at the ac grid side, the converter transformers inevitably suffer the detrimental impacts caused by harmonic penetration. In this paper, combined with an inductive filtering method, a parallel delta-connected filtering-winding scheme for 12-pulse converter transformer units is proposed to reduce the harmonic currents flowing through the transformers and, thus, to attenuate the harmonic effects on them. This paper describes the presented transformer configuration in terms of the wiring scheme of the 12-pulse converter transformer unit and corresponding flowing paths of different harmonic currents with the orders of $3,6n\pm 1\left(\mathrm{n}=1,3,5,\ldots \right)$ , 11 and 13 in the filtering winding circuit. Then, the mechanism of the proposed scheme on harmonic suppression to the converter transformers is explained theoretically. In an attempt to validate the theoretical analysis, simulations based on PSCAD/EMTDC and experiments in a laboratory prototype system are performed to evaluate the harmonic eliminating performance of the proposed scheme.

Vibration measurement has been adopted in some utilities in Australia, Canada, and several European countries for online condition monitoring of the power transformer's onload tap changer (OLTC). By comparing measured vibration signals at different stages, it might be possible to assess changes of OLTC's condition. However, there are still considerable difficulties in correlating vibration signals to the events of OLTC operation, which may impair the capability of vibration measurement for OLTC condition monitoring. Therefore, arcing measurement is proposed in this paper to complement the vibration measurement. Arcing is provoked when the OLTC switching contact closes at a fixed tap position and can lead to electromagnetic signals flowing through transformer windings, finally reaching earth. The arcing measurement is achieved by using a high-frequency current transducer (HFCT) clamping on the transformer's grounding cable. The joint vibration and arcing measurement can provide a better means for interpreting events involved in OLTC operation and facilitating improved OLTC condition monitoring. Since HFCT measured arcing signals can be coupled with noise, a probabilistic wavelet transform is thus employed in this paper to extract arcing signals from noise. Field measurements on two different types of OLTCs are performed using the joint vibration and arcing measurement system to validate the proposed methodology.

This paper presents challenges and potential impacts of photovoltaic (PV) integration in the low-voltage downtown secondary networks (downtown networks). In the conventional secondary networks, substation feeders are the sole source of electric power and establish unidirectional power to the downtown network. The network protectors prevent the flow of power from inside the network to the upstream feeder by disconnecting the circuit to protect the feeder transformer against upstream faults. The assumption of unidirectional power flow can be violated by PV generation due to the possibility of excess power inside the network. It is shown in this paper that a large number of network protector trips can occur and, thus, voltage collapse may follow even in low PV penetration levels. In addition, it is demonstrated that the reclose action of the network protector relays is adversely affected by the PV power. Other adverse effects of such distributed-generation units, such as voltage profile, line overloads, and flicker, are also briefly discussed. Finally, a solution is proposed, based on differential currents, to upgrade the network protector relays in order to avoid false trips due to excessive PV power. Part of the New Orleans downtown network is modeled and the study is performed through simulations.

Proliferation of distributed generations (DGs) and power-electronics-based loads is bringing about more harmonic-polluted power signals. While some failures may occur as a consequence of nonsinusoidal current or voltage waveforms, existing reliability models of protection relays have not taken the harmonic-related failures into consideration. This paper develops a comprehensive Markov reliability model to categorize the possible functional states of a component, say a transmission line, protected by a relay, operating in a harmonic-polluted environment. In order to make the proposed model practically tractable, it is further simplified in two ways, through merging the states of the same consequence. The first simplified model focuses on the power component reliability assessment; while the other is built for reliability analysis of a protection system. Calculation of transition rates and how the resultant model is mathematically evaluated are then discussed. Numerical analyses and outcomes are based on the real-world data taken from Canadian substations. Accordingly, the results and conclusions drawn in this paper would be interesting to both academia and industry. Moreover, the proposed reliability assessment methodology is readily applicable in scrutinizing the impacts of harmonic pollution on the protection devices of other power components.

Several methods of attenuation of very-fast transient overvoltages (VFTOs) in gas-insulated switchgear (GIS) are under development. An approach that is currently gaining high attention among researchers is based on the application of magnetic cores dissipating the energy associated with VFTO due to eddy current losses and magnetic hysteresis losses. This paper presents an advancement in this approach by introducing a new material type (denoted in this paper as “nanoflakes”). This new material comprises macro pieces of nanocrystalline ribbons (with the size ranging in square millimeters), acquired from crumbled nanocrystalline cores of a standard design. The material is characterized by relatively high magnetic permeability, poor electrical conductivity, and high saturation flux level. The material is suitable for application as a coating layer applied to the surfaces of an inner and/or an outer GIS conductor, as a fill-in material in the GIS shielding elements, or as a standard shaped core. This paper presents the material major features, outlines its development process, and reports on validation of the material effectiveness in real VFTO conditions. The validation is based on the measurement of VFTO attenuation in a full-scale 550 kV GIS test setup.

This two-part paper is intended to clarify definitions in dual transformer modeling that are vague, provide accurate modeling guidelines, clarify misconceptions about numerical instability, provide a unified dual model for a specific type of transformer, and introduce new paths of research to the power systems/electrical machinery community for low-frequency transients. Part I discussed the topology of duality-based models and some important issues, such as the most common approaches to derive dual models and their variety, the equivalence of negative inductance and mutual couplings to represent the leakage inductance of three-winding transformers, and the numerical oscillations caused by the use of nontopological models. This part of the paper discusses and compares white-, gray-, and black-box models. The paper also reviews hysteresis models (static and dynamic) and highlights the differences between the air-core inductance and saturation inductance. The available dual models for the representation of the transformer tank are then presented. A unified and accurate model of a three-phase core-type transformer adequate for all low-frequency transients is presented. Finally, concrete guidelines are presented for the appropriate selection of the model topology and parameters for different low-frequency transient studies.

Modular multilevel converters (MMCs) have become the converter topology of choice for voltage-source converter-high-voltage direct-current systems. Excellent work has previously been conducted to develop much needed average value models (AVM) for these complex converters; however, there a number of limitations as highlighted in this paper. This paper builds on the existing models, proposing numerous modifications and resulting in an enhanced MMC-AVM, which is significantly more accurate and which can be used for a wider range of studies, including DC faults.

In China, two modular multilevel converter (MMC) projects were recently commissioned, and another two are currently under construction. The real-time hardware-in-the-loop (HIL) test bench played an important role in validating the manufacturer's controllers. The full detail MMC model does not fit with the HIL test bench, which has to solve the circuit containing numerous switches and handles a large amount of inputs and outputs at a small time step for real-time simulation. The main challenge is to find a method to model and simulate the MMC systems with sufficient detail, accuracy, and speed. This paper presents an equivalent circuit method for the HIL test bench. The circuit inside the submodule is represented by mathematical equations, implemented in CPU or a field-programmable gate array, and solved in parallel to achieve real-time performance. This method-based test bench is used in those MMC projects in China and connected to the manufacturer's controllers for HIL tests. The model accuracy and simulation speed achieved by this method met the requirements of the HIL tests. In this paper, various scenarios are tested in an MMC HVDC study system. The results achieved by the proposed method have high agreement with those of a reference model in EMTP-RV.

The dc cables of the voltage-sourced converter-based HVDC link, including the modular multilevel converter (MMC) configuration, must be protected by surge arresters against transient overvoltages. The selection of the appropriate surge arresters requires a comprehensive understanding of the phenomena that cause the overvoltages and accurate calculation of energy discharged in the arresters. This paper investigates the mechanism that causes transient overvoltages, due to dc-side line-to-ground faults, in a cable-connected MMC–HVDC link. This paper also evaluates the impacts of various system parameters, for example, the prefault power transfer level, the MMC arm inductance, the transformer leakage inductance, the fault location, the ac system short-circuit capacity, the cable inductance, and the MMC blocking delay, on the peak transient overvoltages and the amount of energy discharge in the cable surge arresters. The transient overvoltage studies are also conducted when the dc cable is replaced by an overhead line. The studies are performed in the PSCAD platform. The studies conclude that the prefault power transfer level and the MMC blocking delay have the most significant impacts on the surge arrester energy discharge under dc-side line-to-ground faults in the MMC–HVDC link. The study results also provide essential information for the design of the MMC–HVDC link components.

In this paper, a static synchronous compensator (STATCOM) operation scheme of the modular multilevel converter adopting clamp double submodules (CDSM-MMC) during a pole-to-pole dc fault is proposed. By analyzing the current paths of a CDSM, this paper proposes three new operation states of the CDSM, which allow the CDSM to provide bipolar voltages when the arm current that flows through it is negative. Through control of the CDSMs to operate in these new states along with the blocked state, the arms of the CDSM-MMC can operate alternately in two different modes, that is, the conducting mode and blocked mode even when the dc-link voltage drops to zero. Thus, the CDSM-MMC is capable of working as a STATCOM during a dc fault. Moreover, a control strategy of the dc-fault STATCOM operation mode is designed to control the reactive power provided by the CDSM-MMC, to limit the dc fault current, and to reserve and balance the energies stored in capacitors. Simulation results conducted in PSCAD/EMTDC demonstrate the validity of the proposed STATCOM operation scheme and control strategy of the CDSM-MMC during a pole-to-pole dc fault.

This paper presents the results of a comparative analysis among three methods for attributing responsibilities of voltage unbalance in the electrical network, namely, the IEC, three-phase power flow, and conforming and nonconforming currents methods. To do this, computational and experimental simulations are carried out based on employing a system comprised of a perfectly transposed transmission line and two loads that are not always balanced. The cases selected for this study also permit the identification of the main advantages and disadvantages of each method under evaluation. This work is characterized as a topic of great interest to the regulatory agencies responsible for establishing regulations, and for utilities and consumers. It is relevant because by using the methods investigated in this paper, the identification of the sources, which generates unbalance, the division of responsibilities, and the application of penalties due to limit transgression will be possible.

This paper presents a simple modulation index-dependent Thevenin equivalent circuit model of the voltage-source converters (VSCs) and the associated anti-parallel diode rectifier. The model derivation uses orthogonal transformation and a transformer analogy to transfer impedances to the dc side of the converter. The paper proposes a switched equivalent representation of a three-phase diode rectifier formed due to the antiparallel diodes when VSC switching is inhibited. The proposed model is implemented in Alternate Transients Program–Electromagnetic Transients Program and simulation results are compared with the switched model. The proposed model is shown to under predict the fault currents for pole-ground (PG) faults and to provide a nearly exact match for pole-pole faults. The rectifier equivalent provides increased accuracy in the transient region with the inclusion of the source inductance. A multiterminal dc (MTDC) system model is used to study effects of aggregation which gives increased accuracy of the equivalent circuit. It is shown that the MTDC system simulation with the proposed Thevenin equivalent is about 3 times faster than full-switched circuit simulation for a time step of 5 $\mu$ s. The MTDC system simulation with the Thevenin equivalent circuit is still accurate and 126 times faster when the time step is increased to 750 $\mu$ s.

This paper deals with those measures against overvoltage taken when protecting buildings occupied by people or animals. Such overvoltages can appear in the case of ground fault if the grounding system of the building is located within a potential funnel of a transmission tower's grounding system. This paper presents an entire concept regarding protection against any potential rise within the building. Computations of electrical potential on the grounding system having been performed using data obtained by measurements. This includes equipotential bonding, overvoltage protection of the building, and a transformer substation with metal–oxide varistors and gas-discharge arresters installed on the low-voltage line. A simulation model for the entire overvoltage protection based on known mathematical models has been merged within this paper. Adequate overvoltage protection elements were selected for protection against potential rises in the grounding systems of buildings on the basis of computation results.

This paper presents the analysis of possible transient overvoltages that can be generated during vacuum circuit-breaker (VCB) operation at the connection point of a photovoltaic powerplant. A typical switching event that is related to the VCB concerns energization and de-energization of the unloaded transformer. However, at the typical photovoltaic powerplant, the transformer is fed by an inverter equipped with LC (or LCL) filters that are necessary for limitation of harmonics and ripple in voltage and current. From the point of view of switching operations, the inductance and capacitance of the low-voltage (LV)-side connected filter affect the natural frequency of the transformer, which is reflected by different transient system response during VCB operation. In this paper, laboratory measurement of overvoltages generated during distribution transformer switching by means of VCBs is analyzed. The impact of the LC filter connected at the LV side on the overvoltages suppression was studied. Electromagnetic Transients Program–Alternative Transients Program simulations were conducted in order to verify the possibility of transients elimination by means of the additional series-connected RL choke at the transformer medium-voltage side.

This paper highlights an immediate need for renewable generation developers, manufacturers, utilities and industry standardization institutes to jointly address a challenge of capacitor switching transient immunity in inverter-based wind or solar farms. The need was identified from wide-area operation records which showed partial or full generation loss in coincidence with routine utility-owned capacitor switching. The lack of a clear industry guideline has caused confusion to all involved parties. This paper recommends all entities to address this challenge collaboratively through enhanced equipment specification, design, and test before a well-defined industry guideline becomes available.

Having unveiled the problems associated with distance protection of lines connected to full-scale converter-interfaced renewable energy power plants (CIREPPs) due to remote and intermediate infeed in Part I of this paper, Part II describes why the countermeasures currently deployed by commercial relays to tackle these problems in a conventional power system fail in the presence of a CIREPP. Then, a new formula to calculate impedance is devised for the phase element of the relay to prevent maloperation in the event of line-to-line to ground faults. In addition, a communication-assisted method with minimal bandwidth requirement is proposed for balanced and line-to-line faults. The issues that should be taken into consideration for single-line-to-ground faults are also discussed. The solutions presented in this paper are simple and independent of the host system grid code, and can be applied to industrial memory-polarized distance relays. A comprehensive performance evaluation is presented to verify the reliable operation of the proposed methods.

This paper focuses on the parameter estimation of transmission lines which is an essential prerequisite for system studies and relay settings. Synchrophasor measurements have shown promising potentials for the transmission line parameter estimation. Majority of existing techniques entail existence of phasor measurement units (PMUs) at both ends of a given transmission line; however, this assumption rarely holds true in nowadays power networks with few installed PMUs. In this paper, a practical technique is proposed for the estimation of transmission line parameters while the required data are phasor measurements at one end of a given line and conventional magnitude measurements at the other end. The proposed method is thus on the basis of joint PMU and supervisory control and data acquisition (SCADA) measurements. A non-linear weighted least-square error (NWLSE) algorithm is employed for the maximum-likelihood estimation of parameters. The approach is initially devised for simple transmission lines with two terminals; then, it is extended for three-terminal lines. Numerical studies encompassing two- and three-terminal lines are conducted through software and hardware simulations. The results demonstrate the effectiveness of new technique and verify its applicability in present power networks.

A modular multilevel converter control system, based on converter energy storage, is proposed in this paper for two different control modes: active power and dc voltage. The proposed control system decouples the submodule (SM) capacitor voltages from the dc bus voltage. One of the practical applications is the management of active redundant SMs. A practical HVDC system with 401-level MMCs, including 10% redundancy in MMC SMs, is used for validating and demonstrating the advantages of the proposed control system. This paper also presents a novel capacitor voltage balancing control based on $\max$ – $\min$ functions. It is used to drastically reduce the number of switchings for each SM and enhances computational efficiency.

A new wind turbine transformers protection method against high du/dt switching transients taking place during a wind turbine operation is described. High-frequency transients, characterized by a high rate of voltage rise (du/dt) and overvoltage at the wind turbine transformer high-voltage terminals may be a result of operations of the turbine switchgear typically comprising a vacuum circuit breaker. The protection concept described in this paper is a follow up of research activities on a transients suppressing method dedicated for distribution transformers. The principle of operation of the protective device proposed and parameters optimization procedures for windmill transformers are provided in this paper. The performance of the device was verified by means of Alternative Transients Program–Electromagnetic Transients Program simulations. The physical device was built and experimentally verified by full-scale functional tests reflecting a real wind turbine power network.

This paper proposes a model predictive control (MPC) method for modular-multilevel-converter (MMC) high-voltage direct current (HVDC). To control the MMC-HVDC system properly, the ac current, circulating current, and submodule (SM) capacitor voltage are taken into consideration. The existing MPC methods for the MMC-HVDC system utilize weighting factors to configure the cost function in combinations of the SM capacitor voltage balancing algorithm, ac current control, and circulating current control. Because all combinations of the switch states are considered in order to minimize the cost function, their possible combinations increase geometrically according to the increase of the level of the MMC, which is a significant disadvantage. This paper proposes a new MPC method with a reduced number of states for ac current control, circulating current control, and the SM capacitor voltage-balancing algorithm. The proposed cost functions are divided into three types according to their control purposes. Each cost function determines the minimum number of states for controlling the ac current, circulating current, and SM capacitor voltage. The efficacy of the proposed controlling method is verified through simulation results using PSCAD/EMTDC.

A novel approach for the calculation of transient fault current contributions from capacitive network components in HVDC cable networks during pole-to-ground faults is presented in this paper. Analytic approximations considering the skin effect and the resulting distortion of the waveshape of the fault surge are proposed. Only fault current contributions from capacitive components, such as dc capacitors and adjacent cable feeders, are taken into account in this paper, since they are dominant during the first few milliseconds and yield the highest rates of rise of fault current. The results of the proposed expressions are compared with a benchmark model implemented in PSCAD and exhibit an accurate representation of the time development of the fault current. The derived approximations may serve as a starting point for a short-circuit calculation standard for HVDC networks and the specification of HVDC circuit-breaker requirements.

A 240-MVA, 165-GRDY/20-kV generator step-up transformer was insulated with Envirotemp FR3 fluid. The transformer was energized during a cold temperature period in mid-December 2010. This is believed to be the largest generator step-up transformer filled with a natural ester fluid that was energized at low temperatures. The lack of cold start experience of natural ester-filled power transformers created questions and at least some uncertainty as to the outcome. This paper describes the instrumentation used and data collected during the cold startup procedure. In addition, a computational fluid dynamics analysis of this transformer design was performed that provided calculated temperatures and elapsed times until steady-state temperatures were achieved. Comparisons were made between the actual and calculated values. The results indicate that the cold startup procedures for a power transformer are the same whether filled with mineral oil or natural ester fluid of similar characteristics as used for this paper.

This paper describes the application of the finite-elements method for determining the distribution of losses over windings and a detailed thermal-hydraulic network model for determining the value of the hot-spot factor and the hot-spot temperature. The influence of nonuniform losses in the conductor of the windings and the influence of nonuniform cooling are clearly separated. A detailed discussion of the components included in the hot-spot factor is given in this paper, enabling proper understanding of this complex and important quantity. The case study transformer was a 100-MVA oil power transformer with oil-directed water-forced cooling. The method was applied within the scope of the project of increasing the power of a hydro power station by 12%. The task was to calculate the temperatures and to recommend whether the reconstruction of the transformer is necessary. The developed calculation method and software were hitherto used for transformer design (in the production or the reconstruction stage of transformers) and it is the first application to give an answer as to whether it is possible to increase the power of an existing transformer. In this specific application, necessary onsite measurements for determining the characteristics of the coolers were performed.

Over the years, there have been a number of cases where users experienced higher noise levels of transformers on-site versus what was measured in the factory acceptance tests. In some of these cases, measurements have shown that a power transformer can have significantly different noise characteristics (both noise level and frequency spectrum) on-site compared to what was measured in the factory. This difference can range from a few decibels to more than 10 dB in some cases. Some of these differences are found to be caused by local environmental factors that influenced the on-site measurements. Others were caused by operating conditions on-site that are different from those of the factory tests. This paper presents results of investigations and measurements performed, over the past several years, on the main contributors to this difference. For each of these contributors, data are presented that demonstrate the impact of this contributor as well as quantify typical magnitudes of this impact. It is hoped that the material presented in this paper will assist users in: 1) understanding some of the pitfalls of on-site measurement of transformer noise; 2) how operating conditions contribute to the noise performance of transformers on-site; and 3) how significantly higher on-site noise levels can, in some cases, indicate an issue with the transformer or any of its components/accessories. It is also hoped that the material in this paper will contribute to industry standards/application guides on noise determination and abatement.

A typical maloperation case of the differential protection utilized in the ultra-high voltage (UHV) voltage-regulating transformer (VRT) under inrush conditions is investigated in this paper. It indicates that the second harmonic restraint principle cannot identify the inrush current reliably under such conditions. This paper proposes a normalized equivalent instantaneous inductance (NEII)-based inrush blocking method, which improves the equivalent instantaneous inductance (EII)-based method in the UHV VRT protection. Although the EII-based method is capable of discriminating inrush currents and internal faults, some factors constrain its application, such as the field limitation of measuring the terminal voltage of the transformer and the difficulty of determining a reasonable threshold due to the diversity of transformer magnetizing characteristics. The proposed NEII-based method overcomes limitations of the EII-based method and can be adopted in UHVVRT protection. A number of simulation tests have verified the effectiveness of the proposed method.

This paper proposes a technique for preventing saturation in series transformers from dynamic voltage restorer (DVR) systems. The method consists in correcting the voltages which are injected through the transformers into the power system to compensate voltage sags. It restricts the compensating voltages during the sag whenever it predicts that a maximum limit for the flux linkage is about to be exceeded. The prediction is carried out at the beginning of a stabilized voltage sag. Moreover, the technique allows a certain level of sag compensation even when the estimated flux is expected to exceed the saturation limit. The voltage sag level and phase are computed through an adaptive recursive least squares (RLS). The RLS estimation incorporates a transient period before it achieves a stable state whenever there is a sag event. The DVR is not supposed to operate in this period. Therefore, this paper also outlines a simple procedure for detecting the RLS estimation stable level. Simulation of different scenarios of voltage sags and the results from the experimental implementation of a DVR show the effectiveness of the method.

This paper presents a practical assessment and validation for the new proposed approach based on the continuous wavelet transform (CWT) to obtain the frequency response from online transient signals for an actual transformer. Apart from the mathematical procedure, an electronic system was also designed and implemented in order to inject superimposed controlled pulses to the power system (50/60 Hz) wave. The results show improved performance of the wavelet transform compared to the Fourier transform, for transient signal analysis to be applied on a nonintrusive transformer monitoring approach. Particularly, this new approach enables overcoming some problems related to the signal filtering and the signal processing in an online frequency-response analysis transformer diagnosis. This paper is presented in three parts. Part I showed the main theoretical basis of this approach. The current Part II shows a practical assessment based on tests performed in a three-phase transformer of 1150/345 V-5 kVA and in a single-phase transformer of 13200/240 V-15 kVA. Part III shows the performance of this new approach on transformers being diagnosed under real conditions.

This paper presents application guidelines for a new generic plant controller model that can be used for renewable plant dynamic system analysis that may include the control of static com- pensators (STATCOMs), static var compensators (SVCs), and/or energy storage devices. The authors originally presented the need for such a model to the Western Electricity Coordinating Council (WECC) and contributed to various stages of the new model development led by WECC. As Independent System Operators (ISOs)/Regional Reliability Organizations (RTOs)/Transmission Service Providers (TSPs) are moving toward acceptance and use of generic models only the authors have for the first time applied this new model for an operating renewable plant. The new model was configured, parameterized, and tuned to control two STATCOMs and multiple switched capacitors/reactors to represent a mas- ter/auxiliary control strategy as implemented in the plant. Dynamic simulations show that the configured and parameterized model is performing properly as expected. This tuned model was submitted to the system operator for compliance with the interconnection requirements of the plant. This is the first submission of the new generic model for this type of control for a renewable project. The submitted model is appropriate for dynamic simulations, which are routinely performed by ISOs/RTOs/TSPs in system impact stud- ies or transmission planning studies. This paper provides useful guidelines in applying this new model for similar projects.

An online approach based on application of search coil for transformer winding fault detection is presented in this paper. Earlier attempts report use of search coil for the detection of transformer interturn fault. This paper extends application of search coil for detection of winding displacement in addition to interturn fault identification. Experimental results based on specially designed 10-kVA transformer are presented to analyze search coil voltages during healthy and faulty conditions. The algorithm detects phase ( $U$ , $V$ , $W$ ), side (LV, HV), and precise location of the fault. Discrimination between interturn fault and winding radial displacement is also achieved by search coil voltage analysis. The results are experimentally verified based on impedance analyzer plots. Experimental analysis indicates that the proposed method is independent of load on the transformer, vector group, nature of winding, balanced or unbalanced source voltage, source voltage harmonics, inrush, and overexcitation conditions.

Room Temperature Vulcanized (RTV) silicone rubber is widely used on the surface of insulator of power transmission and transformation equipment because of its excellent hydrophobicity. With the increase of operating life, RTV coating is subject to the influence of high temperature, high humidity and complex outdoor environmental factors, which tends to fall off, seriously affecting the working state of the insulator. In this paper, based on the realization of insulator identification, the defect in RTV coating is identified, and the damage area percentage of RTV coating in the insulator shed is calculated. Experiments show that this method can effectively characterize the severity of defects in RTV coating, which provides a technical basis for the online detection and evaluation of RTV coating status.

A power transformer is one of the most crucial items of equipment in the electricity supply chain. The reliability of this valuable asset is strongly dependent on the condition of its subsystems such as insulation, core, windings, bushings and tap changer. Integration of various measured parameters of these subsystems makes it possible to evaluate the overall health condition of an in-service transformer. This paper develops an artificially intelligent algorithm based on multiple general regression neural networks to combine the operating condition of various subsystems of a transformer to form a quantitative health index. The model is developed using a training set derived from four conditional boundaries based on IEEE standards, the literature and the knowledge of transformer experts. Performance of the proposed method is compared with expert classifications using a database of 345 power transformers. This shows that the proposed method is reliable and effective for condition assessment and is sensitive to poor condition of any single subsystem.

The protective effect of a shield wire against direct lightning flashes to a buried power cable has been clarified in this paper. The finite-difference time-domain (FDTD) method is used in the investigation. The FDTD model considers a 1-km-long insulated cable protected by a shield wire, and a lightning strike right above them. The model has been validated by comparing its results with experimental results available in the literature. The computed results show that the presence of the shield wire can effectively prevent the cable from being directly struck by the lightning flash. The voltage across the cable insulation at the striking point depends on the ground resistivity (the higher the resistivity, the higher the voltage), but this dependence is not linear. It is also shown that the soil surface ionization is relevant to the current delivered to the shield wire and to the voltage developed across the cable insulation. On the other hand, the soil ionization around the shield wire has negligible effects. The intrinsic resistance of the shield wire does not influence significantly the peak voltage across the cable insulation, but it influences its duration (the higher the intrinsic resistance, the longer the voltage duration).

In order to improve features, such as fault current blocking and capacitor voltage balancing, modular multilevel converter (MMC) topologies incorporating multiport submodules (SMs) are being considered as candidates for HVdc transmission applications. This paper presents high-speed and accurate electromagnetic transient (EMT) models for MMCs composed of such multiport SMs. The approach uses the Schur's complement technique to recursively eliminate internal nodes of the converter structure to create a multiport Norton equivalent that connects to the external network. Thus, the final admittance matrix seen by the EMT solver has a dimension orders of magnitude smaller than that of the unreduced structure. As with previously developed approaches for MMCs with single-port SMs, all internal information, such as individual SM capacitor voltages, is preserved and can be output by the program if needed. This increases the bookkeeping effort, but the overall reduction in matrix size more than compensates for any resulting time penalty. Approximately two to three orders of magnitude speedup over a straightforward implementation in an EMT program is achieved.

This paper presents a comprehensive analysis and a novel solution for the discharge problem on the half-bridge submodules’ (HBSMs’) capacitors in the hybrid modular multilevel converter (MMC)-based high-voltage direct current transmission systems under a long-term (e.g., several minutes or longer) pole-to-ground fault condition. The mechanism of this issue is analyzed, and then a solution based on fundamental-frequency reactive circulating current (FFRCC) injection is proposed. The influence of FFRCC injection is analyzed in terms of the fluctuated arm capacitor energy and the current stress of semiconductors. Moreover, an enhanced arm voltage modulation method based on HBSMs and full-bridge submodules coordinate control is proposed to accelerate the capacitor voltage balance of HBSMs. Finally, the performance of the proposed control strategy is verified by the simulation results.

With the integration of the renewable energy-based distributed small power plants, whose generated energies are uncertain, in addition to the increasing usage of the charging stations for electric vehicles, the determination of the dynamic loading of distribution transformers appears to be crucial. Therefore, the issues of the model-based online monitoring of distribution transformers may achieve acceptance, if a cost–benefit compromise is reached. In this paper, a new thermal model is proposed to calculate the top-oil temperature of indoor distribution transformers using calculated tank temperature which accounts for convection, radiation, conduction, and ventilation heat transfer phenomena. The model is validated using measurements during the in-service operation of two indoor distribution transformers. Moreover, a method for calculating the loading capability of indoor distribution transformers without sacrificing the transformer's life is presented which can be easily adjusted for other distribution transformers. Furthermore, the results of the hardware implementing the proposed model and method for a model-based online assessment of the loading capability during the in-service operation of an indoor distribution transformer are presented.

The inductor-capacitor-inductor ( LCL ) dc/dc converter has been extensively studied for high power and stepping ratio because of elimination of internal transformer, lower footprint/weight, higher efficiency, and most importantly providing dc fault isolation from both dc sides. This paper presents a two-channel, two-layer controller including two inner current loops, which is symmetrical for each bridge of LCL dc/dc. The real-time implementation of the control scheme and its performance under normal conditions and during transient dc faults at both sides are studied on a 30 kW 200 V/900 V 1.7 kHz prototype. The prototype development is presented in some depth. The experimental results show that the converter with closed-loop control operates well at full power and under fast power reversal. Further dc fault testing concludes that there is no need for blocking since the internal voltage and current variables are within the rated values. Detailed study of converter losses is performed and results show that full power efficiency is around 93.4%.

This paper presents a novel online diagnostics method capable of detecting winding deformations in two-winding single-phase transformers. The main idea is to identify changes in the short-circuit impedance. The combination of 3-D Lissajous curve methods with a Butterworth low-pass filter allows for the accurate determination of winding deformation of large power transformers in real time. The method is very robust and capable of detecting deformations at the early stage even when the measurements are noisy. Only information already available to the differential protection relay is needed. The proposed diagnostics method has been validated with circuit and finite-element simulations plus a lab experiment. The results show that the proposed online diagnostics technique has the ability to identify winding deformation problems under severe conditions, such as nonsinusoidal input, nonlinear loading, and measurement noise. Under ideal conditions (no signal noise), the inductive identification error of the proposed online diagnostics method identifies the parameters with less than 0.09% error. When accepting a measurement noise of 1%, the error on the identification of inductance is less than 0.13%.

This paper proposes an evolved capacitor commutated converter (ECCC), embedded with antiparallel thyristors based dual-directional full-bridge module (APT-DFBM), to mitigate the commutation failure of line-commutated converter-based-high-voltage direct-current (LCC-HVdc) transmission and enhance the control flexibility and dynamic response of capacitor-commutated converter-based HVdc (CCC-HVdc). The operating modes and parameter selection approach of the APT-DFBMs are provided, and the coordinated control approach between the APT-DFBMs and converter valves is presented. Then, an ECCC-HVdc system, adopting an LCC station as the rectifier and an ECCC station as the inverter, is developed in PSCAD/EMTDC. The transient responses including commutation failure probability, voltage and current stress, and fault recovery performance are compared under ac fault conditions among the proposed ECCC-HVdc, LCC-HVdc, and CCC-HVdc. The results show that the proposed ECCC-HVdc can effectively reduce the commutation failure risks of LCC-HVdc, and enhance the control flexibility and improve the dynamic responses of CCC-HVdc.

This paper presents a new method based on the relay agent for single-phase earth fault protection in radial distribution systems with distributed generators. In order to get the most informative fault features to discriminate fault for this system, traditional protection schemes based on different fault features are analyzed. The selection of fault features is discussed. Moreover, the cluster center of historical fault data is calculated to analyze the space distribution of fault data for each feeder by applying a fuzzy clustering algorithm. The space relative distance between the online sample data and the cluster center and the minimum value of the space relative distance among all of the relays at the relay agent are obtained. Finally, the coordination strategy of the relay agent is proposed to discriminate the faulty feeder. The proposed protection scheme is evaluated in a radial distribution network with distributed generations using PSCAD simulation.

This paper presents a novel protection method for monopolar high-voltage direct current (HVDC) converters. This method uses the secondary and tertiary currents of the converter transformers and based on dc and harmonic contents of the currents decides for the security of the converter. The new protection method detects the location of the fault inside the converter and based on that issues a signal for single-phase disconnection in the ac system to eliminate the fault. It will be shown that the new method is not only faster compared to the conventional protection methods but also can be easily applied to the existing HVDC systems without imposing any significant requirement. The Hydro-Quebec HVDC test system is used to present the simulation results and to ensure the validity and functionality of the proposed protection method.

Grounding faults are common faults on dc transmission lines. Characteristics of the short-circuit current, such as amplitude and frequency, and its following distribution are very important for analyzing the influence of the fault. Based on the ±500 kV Yongfu dc transmission project in China, several artificial grounding short-circuit tests at the terminal or middle of the dc transmission line were carried out. This paper reports the measurement and analysis of the short-circuit current and its distribution in a dc system for the first time. The short-circuit current distribution includes currents in the ground wires of the fault tower and currents in ground wires, which are connected to the converter station grounding grid, such as the optical ground wire (OPGW) currents of ac outgoing lines, neutral point currents of different transformers, and the grounding electrode bus arrester current. In addition, the frequency components of the fault currents are analyzed. Based on the PSCAD/EMTDC software, the current distribution is simulated and discussed. The simulation results are in good agreement with the test results.