Supraharmonics in transmission and distribution networks

September 4th, 2017, Published in Articles: Energize

Standards for harmonics in power networks (transmission and distribution) usually deal with the range below 2 kHz, as most of the harmonics due to waveform distortion fall into this range. Harmonics or disturbances in the range above 2 kHz are becoming noticeable in networks due to the introduction of inverter based distributed generators, and are a cause of concern because of interaction with other devices, such as power line communications (PLC) connected to networks.

 The term “supraharmonics” (SH) is used to refer to any type of waveform distortion of voltage and current in the frequency range between 2 and 150 kHz. This is not an official definition adopted by any organisation, nor is there general agreement about the term. There is no abrupt change in phenomena at 2 or 150 kHz, so that the boundaries can be seen as being arbitrary. The same frequency range is also used in several European countries by network operators for PLC applications, which further increase the challenges to ensure EMC in this frequency range.

This family of disturbances is becoming an increasing concern in the industry, especially with the growth of distributed and embedded generation using inverters.  The large number of distributed renewable energy sources with their fluctuating power infeed can have an increasingly negative influence on the electricity-supply system. High-frequency emissions in future grids and the impact on connected consumers are expected to increase [1].

The subject is relatively new and knowledge of the phenomenon is limited, although several studies are ongoing and standards are still in the process of production. For example IEC standards for limiting values of supraharmonics in low-voltage power systems are expected to be published in the near future [1]. SH can highly affect neighbouring devices and influence them. The sensitive devices include smart meter gateways, led lighting, dimmers for lamps, and PLC communication. This can also be the cause of failures of touch-controlled operator elements [2].

Consequences of disturbances in the frequency range from 2 to 150 kHz

SH are receiving attention because of the possible impact on other devices connected to the network [2]. The following list identifies some of the problems experienced:

 Increasing capacitive currents that can damage the power supply, increase the neutral current and thus increase the safety risk

  • Failures in touch-controlled operator elements and dimmers for lamps
  • Reduction in the service life of LED lamps
  • Communication problems (for example, PLC communications)
  • Overheating of capacitor banks and transformers
  • Failures in protection devices

The effect of SH on other network components and on network control and management devices is also under study. SH are known to cause instability in weak networks with PV inverters, with the consequent spurious tripping of the inverters.

Source of supraharmonics

Broadly speaking, these HF components are due to the normal operation of electronic converters used in inverters and the switching techniques employed. The well known low frequency harmonics result from the nonlinear behaviour of loads.

Although referred to as harmonics, the SH disturbances are not caused by distortion of the fundamental frequency waveform, but are due to the switching of the inverter output circuits (although this does cause distortion of the waveform). The presence of inverter circuits in distribution networks can also increase the level of supraharmonics originating from other devices, ie act as both a source and a sink.

The generation of supraharmonics is related to the shift from grid-commutated to self-commutated power-electronic converters. The development of power electronics (PE) has been mainly driven to improve the voltage and current-handling capability and the switching speed of power semiconductor devices.

Nonetheless, it is hard to connect a single power semiconductor switch directly to medium voltage grids. The series connection of standard low-voltage switching devices enables synthesising of a medium voltage output, while the individual power semiconductors need to withstand only part of the voltage.

The addition of several low voltage cells per arm provides high scalability, leading to reduced cost and volume of the entire solution. Moreover, it allows a more creative use of these additional switches in novel modulation strategies, which enable to enhance the quality of output voltages and input currents, resulting in what is known as multilevel converter (MC) technology.

All of these developments use switching of power electronic devices at frequencies well above the fundamental grid frequency, which results in step discontinuities in the current flow manifesting as supraharmonics.

PV inverters

PV inverters are of particular interest because of the large number of embedded and small rooftop systems in distribution networks. The switching frequency of a PV inverter often lies below 20 kHz. Smaller single phase inverters often lie in the upper range and bigger three phase inverters commonly below 5 kHz [3]. Fig. 1 shows the measured harmonics from a typical inverter.

Fig. 1: Harmonic spectrum from a typical inverter [5].

Characteristics of supraharmonics.

The effort to increase the power factor and decrease the harmonic content in the lower-frequency range of the output current of inverters used in grid connected devices, has led to an increase of the emission in the supraharmonics range [3].  The SH in this case originate from the switching circuits in the inverter and will be injected into the grid as long as the inverter is operating [4].  When the inverter is not operating or producing output the device can become a sink for SH.

 Primary and secondary SH

Devices using inverters as output can be both a source and a sink of supraharmonics. There are two driving forces for the currents at the interface between the inverter and the grid. The resulting currents are referred to as “primary emission” and “secondary emission” [4]. The primary emission is the part of the harmonic or supra-harmonic current driven by power electronic or other sources inside of the device or installation (driven by ?1 in Fig. 1-2). The secondary emission is the part driven by sources outside of the device or installation (driven by ?2 in Fig. 1-2). The latter one plays a much bigger role for supraharmonics than for (low-frequency) harmonics. [1]

Fig. 2: Model for the connection of a device or installation to the rest of the power system [1].

Transmission of supraharmonics through the network

Emission in the supraharmonic frequency range differs from low frequency harmonic emission in several ways. One of the main differences is the propagation of the emissions. Harmonic currents propagate into the grid while supra-harmonic currents tend to stay within the installation and propagate to a high extent towards neighbouring equipment. The reason for this is the low impedance offered by many connected devices. The interface between the device and the grid is in many cases an EMC-filter with a capacitor connected between phase and neutral and hence the impedance will decrease with increasing frequency.

In the supraharmonic frequency range the interaction between devices can be substantial.  As many small photovoltaic installations are close together, the interaction between the inverter and other devices becomes an important aspect to consider [4]. In the case of isolated utility scale PV and wind generators connected to the grid, transmission of SH will depend on the harmonic impedance of the network.

 Harmonic impedance

The impedance of a transmission or distribution network will vary with frequency, and hence for evaluation of the effect of SH, the network impedance at the SH frequency needs to be known.

Fig. 3: Impedance vs. frequency for urban power lines [5].

Harmonic impedance is an important parameter used to analyse the stability of power system, when analysing the impact of high frequency harmonics. Harmonics are usually injected into the network as current harmonics, and are converted to voltage harmonics by the impedance of the network.  Because transmission line impedance is dominantly inductive in nature, the line impedance increases with frequency and high frequency harmonic currents can result in significant voltage harmonics (Fig. 3). Transmission and distribution lines also exhibit resonant frequency points, some of which may coincide with SH frequencies.

Measurement or calculation of harmonic impedance is complicated and requires careful planning. For compliance testing in South Africa, an IPP can assume that the network harmonic impedance at the point of connection (POC) will be less than three times the base harmonic impedance for the range of reference fault levels at the POC, i.e. the network harmonic impedance shall not exceed a harmonic impedance of:

Z h= (3·Vll2/Ssc )·h

Where:

h is the harmonic number, Vll is the nominal voltage in kV, and Ssc is the fault level in MVA.

PV inverters can have a significant impact on harmonic network impedance. In larger PV installations the grid-side filter circuits can cause significant resonances at low frequencies. The resonant frequency decreases with increasing number of inverters. Consequently, input impedances of PV inverters should be considered in harmonic network impedance studies [3]. In a distribution network, the harmonic impedance can also be influenced by the loads connected to the network.

Measurement of supraharmonics

For utility PV and wind energy systems, compliance to grid codes is required before connection can be granted. This involves measurement of parameters including harmonics, and in future, supraharmonics.

In a network where a large number of PV systems using inverters are connected to the network, the output filter of the inverter offers a low impedance to supraharmonics and can also affect the flow of high frequency harmonics in the network. Resonance can also result in situation where a large number of PV installations are installed close together. Harmonic currents will flow into the inverter if it is not producing current. For this reason it is difficult to determine SH levels using a single measurement, and time based measurements must be used. SH could be time variant and intermittent and measurement can be difficult. Measurement over a period with a recording type instrument is often necessary.

Many systems rely on actual measurements at site using existing voltage and current transformers. This can be problematic as these transformers are designed for accuracy at grid frequency and can exhibit inaccuracy at SH frequencies [7].

References

[1] J Smith, et al: “Power quality aspects of solar power”, Cigre TB 672.

[2] Siemens: “Line conducted emissions in the range 2kHz to 150 kHz”, application note PQ application HF 200

[3] A Mureno-Muratz, et al: “Ongoing work in Cigre working groups on supraharmonics from power electronic converters”, CIRED 23rd International Conference on Electricity Distribution , Lyon, June 2015.

[4] AM Blanco: “Some European experiences with renewable installations”, SAIEE workshop on the grid integration of renewable energy: Grid code compliance assessment, Johannesburg, August 2017.

[5] H Cavdar and E Karadenitz: “Measurements of Impedance and Attenuation at CENELEC

Bands for Power Line Communications Systems”, Sensors 2008, 8, 8027-8036.

[6] I Angulo, et al: “Study of Unwanted Emissions in the CENELEC-A Band Generated by Distributed Energy Resources and Their Influence over Narrow Band Power Line Communications”, Energies, 2016.

[7] J Meyer: “Measurement accuracy due to VTs and CTs”, SAIEE workshop on the grid integration of renewable energy: Grid code compliance assessment, Johannesburg, August 2017.

Send your comments to energize@ee.co.za

 

 

Subscribe to our leading email newsletters

FREE-OF-CHARGE

CLICK for other EE Publishers information products