Researchers have established a technique that will help ensure a stable supply of electricity even as new renewable energy sources come on line. The trick is helping all of the subsystems to work in concert.
You definitely want to have a stable supply of electricity, because without it, all your gadgets could stop working, or even worse, if the variation in voltage and frequency is too large.
But delivering a stable supply of electricity is getting more and more difficult, especially as households begin to contribute with electricity from solar panels and small wind turbines.
One of the most important aspects of ensuring the reliable operation in any electric grid is getting correct measurements and information from the system. Those measurements affect everything else.
In practice, an entirely stable frequency is nearly impossible. No system is perfect, and the frequency of the electricity supply is actually constantly changing.
The challenges posed by these variations have led researchers at the Norwegian University of Science and Technology (NTNU) to use a new method to analyse this phenomenon in electric power systems.
- You might also like: A major research effort into tomorrow’s digital power grids
More precise descriptions
The NTNU researchers’ method can provide a more precise description of variations in the voltage frequency of the electric grid. These variations may oscillate around a mean value, but the lengths of their periods may differ. If these differences are not accounted for, they can, in the worst case, result in a system collapse.
To analyse and manage a power grid, you have to have precise measurements and represent them correctly. Everything else is based on those measurements and their interpretation being accurate.
“The new analysis method for time-varying frequency has big advantages over methods used previously and may have significant impact,” said Olav B. Fosso, a professor in the Department of Electric Power Engineering.
- You might also like: Helping today’s power companies anticipate tomorrow
Assumptions are inadequate
Today, the most widely used methods for analysing periodic variations in grid voltage measurements are based on the Fourier series. This approach relies on assumptions that are not always met in practice, but that have worked well enough with the structure of the electrical grid up to now.
However, the extensive use of power electronics with inverters that has blossomed in recent years due to the increased penetration of renewable energy sources poses challenges for both small and large electricity networks.
Some large wind farms, for example, have been shut down for long periods because the control and regulation subsystems haven’t worked together properly. But finding the reasons behind these kinds of problems has thus far been difficult.
The strength of the new analysis method is that it can more reliably identify different parts of the grid voltage frequency, so that they can be factored into the design of control systems.
- You might also like: Major climate benefits from shifting to renewable energy
Ensuring reliable operation of microgrids
Frequency time variations are more significant in determining the reliable operation of small systems, such as solar panels on your house or windmills that constitute a microgrid.
The power supply in many parts of the world operates at a frequency of 50 Hz, and is considered stable within a small range of variations. All electrical household appliances depend on the quality of the voltage supply, and on frequency variations being within the norms. If the quality isn’t good enough, the appliances either run less efficiently than they should, or simply stop working.
Small electrical systems such as rooftop solar panels and wind turbines are now a familiar sight. In the future, the norm may be that every household generates and uses its own power or contributes electricity to a larger network. Under this future scenario, it will be essential for each of these small players to deliver power that meets quality standards.
- You might also like: Two hundred and seventy-eight wind turbines won’t cut it
Collaboration with Norden Huang
The NTNU researchers used the Hilbert-Huang Transform (HHT) for detecting frequency-related phenomena to interpret their voltage measurements.
The method was originally developed to analyse ocean waves, but has since become much more widely used, including in the measurement of brain signals. This is the first time HHT has been used to analyse the frequency variations of the voltage in electrical power systems.
The NTNU researchers are working with Professor Norden Huang at National Central University in Taiwan, for whom HHT was named.
Professor Huang has been responsible for the Data Analysis unit of NASA Goddard Institute for Space Studies and today collaborates with Harvard, Johns Hopkins, MIT and Oxford universities. He also works with researchers and students from NTNU, including Geir Kulia, a master’s student in the Department of Electronics and Telecommunications.
- You might also like: Prime-time charging of electric cars could be a problem
Data from Bhutan
Kulia is currently writing his master’s thesis, supervised by Lars Lundheim and Marta Molinas from the Department of Engineering Cybernetics. His thesis is part of an MSc programme called Knowledge for Humanitarian Purposes that is offered by NTNU.
Kulia adopted the HHT method to analyse data from a microgrid in Bhutan.
To do this, Kulia developed an algorithm based on Huang’s HHT method. He decided to do this because he hadn’t been able to identify the frequency-related variations in his voltage measurements when he used traditional approaches. His analysis made no sense.
“At least they made no sense to me,” Kulia said, an observation confirmed by Molinas.
But the Hilbert-Huang Transform provided a completely different picture.
Identifying errors rather than masking them
The method proved far more suited for identifying possible problems than existing approaches. Suddenly Kulia had some clues to go on.
“In this case, we suspect that the problem lies in the control unit itself,” says
The controller is the part of the generation system that coordinates the rest of the power system based on measurement data.
If the controller design does not respond properly to the measurements, it may try to compensate for frequency components that do not really exist, but that are a result of how the algorithm analyses the data.
Then the part of the network that is supposed to regulate variations actually introduces new oscillations or fluctuations in voltage and frequency. This is not something you want happening in your electrical system.
It can also have serious consequences for the industry. Manufacturers of control systems don’t want to hear that a control unit is creating additional problems.
Method is drawing attention
The new method is also able to detect other phenomena. The new HHT method appears to be far better suited for analysing phenomena related to frequency variations than many traditional methods. It effectively detects the frequency components of the measurement data, which is essential in designing control systems.
One article about the findings, with Kulia as first author, has already been published and another has been accepted for publication. One of the reviewers of the soon-to-be published paper noted that “as solid state control equipment continues to expand, traditional ‘fundamental frequency’ analysis is no longer sufficient, especially in microgrids.”
Instantaneous Frequency in Electric Power Systems. Non-stationary signal analysis in physical (including man-made) and biological systems. ResearchGate.
Towards a Real-time Measurement Platform for Microgrids in Isolated Communities. Geir Kulia; Marta Molinas, Lars Lundheim, Bjørn B. Larsen, in Humanitarian Technology: Science, Systems and Global Impact 2016, HumTech2016, Boston, USA
Tool for detecting waveform distortions in inverter-based Microgrids: a validation study. Geir Kulia, Marta Molinas, Lars Lundheim, in IEEE Global Humanitarion Technology Conference 2016, Technology for the Benefit of Humanity, Seattle, USA