Conference Agenda

Some petrochemical plants (low density polyethylene type) always require the use of big synchronous motors to feed large power compressors (typically from 10 to 30 MW).
In certain supply network conditions, it could happen that the motor undergoes a critical starting which takes several tens of seconds to be completed. “Critical” means that the starting time is higher than the time the rotor can withstand in blocked condition: therefore, it is necessary to assess the actual motor rotation to be sure that the motor is being self-ventilating during its acceleration and can survive the critical starting. By means of numerical simulations, it is demonstrated that the solution of using an under-impedance relay (ANSI code 21) for starting supervision, as is proposed in technical literature since many years, gives some troubles for the correct and safe protection of the starting of the motor, with respect to the most common solution of adopting a supervision switch relay (ANSI code 14) based on rotor speed measurement.

Due to the vast number of substations at the distribution level and increased costs of differential busbar protection, DSOs are in search of cost-effective protection schemes for busbar protection. This includes the use of various communication-based protection schemes, such as the reverse-blocking schemes used at Stedin. However, due to impedance grounding, the single-phase-to-ground short circuit currents have small values in medium voltage impedance-earthed distribution grids. As a result, the reverse-blocking scheme fails to detect this type of fault. This paper introduces a novel distributed protection scheme based on the detection of zero-sequence components of the currents and voltages and the negative-sequence current component. The proposed scheme successfully detects single-phase-to-ground busbar faults by using the standard settings of the widely available overcurrent IEDs, and an IEC 61850 communication between them. Firstly, the detection of the zero- and negative-sequence current components is used to distinguish between a busbar and a feeder fault. Secondly, zero-sequence voltage detection is used to distinguish between the faulty and healthy sections of the busbar when the busbar coupler is opened. This also increases the proposed scheme’s reliability by avoiding miss-operation due to human errors during maintenance or testing. The grid is modeled in a Real Time Digital Simulator (RTDS), and a Hardware-in-the-Loop (HiL) simulation is carried out to test the protection scheme. The extensive simulations show the strengths and the limitations of the proposed scheme. Based on the research results, the developed protection scheme is implemented as a standard protection scheme in all of Stedin’s new distribution substations.

With increased integration of renewable energy resources, FACTs devices and series compensation, sub-synchronous oscillations (SSO) have become more common in electrical power systems in recent years. Specially designed relaying devices are often employed to detect and isolate harmful SSO conditions as when unconstrained, they can lead to widespread equipment damage and system instability. This paper presents design and implementation of a SSO relay model that can effectively extract sub-synchronous components in system measurements to quickly detect SSO conditions. Dependability and security of the developed relay model are validated using electro-magnetic transient (EMT) type simulations. In addition, performance of the developed relay model is compared against a commercial (physical) SSO relay. Obtained results demonstrate the effectiveness of the implemented relay model in detecting SSO and protecting electrical power systems against SSO conditions.

Unearthed neutral is often used in industrial networks, which require continuous power supply, and indistribution networks where the network capacitive current is not too high; or it is uneconomical to move to compensated grounding. Unearthed networks can remain in operation during an earth-fault, but fast determination of the faulty line is the key for prevention of further fault escalation. Signal injection is one of the fault location methods often used in LV unearthed networks. The possibility of application of this method in MV networks isdependent on how to inject the signal into unearthed phases with voltages ranging from 10 to 35 kV. This paper presents the signal injection circuit, which consist of three inductive voltage transformers (IVTs). The specifics of IVT design for signal injection are discussed. The application of the LF signal injection during an earth-fault was simulated on the model of an unearthed distribution network in EMTP. An intermittent fault and ring-type network connection were considered as possible cases.

This paper presents a two-terminal traveling-wave-based protection algorithm applied to non-homogeneous transmission lines comprising any number of sections with different topologies and considering the effect of the sampling rate. Existing two-terminal traveling wave protection functions cannot protect the line under close-in faults and present limitations in non-homogeneous transmission lines. However, the effects of the sampling rate, considered in the proposed method, result in well-defined protection and unprotected zones, essential for protection security and coordination to deal with the issue of close-in faults in non-homogeneous transmission lines. Furthermore, the algorithm can accurately detect the faulted section, allowing its use in advanced protection functions such as adaptive automatic auto-reclosing and high-speed protection schemes. A protection device was modeled considering a sampling frequency equal to 1 MHz, including functions to detect traveling waves via wavelet transform, and the proposed protection algorithm to discriminate line internal from external faults, and to select the faulted section. The algorithm was evaluated using a large number of ATP fault simulations. The results show that the algorithm is robust and reliable for protection devices installed in non-homogeneous lines.

In this paper, the performances of fixed and variable window size phasor-based techniques reported in the literature are evaluated on a classical one-terminal impedance-based fault locator, pointing out the challenges and alternatives for applications in transmission systems equipped with high-speed protection functions. To do so, several Alternative Transients Program (ATP) fault simulations were carried out in a 230 kV/60 Hz power network, varying their parameters such as type, location,
inception angle and resistance. The obtained results attest that phasors estimated from variable window size strategies may improve the fault locator performance, even in the first cycles after
the disturbance occurrence, appearing themselves as potential alternatives to be applied in combination with high-speed phasorbased protection routines.