The goal of Miao He’s research, funded by the National Science Foundation and the Energy Reliability Council of Texas, is to mitigate power loss from sudden wind changes.
Wind always seems to be the one thing, no matter the season, the South Plains has in abundance. Because of that, wind turbines rise prominently above the horizon all across West Texas, making the region one of the leading producers of wind power.
But while the presence of wind is consistent, the speed and direction of those winds seem to change quickly, especially when weather-related events roll through. Those occurrences, called wind ramp events, not only affect people and property on the ground, but also the efficiency of those wind turbines.
It is that efficiency that Miao He, an assistant professor in the Texas Tech University Department of Electrical and Computer Engineering, is studying in an effort to maintain wind power reliability when sudden changes in wind speeds and direction occur.
“My research focus is on how to predict changes in wind 10 to 30 minutes ahead of that change so that wind farm operators and power system operators can be prepared for these events,” He said. “There are a lot of signatures we can use from real-time measurements from wind farms to tell when this wind ramp event is going to happen and how significant a drop or increase it will be. It's sort of prediction quantification.”
To help with his research, He received two large grants totaling more than $618,000. He received $318,610 from the Electric Reliability Council of Texas (ERCOT), and $299,960 from the National Science Foundation (NSF).
He will develop data analytics tools to measure sensory data from wind farms, focusing on developing algorithms and software tools to detect wind ramp events as well as extracting real-time data information to help wind farm operators determine how much to increase or decrease power output from wind turbines depending on the disruptiveness of the wind ramp event.
Wind Farm Power Output
West Texas has become a significant producer of wind power. But the turbines are also greatly affected by changes in weather or wind ramp events.
“In the NSF-funded project we focus on one of the dominant events that wind ramp events are caused by, which are weather fronts,” He said. “Weather fronts are events that, ahead of them and behind the fronts, wind speeds and weather conditions are totally different. They can cause a sudden change of wind power. Even extreme weather conditions like icing can cause wind power ramp events on wind farms.”
Maintaining the level of output from wind turbines in the face of wind changes is critical to the reliability of regional power grids. That makes measuring power reduction or increase from wind ramp events paramount to maintaining that output.
He's research will involve utilizing sensors on wind turbines able to measure the changes in weather and wind. Together with data that show how much the event in question reduces or increases power output from wind turbines, He hopes to develop an algorithm that can tell wind farm operators how much to increase or decrease turbine output on one end of the farm to compensate for the power loss or increase on the other end.
Using signal processing tools and data mining technology, He said the first step is to discover the signature of a specific wind ramp event, such as its direction and magnitude. Then the goal is to determine which region of the wind farm will first incur the wind power reduction, followed by technology to estimate how fast the event will move through the wind farm.
“That information will be critical in determining in the next 10 minutes, which area the front will cover and how much reduction will be induced,” He said. “We need to detect these events as early as possible and quantify them so that the power system generator and the wind farm generator can adjust accordingly.”
Wind ramp events, though, are not the only reasons for power reduction or increase with wind farms.
Another factor is what is called sub-synchronous interaction, which involves power loss in transmission from wind farms to power stations. Because most large wind turbine farms are built in remote, rural areas, they require lengthy transmission lines to reach power stations, and there is a subsequent amount of energy that is lost in that transmission.
He said a fundamental technology to reduce the amount of power lost is to deploy capacitors at different locations along the line so that the aggregate transmission is near capacity. However, motors in wind turbines can appear inductive, and this can cause oscillation of current in the transmission lines which limits power transfer.
One goal of the study funded by NSF is to find a way to reduce this oscillation and subsequent loss of power. He said the amount of power lost is not large, but it is enough to warrant studies on ways of reducing it.
“It is a fairly new event and people have been looking at how to mitigate these events,” He said. “We need to mitigate these events so more power can be transferred from wind farms through long transmission lines.”
He is hopeful the project can be finished and the detection algorithms can be tested and developed and in the marketplace within three to five years. The ultimate objective of the study, He said, is to build an online data analytics tool to integrate into the control rooms of wind farm operators and power system operators to aid in decision making during wind ramp events.
“The goal of the project is to deliver the algorithm and support tools for power system operators and wind farm operators to handle these wind power events,” He said.