By using our site you accept the terms of our cookie policy

Search

Make renewable energy more reliable with analytics

June 26, 2018

Image

Richard Drake, Energy Practice Lead, FICO

Richard Drake has nearly 20 years’ experience working in the energy sector. At FICO, he helps energy firms in Europe, the Middle East and Africa make better decisions about energy using advanced analytics.

Consumer and political demands, alongside improved engineering have resulted in greater global uptake in the use of renewable energy[1]. According to a report published earlier this year, it will also be consistently cheaper than fossil fuels by 2020[2].

But cost isn’t the only factor. Reliability remains a key limiting factor for renewables, and new types of power generation come with diverse new maintenance requirements. In the search to make renewables reliable enough to truly replace fossil fuels, predictive and prescriptive analytics are key, working hand in hand with sophisticated optimisation algorithms.

Power at the mercy of the elements

One of the significant challenges with renewable energy is that it can be impacted by factors beyond human control; solar farms take a significant productivity hit on a cloudy day for example, and wind turbines are – unsurprisingly – entirely reliant on wind.

Currently, when renewables produce less energy due to poor weather, other sources of energy can ‘pick up the slack’. But optimising energy networks to handle unpredictable shifts in consumer demand is a complex process and introducing renewables into the mix adds another variable to consider.

Power plants take time to reach full capacity and doing so is costly. To avoid power outages and maximise value, we must build intelligent models that optimise our energy mix. By using predictive and prescriptive analytics, as well as optimisation algorithms, we are able to do this.

Models can be built to compare the energy flow between renewable and non-renewable sources, as well as data streams that predict the weather. Doing so will enable accurate predictive models that determine exactly when your energy supplies will need to be augmented with fossil fuels, and by how much.

But it’s not just short-term weather prediction models that governments and organisations should employ if they’re looking to secure the power grids of the future.

Recognising that we sit at a crucial turning point in the history of energy production, one Norwegian energy supplier called Statnett has been using advanced analytics to forecast future energy requirements and market behaviour through 2040 and beyond[3]. The insights they gain today are informing the investment decisions they will be making for decades to come, ensuring they’re able to manage the transition from non-renewables to renewables smoothly.


Decisions without precedent

Moving to a greater reliance on renewables is a step into the unknown for the energy industry. Maintenance is difficult: how do you know what your maintenance schedules and practices should look like for a new system if you don’t have historical data to base your decisions on? There are ways for analytics-based energy optimisation solutions to help here too.

One example of this challenge is that a wind farm is much larger than a coal plant, meaning it is much harder to inspect. To do so in person is a massive undertaking, but crucial. Wind farm operators must build an accurate picture of the operating health of each turbine and calculate any impact on energy output based on this.  

Deploying drones to inspect the operational health of wind turbines and using high-resolution cameras with thermal imaging helps energy companies to get a more accurate picture that can be fed into maintenance optimisation algorithms. This is particularly valuable for wind turbines and solar panels in remote locations (which tend to be prime locations for renewable energy developments), because it minimises the environmental impact of human activity.

Wind power optimisation 

Since renewables are a nascent form of energy production, there are also opportunities to refine the design processes used to build plants.

For example, optimisation techniques can be applied to ensure that wind farms are set up to maximise power output and minimise environmental impact. The layout of a wind farm has a profound impact on its productivity; developers need to balance visual influence, noise, wake interactions, turbine loads and more if they are to extract the maximum value from their facilities. Doing so involves huge, complex calculations, which require cutting edge analytics technologies.

Norwegian wind farm designer Markedslabben has achieved this using advanced analytics to optimise the layout of its wind turbines[4]. The results were significant, increasing energy yield and profitability by 2-5 percent. If wind power was to replace coal entirely in Norway, and all new wind farm projects were optimised in this manner, annual carbon dioxide emissions could be reduced by 20-50 million metric tonnes.

Where next?

The world’s migration from fossil fuel-based energy to renewable energy is a bold but important goal. It will require vast technological adaptations to ensure humans can harness the power of nature. Advanced and predictive analytics can minimise uncertainty and help energy companies to better understand and predict the behaviour of the environment, and above all build a sustainable future.

7

Contact Us

+44 (0) 203 725 6841