A whole new ball game: forecasting renewable energy
The SA energy sector faces pressing challenges, and needs to act with urgency. Policy commitments to a net-zero future, such as the Paris Agreement, mean the transformation to a lowcarbon economy must come at a pace that can only be achieved by using technology.
Disruption to the electricity sector is on the cards as governments worldwide ramp up renewables and transition away from fossil fuels. While renewable energy looks set to flourish, its intermittency means solutions will need to be found to keep grids stable. In addition, the industry is changing from a market based on commodity pricing to one based on technology solutions for integrating renewable energy.
As the energy industry continues to use more variable generation sources, accurate forecasts of power generation and net load are becoming essential to maintain system reliability, minimise carbon emissions and maximise renewable energy resources.
As we move into the fourth industrial revolution, grid operators, developers and consumers are harnessing artificial intelligence (AI), paving a path for a smooth transition to greater use of renewables. AI’s ability to provide better prediction will improve demand forecasting and asset management, while its automation capability drives operational excellence
— leading to competitive advantage and cost savings.
Supported by other emerging technologies, such as the internet of things (IoT), sensors, big data and distributed ledger technology, AI can unlock the vast potential of renewables while failure to embrace it would leave the renewable energy sector falling behind.
AI is far superior to humans when it comes to complex tasks at speed. Given that an energy grid is one of the most complex machines ever built and requires splitsecond decisions to be made in real-time, AI algorithms are a perfect fit.
As an increasing amount of megawatts feeds into the grid from variable renewable energy sources, predicting capacity levels has become paramount to secure a stable and efficient grid. Because renewables take up a greater share of the grid, there is a loss of base load generation from sources such as coal, which provide grid inertia via the presence of heavy rotating equipment.
Without grid inertia, power networks are unstable and susceptible to blackouts. Now, with the application of sensor technology, solar and wind generation can provide an enormous amount of realtime data, allowing AI to predict capacity levels.
Before harnessing AI, most forecasting techniques relied on individual weather models that offered a narrow view of the variables affecting the availability of renewable energy. Now, AI programs have been developed — such as IBM’s for the US department of energy’s SunShot Initiative — that combine self-learning weather models, sets of historical weather data, realtime measurement from local weather stations, sensor networks and cloud information derived from satellite imagery and sky cameras.
The result has been a 30% improvement in accuracy in solar forecasting, leading to gains on multiple fronts. Improved solar forecasts lowered operational electricity generation costs, start and shutdown costs of conventional generators, and solar power curtailment.
Forecasts of the base variables, such as wind speed, and the resulting power output, allow for a view on a range of time horizons, from minutes to hours ahead. This helps maintain grid stability to day-ahead (optimising plant availability) and several days ahead (scheduling maintenance).
With larger data sets becoming available predictions can now go far beyond the weather to train algorithms to predict more remarkable outcomes: for instance, how much extra power is used during a festive holiday, a large-scale international event, or how much altitude affects a community’s energy use.
For generators, distributors and energy traders, more accurate forecasting of variable renewable energy at shorter timescales allows better forecasting of capacity and output requirements, and bidding in the wholesale and balancing markets.
The earlier and more accurate the predictions, the more efficient it is for energy traders to rebalance position. AI provides a way of dealing with many more sites using more granular and diverse data than historical methods.
Meanwhile, for grid operators AI algorithms with vast amounts of weather data can ensure optimal use of power grids by adapting operations to the weather conditions at any time. More accurate short-term forecasting can result in better unit commitment and increased dispatch efficiency, thereby improving reliability and reducing required operating reserves.
With AI it is possible to predict more accurately what renewables are likely to do, creating greater control over other power plants such as coal plants, which take many hours to ramp up.