Managing Renewable Energy Generation by Better Predicting Output
- March 6, 2019
- 417 views
A common, and deserved, knock on many emerging renewable energy resources comes on their main weakness-- the inherent intermittency. When the wind isn't blowing, wind turbines aren't helping you. When it's cloudy or nighttime, solar panels aren't producing critical power. So when looking at the nameplate capacity of wind and solar installations, it's important to recognize how infrequently those installations will actually be generating that much power.
While the emerging technologies in energy storage are ramping up to try and fill in those gaps and allow excess solar power generated in the daylight be used to power critical resources at night, there's no doubt still a gap in the process. This intermittency would be one thing to handle if that was the only issue with it, but a critical part of the intermittency of renewables comes from the apparent unpredictability of that intermittency. If weather unexpectedly adjusts from the forecast and the day is cloudier than anticipated, solar power will be less effective and more energy is needed to power the grid from baseload sources like gas, nuclear, and coal. Recognizing that the worst thing the grid can be is unreliable, energy markets will typically not count on the exact renewable output that will probably come just to avoid not having the necessary amount of electricity on hand-- adding to issues of overgeneration and inefficient energy management.
One way to reduce the negative effects of this unpredictable intermittency, though, is to reduce its unpredictability. This might seem like a difficult challenge to tackle, but I came across two recent stories of organizations doing just that:
One of the first applications of DeepMind at Google was to reduce and now control power usage at its data centers. Google is now leveraging machine learning to make more efficient wind farms by predicting wind power output 36 hours in advance.
The two Alphabet divisions worked together to train a neural network on weather forecasts and historical turbine data. A DeepMind system was then able to “predict wind power output 36 hours ahead of actual generation.”
By reducing the variability of wind power, the renewable energy source becomes “sufficiently more predictable and valuable,” giving wind farm operators more data-driven assessments of how to meet upcoming electricity demand.
The Sun is becoming an increasingly important source of clean electricity. Accurate sunlight forecasts being developed by A*STAR researchers could greatly improve the performance of solar energy plants, making it a viable alternative to carbon-based sources of power.
A photovoltaic power plant can cover up to 50 square kilometers of the Earth's surface and can generate up to a billion Watts of electricity. This capacity depends on the amount of sunlight at that location, so the ability to predict solar irradiance is crucial for knowing how much power the plant will contribute to the grid on any particular day.
"Forecasting is a key step in integrating renewable energy into the electricity grid," says Dazhi Yang from A*STAR's Singapore Institute of Manufacturing Technology (SIMTech). "It is an emerging subject that requires a wide spectrum of cross-disciplinary knowledge, such as statistics, data science, or machine learning."
Yang, together with Hao Quan from the A*STAR Experimental Power Grid Centre and colleagues from the University of Tennessee at Chattanooga and the National University of Singapore, has developed a numerical approach to weather prediction that efficiently combines multiple datasets to improve the accuracy of solar irradiation forecasts.