Here come the Meteorologists! Business Critical Forecasting for Renewables

As the European energy market homogenises and is used for balancing demand and supply, the need to predict renewable generation becomes critical.
Published: Fri 04 Mar 2016

In April 2010, Dow Jones launched Lexicon, a service which “converts news content into measurable, actionable data, examines massive volumes of Dow Jones real-time news, identifying crucial words and phrases that signal trends or patterns analyses data” and sends real time information to professional investors. This in itself is not remarkable; the true story lies in the fact that many of the users of this service were not human, but algorithms there are able to analyse and act on the information received.

While the bear pit trading environment of the ‘80s is fast becoming a fading memory, the same cannot be said for increasingly active energy trading market. In this era of renewable energy integration, data companies are betting big on big data to provide business intelligence based on weather data.

We spoke to Robert Boucher, Product Director, Energy Services at The Weather Company, an IBM business, who shared insight into the importance of providing energy traders with accurate and easy-to-use meteorological data so that they can make decisions which could lead to greater profitability.

Energy traders in markets with a large number of embedded renewables are increasingly interested in being able to predict the availability of energy resources in ‘as close to real time as possible’.

Finding ‘the edge’ in the energy market

Today, the biggest variables in the electric market are volatility and variability. This affects energy trading, as well as demand and load management for the distribution network.

This opportunity to provide an edge is  not lost on big data companies like IBM who have just completed the acquisition of The Weather Company.

Mr Boucher explains that it is becoming increasingly difficult to find an ‘edge’ in the market as many of the inequities have been regulated out of the market, but that organisations that are willing to think outside the box and do things differently, are the ones that will benefit. He cautions though, that those that have the edge may only have it for a very limited time span before the others start catching up.

“We are trying to transform ourselves from a company that was in the business of delivering the best weather guidance and forecast information to one that uses that information and marries it up with industry data to provide clearer insight for better energy predictions.”

Energy trading companies have to transform themselves if they want to stay ahead of the game. This can be done by tapping into the best in weather forecast data and marrying this up with industry data such as the efficiency of a turbine blade.

“It’s not just about the wind speed any more. It’s more about predicting what wind power that wind speed can actually deliver by linking manufacturer data to new real time models.”

While traders have yet to pin an ideal tolerance level when it comes to accuracy of predictions, they prefer to avoid steep fluctuations since this can be challenging when trying to develop a trading strategy, explains Boucher. “The aim is to reduce data volatility so that traders’ can make critical decisions more confidently."

Thriving in the energy trading market

The energy trading market has become more complex and competitive as balancing authorities are snapping up the opportunities in trading.

Data has to be very accurate for energy traders to thrive, especially when it comes to smoothing out price fluctuations. To reduce the margin of error and uncertainty, data capturing companies have to keep going back to the drawing board to find ways of improving the data collection and analysis for traders.

Finding the right skill sets can also be challenging for the energy trading industry. It’s not just about hiring a data scientist anymore. Boucher explains: “Skill levels need to include expertise in both statistics and computer programming since these must be married up in order to develop accurate and efficient trading models.”

In addition to this, data companies are expected to build two models for the energy trader: predictability and financial viability. Data vendors also have to sell their product strategically because if it is sold to the entire industry, its value will be diluted. But often, traders are not willing to pay the price for a customised or ‘limited edition’ solution.

Energy trading needs reform

Renewable generation and integration has resulted in a more complex energy system which requires a more detailed and sophisticated level of management if it is going to work. The introduction of smart meters has led to a greater quantity of data which, if analysed correctly and efficiently, could help drive greater operational efficiencies and cost savings.

The energy trading industry has been characterised by the emergence of new markets, higher trading volumes and more stringent risk and regulatory reporting requirements. To adapt, it has become clear that spreadsheets alone are neither scalable nor robust enough to handle the escalating trading volumes or middle and back-office requirements. [Tough Energy Trading Environment Drives Need for Sophisticated IT].

According to Boucher, the energy trading community tends to operate in a very traditional manner, usually because the amount of time available for innovation or exploring opportunities for increasing efficiency is limited and changing workflow and processes is a luxury, rather than a necessity. Boucher believes however, that this is where there is an opportunity for vendors to step in.

He continues that “if traders find that someone is doing something that gives them an edge, they will actively try to replicate it. Traders who are willing to put themselves out there, who are forward thinking, are the ones that are going to capitalise on the opportunities.”