Smart Learning Software Predicts Renewable Energy Output with 90% Accuracy

When it comes to easing renewable energy into power grids, smart grid technologies that can smooth out the inconsistencies in solar and wind power are crucial. Smart meters, large-scale energy storage and energy management and demand response software all help, but energy forecasting has started to be a major focus. If utilities and other grid operators can know in advance what the power demand versus power supply from renewables will be, they can fully stabilize the grid.

Siemens Corporate Technology has taken on this challenge and created a new self-learning software that can predict the energy output from renewable sources over a 72-hour period with 90 percent accuracy. The software was based on an artificial neural network, which is modeled after how the human brain works. Using historical data, the algorithm can make increasingly accurate predictions.

Siemens has developed neural networks in order to calculate the behavior of highly complex systems, like renewable energy projects and even stock markets. Siemens says the system’s learning capability "makes it particularly suitable for adjusting grid operation to the fluctuating power outputs associated with renewable energy sources".

The new software was developed at the request of Swissgrid in Switzerland, which sees a large amount of energy transmitted through on the way to other countries in Europe. It needed a tool for predicting energy lost as it made its way through power lines so that it could order the right amount energy to make up for it.

According to the press release, "The new algorithm developed by Siemens researchers derives the projected transfer losses directly from electricity consumption forecasts. Along with data from the past, the system also uses variables such as current load flows, power generation figures for renewable sources, weather data, and water levels in pumped-storage hydroelectric power stations."

While more accurately predicting energy consumption has already saved Swissgrid 200,000 euros per year, the benefits for the environment could be much greater. With smart software like this, utilities can feel more confident about connecting renewable energy projects to the grid, which will help more clean energy projects be developed and more clean energy make its way to out in the world.

Smart Learning Software Predicts Renewable Energy Output with 90% Accuracy
Software based on the brain's neural networks can predict energy supply from renewable sources and energy demand to stabilize power grids.

Related Content on