Using asset data to secure the best price for wind projects ahead of acquisition

energy data

The end of last year signaled a step change for energy investment. Decarbonization targets and the drive for sustainability in investment have made renewables a safer bet in the long run than traditional methods of energy generation.

As such, many shareholders are driving their boards to ensure their investments are sustainable, and to reflect this, a number of high-profile investment firms have announced their commitment to divest from fossil fuel assets.

To make the most of the subsequent increase in investment and acquisition activity around renewables, the industry will have to adapt to the asset management strategies and expectations of firms that have previously invested in conventional power ” which don’t traditionally account for intermittency or variation in output.

While the majority of wind asset acquisitions to date have been pre-construction or pre-operational projects, operational assets are increasingly interesting to new investors looking to enter the market. As there is historical data detailing the output of the project to date, the risk profile of these projects is more concrete as energy yield assessments are based on real-world project data rather than assumptions from pre-construction modelling.

As such, many investors are seeking operational projects that have clearly been finessed to get the best from the available resource, or projects which have the potential to maximize generation with a few tweaks.

This presents a key challenge for the renewables industry, wind power in particular, as the majority of projects are not achieving their generation potential. While it is tempting to blame low energy production on low winds, the industry must face the fact that in many cases persistent underperformance is a result of the turbine interacting poorly with the wind resource, not the level of wind resource itself.

Asset owners must rectify turbine underperformance ahead of sale, in order to be in the best possible position to take advantage of increased investor interest in renewables.

Digging into the data

Historical data is key to determining the project’s price ahead of acquisition. However, a roundup of annual energy production (AEP) is not the only consideration potential asset owners will bear in mind. By putting turbine health and performance data in the context of environmental data and analyzing it through an AI-driven platform, potential buyers can identify and quantify whether low energy production is due to the resource or the turbine, and determine avenues for optimization should they decide to acquire the asset.

When onboarding a project onto Clir’s platform, we digitize all available atmospheric, meteorological, and geospatial data relevant to the project site, and use AI to analyze SCADA data in context. Ultimately, this reduces uncertainty around future performance, raising the value of the asset and enabling a better price for sellers and better leverage for buyers.

Crucially, wind farm owners looking to sell an asset can also use this data to guarantee the best possible price. By identifying the causes of underperformance quickly, and either acting to fix them pre-sale or demonstrating to the buyer that the fault is easily solvable, owners can demonstrate that the project’s AEP could increase by 5%. This raises the value of the project with lower CAPEX needed than retrofitting new technology or committing to repowering.

In-depth analysis of a project can also demonstrate the condition and value of an asset against that of the broader portfolio, maximizing the asset’s sale value should it be clear that it is performing above the portfolio benchmark. This level of transparency can help assure potential buyers that the project will generate strong returns for the expected asset lifetime or beyond. Conversely, if the project is performing highly against the portfolio benchmark but individual asset data indicates that the components are showing significant wear and tear, this may well negatively influence the sale value of the project.

Red flags for future faults

Prospective buyers are also using AI-driven software to identify any red flags in turbine data that indicate significant faults with the turbine ” whether the fault is immediately evident, or if there are indications that a problem is developing.

For example, Clir’s software is able to identify gearbox failure early on by recognizing unusual patterns of temperature and vibration. While the gearbox may still be functional, prospective owners may hesitate to invest in significant repairs and replacement parts ” all of which will take the turbine offline. This ultimately drives down the value of the project.

With an in-depth picture of mechanical health of the turbine, current owners of projects can make any repairs early on and before the faults become critical, reducing the cost of replacement parts while increasing the value of the wind farm. In the longer term, this data analysis can also help determine whether individual turbines may be run harder than could be assumed from their design.

Small but significant improvements

While significant causes of underperformance can be quickly identified in routine inspections, many smaller, more subtle causes of underperformance can slip under the radar and cumulatively reduce AEP. For example, while an error in blade pitching for a single asset may fractionally impact performance, if this error is farm-wide the asset owner may find themselves losing out significantly – this will be reflected in the value of the wind farm at sale.

Increasingly, asset owners are using advanced analytic software to track down these smaller causes of underperformance and direct their operators to rectify them, increasing AEP. Parsing the root of underperformance out from the “noise” of wind resource and other turbine data is impossible through traditional data analysis methods, but the right software can properly label turbine performance data, place it within the context of that particular environment and wind resource, and ultimately give owners a full picture of asset performance and value.

AI-driven data analysis platforms also allow owners to model turbine performance from an individual turbine’s historical data, rather than through peer-to-peer comparisons. This prevents small but common problems ” such as pitch and yaw errors – from being missed. It is also possible to use Clir’s platform to assess the potential for asset upgrades than will trigger improvements in performance post-sale, and outline the value that these can add at the financing stage.

Time waits for no deal

Properly analyzing wind farm data can be a complex, arduous process through more “traditional” means. It is quite easy to lose weeks going through asset data to achieve an accurate picture of turbine health and performance. However, stretching out the timeline like this can lead to both the buyer and seller sinking unnecessary costs into the deal. Human error also increases the chances of key risks and opportunities being missed.

Data analysis powered by artificial intelligence significantly cuts down the time needed to analyze renewable energy project performance. Clir’s platform cuts the time needed to analyze multiple streams of data and build a full picture of turbine operations and health down from weeks to hours.

All parties involved in the acquisition will benefit from having more immediate access to this information, as it allows decisions to be made quickly based on a true understanding of the performance of that particular asset.

Ahead of the curve

It may be tempting to sit back and assume consistent returns mean your assets are producing as much energy as they could be. However, large asset management firms will be combing through the data during due diligence ahead of an acquisition, looking to identify avenues to achieve higher returns or to negotiate a more favorable price due to underperformance.

Proactive renewable energy project owners are already taking advantage of software that can assess gigawatts of assets in a matter of days. By using an AI-driven analysis platform to gain a full understanding of the health, performance, and potential of their assets well ahead of negotiations, these owners will be in a very strong position.

Author: Gareth Brown, Clir Renewables

No posts to display