Energy forecasts for net demand are crucial for participants of competitive markets, such as the Texas Electric Reliability Council (ERCOT) and similar markets, for several key reasons. For example, precise prognosis helps to predict when the imbalance of supply and demand will create jumps or demolition of prices, allowing traders and generators to optimize their license strategies. It is also important for the optimization of property. Electricity generators need to know when resources to market and price levels need to be committed. Poor prediction can lead to missed profit or operational assets when prices do not cover costs. Ercot region, specifically, has a huge wind and solar capacity. Net demand forecasts (total demand minus renewable generation) help predict when a conventional generation will be required to fill in the gaps from changing renewable resources. Market participants also use forecasts as a risk management tool. The exact projections allow participants to protect their positions through bilateral contracts or financial instruments, protecting against unstable market conditions. In the meantime, forecasts can provide an insight for operating planning. Having market predictions up to 15 days can help managers with the decisions on the obligations of the unit, the maintenance schedule and the distribution of resources in the generation property portfolio. In Texas, a competitive market design only for energy puts even more importance on prediction, as there is no payment of capacity-genedators generate income exclusively when they produce energy. The isolated network of states, extreme weather events and high renewable penetrations make a precise forecast and more challenging and financially due than in many other markets. Fortunately, artificial intelligence (AI) is now able to produce very accurate forecasts from the growing amount of meters and time data available. Complex and robust budgets derived from these machine learning algorithms are significantly more than what human analysts are capable of, making prediction of the system key to utilities. In addition, they are increasingly valuable for independent electricity manufacturers (IPPS) and other energy traders who make decisions about their positions in wholesale markets. Sean Kelly, co -founder and executive director SmartA company that provides solutions for a AI forecast, said the use of the Excel budget table as a prediction tool in order in 2005 when it started with work as an electricity trader, but this type of system is no longer appropriate today. “Now, we literally run in an ampere of four to six models behind the scenes, with five different suppliers of the weather that runs the ensemble each time,” Kelly said as a guest on Podcast of power. “So, as everything becomes confusing, we have to stay at the top and this is actually driven by machine learning.”
The consequences of poor preparation can be terrible. Having wounds and accurate forecasts can mean a difference between surviving business or failure. Effects of winter storms offer a case. Usually, wholesale prices Ercot -a fluctuate with about $ 20/MWH up to $ 50/MWH. During the Winter Storm URI (13–17, 2021 February), Ercot set a wholesale price of electricity on its limit of $ 9,000/MWH due to extreme demand and widespread failures in a generation caused by a storm. This price remained in effect for about 4.5 days (108 hours). This increase in prices of 180 times has had devastating financial impacts on the Texas electricity market. The financial outburst was strong. Several electricity providers in retail in retail were to go bankrupt, most notably Griddy Energy, which conveyed wholesale prices directly to customers, which resulted in receiving an account of more than $ 10,000 in just a few days of power. Brazos Electric Power Cooperative – the largest and oldest electric cooperative Texas – is subjected to bankruptcy protection. Chapter 11. After facing Ercota with $ 1.8 billion. Rayburn Electric Cooperative faced more than $ 1 billion of energy costs during the storm. CPS Energy Communal Municipal Municipal Services Antoni-Ercota has been reported over excessive prices and faced a billion dollars of storm-related costs. “Our clients were very appreciated by the job we had on the Amper,” Kelly recalled. “We probably had a dozen clients at the time, and we told them on February 2 to come,” he said. With that early warning, Kelly said that AMPeron’s clients were able to get out in front of the prices swing and buy strength at much lower prices. “Our forecasts come out for 15 days, Ercot’s forecasts only come out only seven,” Kelly explained. “So, we told everyone,” A warning! Warning! This is coming! ‘Dr. Mark Shipham, our internal meteorologist, was screaming from the roofs. So, we had a lot of clients who bought $ 60 per megawat. So, consider buying the 1960s, and then the opportunity is 9,000. So, a lot of merchants made money, “he said.” All LSE -II – burdened subjects – they were still affected extremely bad, but they hit a lot less bad, “Kelly continued.” I remember I said one client: “I bought electricity at 60, then bought it at 90, and then I bought it at 130, and then bought it in 250 bad. I managed to keep my company as a retailer supplier. “And, these are just some of the ways in which these forecasts are extremely useful.”
After the URI Winter Storm, Texas legislation has brought accounts that allow utility services to secure their URI debts through bonds supported by bonds, expanding costs over the decade. This may have saved some bankruptcy companies, but did not remove the financial burden. Some municipal services owned by the City received the financial support of their municipal governments. Many cooperatives and other utility services have eventually transferred costs to customers through increasing rates that have spread over the years. The crisis has discovered significant vulnerability in Ercot’s market design, especially the way the financial risk is assigned during extreme weather events, and has led to regulatory reforms regarding time and market rules. However, precise prognosis is still vital for the electricity industry. Because more and more renewable energy sources are added to the net, Kelly said he saw the market to be binary. “It will be zero or it will be one. And by that, I mean, it will be a power of $ 10 or it will be $ 1000,” he explained. “This job is more difficult and harder day by day – for software companies, but really for those loads that serve the load,” Kelly said. “So, we have to adopt new technologies here and always continue to improve ourselves, improve our knowledge of new things that get down into the tube and just work together to make the network a much more stable place.” Hear the whole interview with Kelly, which contains more about how energy markets work; Changing market dynamics; Other examples from Australia, California and the Elliott Winter Storm; challenges with precise prediction; how AI improves the process; And more, listen Podcast of power. Click on the SoundCloud Player below to listen to your browser now or use the following connections to reach the show on your favorite Podcasta platform:
Podcast of power · 185. Energy Energy Prognosis on AI: In order for accurate predictions to save your business
For more Power Power, visit Podcast of power archives. –Aaron Larson Is Power’s executive editor (@Aararonl_power, @powermagazine).
