The global commitment to cope with climate change by accelerating and intensifying the actions and investments needed to meet binding targets for a sustainable low carbon future agreed under the Paris Agreement in 2015, has fostered the increase in the penetration levels of renewable energies into the electrical grid.
In particular, wind energy has experienced a very significant growth over the last decades (from 6.1GW of global cumulative installed wind capacity in 1996 to 539GW by the end of 2017 ) which has led to play an increasingly important role in the energy mix. In 2017, for example, wind energy generated enough electricity to meet 11.6% of the EU-28’s total electricity demand  and in some countries such as Denmark, Portugal, Ireland, Germany or Spain, the percentage of the average annual electricity demand covered by wind that year was especially noteworthy (Denmark 44.4%, Portugal 24.2%, Ireland 24.0%, Germany 20.8% and Spain 18.6%).
As the level of wind power penetration into the utility networks increases, more restrictive grid codes required by transmission system operators (TSOs) has to be fulfilled by modern wind power plants (WPPs), in order to ensure stable and secure operation of the power system. Thus, WPPs must be able not only to generate active power, but also to provide ancillary services to the grid such as fault ride-through capability, frequency support and voltage regulation, as do conventional power plants based on synchronous generators.
Thus, it is clear that the wind energy sector has transformed from a niche sector in its beginnings in the 70s to a mainstream industry. Just take a look at Figure 1 where cumulative power capacity in the European Union from 2005 until 2017 is shown. As it can be seen, wind climbed from the 6th (2005) to the 2nd (2017) largest form of power generation capacity in Europe, closely approaching gas installations.
But … what are the limits of wind energy? Has it reached its maximum capacity? Of course not. Even with all this growth, wind energy still has to reach its full potential. Today, the LCOE for wind energy, especially offshore, can be further reduced to be more competitive, as well as the way wind farms operate is suboptimal and it is far less efficient and profitable that it actually could be.
Because wind turbines are increasingly larger - which means that fewer turbines are needed to generate the same amount of energy in the wind farm - their reliability and availability become crucial. O&M can be extensive and costly, particularly at sea, if the failure of a component occurs in a period of adverse weather where accessibility to the WPP may be limited or none. Likewise, the unavailability of appropriate vessels at a specific time can lead to long downtimes, thus impacting significantly on power generation, revenue and profitability. At the same time, the energy sector in general, but especially wind, has not fully embraced the connected era. Nowadays, wind farms generate terabytes of data through multiple sensors. However, most of such data is never consulted until a failure occurs. Decisive technological steps still need to be taken towards increased competitiveness for the wind industry. Leveraging of the combination of new sensing and advanced condition monitoring techniques, the rise Internet of Things (IoT) maturity and the possibilities of Big Data analytics, the lifetime and efficiency of the wind farm can be greatly increased and the cost substantially reduced. The wind is changing, and digitalization is coming...
How wind energy sector can be digitalized?
Over the last decade, many sectors, such as automotive or healthcare industries, has been disrupted by digital solutions, and wind industry has not remained indifferent to this revolutionary trend. Thus, by integrating data analytics, machine learning techniques and harnessing of advanced condition monitoring and the boom of IoT, wind actors has developed the Digital Twin technology to adapt its business model to the digital era.
A Digital Twin is a dynamic virtual representation of a physical object (e.g., a wind turbine) which makes use of real data to optimize maintenance strategy, as well as to improve reliability, availability and performance of the system. It receives data through multiple sensors (or even fully autonomous drones) and performs diagnosis and prognostics to optimize asset performance and improve productivity based on predictive analytics and machine learning techniques. Thus, the predictive model will aim not only to anticipate that a failure is going to occur and when, but also to identify what part is going to fail, in order to save time and money.
To get full visibility not at a turbine level but at a wind farm perspective, a Digital Twin is developed for each wind turbine within the wind farm and all of them are connected to a common cloud-based platform. Each Digital Twin carries out sequentially the follow 3 stages:
- See: This step consists in gathering data – operation and environmental – through various sensors that measure critical inputs from the physical process and its surroundings. Then, from all this information, messages or warnings are send out to the operators, so they can take actions before failure occurs.
- Think: In this second step, the model runs simulations and, based on historical data, fleet data and forecast for revenue, gives a number of options and reasons among the options with an associated risk and confidence interval.
- Do: Finally, the Digital Twin model inform us and executes what needs to be done.
Why is wind energy digital transformation crucial?
As it has been previously mentioned, digitalization will enable wind farm operators to carry out a better diagnosis of system performance, increase asset availability and thus reduce total costs. More specifically, it will bring the following advantages:
- Energy lost due to unplanned downtimes could be reduced by anticipating degradation and failures and taking proactive actions. This could be particularly critical if, as previously mentioned, the failure of a component coincides with a period of harsh weather or no logistical resources are available.
- Reduces costly emergency repairs by predicting minor issues before they become major problems.
- Turning from a routine preventive maintenance (often unnecessary) to a condition-based strategy enables to reduce costs, risks and to increase availability.
- The system will become more observable and controllable which will enhance its performance by optimizing the entire wind farm output instead of each wind turbine individually.
- Similarly, wind power plants will be able to improve their provision of ancillary services to the grid such as frequency and voltage support, as well as their ability to act as stand-alone generation.
- Digital Twin models based on enhanced predictive analytics and machine learning techniques will allow wind power operators to obtain economic profits by increasing the value of wind energy through data-driven trading strategies.
- Workforce productivity will be improved by rescheduling resources around specific actions that need to be taken based on its criticality.
- Health and safety of employees will be less exposed as the number of maintenance actions will be reduced.
- Stock list costs could be lower by determining spare parts requirements based on failure rates, logistical constraints and associated costs.
However, moving towards a digitalized world also implies certain risks. Relevant issues such as the threat of cyber-attacks or concerns around IPR considerations and data privacy can be very challenging and sensitive matters and should be taken into account.
Where are we and what is the next?
Despite wind energy lags behind other industries, such as media and retail, in its adoption of digital technologies, very relevant players within the sector, such as General Electric (GE), Siemens, Vestas, Nordex, among others, are starting to invest large amount of money to transform their core businesses into digital.
GE was the first to realize the full potential of the Industrial Internet[i] and now is leading the wind energy transformation into digital. GE launched in 2015 the world’s first Digital Wind Farm . It is comprehensive hardware and software solution build on the Predix cloud-based platform, which allows wind farm operators to connect, monitor, predict and optimize unit and site performance.
Vestas has decided to digitalize its business by acquiring Utopus Insights, a leading energy analytics and digital solutions company . With this purchase, the Danish wind turbine manufacturer company will have a greater predictability of the system allowing a better asset management and cost reductions.
Digital solutions have also arrived to the North Cape, in Norway. Arctic Wind, the Finnish wind farm operator, has developed a Digital wind farm to keep track remotely of the state of health of their turbines and increase the cost-effectiveness of their wind farms. For instance, a virtual replica of Havøygavlen, the world's most northerly wind farm, has been built to cope with the difficult and costly maintenance tasks under extreme weather conditions (temperatures of -25ºC!) and long periods of darkness that its technicians must withstand .
In addition, wind energy digitalization can harness of the insights from collaborative projects such as WindTwin: a project that aims to optimize wind turbine operation by developing a Digital Twin model whose consortium comprise the following engineering technology experts: Agility3, Brunel University London, Dashboard, ESI and TWI Ltd .
Digital Twin, Big Data, IoT, etc. are trending topics in wind energy these days. This can be reflected in different events and wind energy conferences where organizers are starting to include dedicated sessions into their programs to discuss about them. For example, WindEurope 2018 Conference at the Global Wind Summit, held in Hamburg, focused the Day Two on debating about the impacts of digitalization on the wind sector, generating a lot of expectation.
The wind energy sector still has a long journey to make to fully embrace the Digital era. Small steps have already been taken, but this is just the beginning… the transformation of the wind industry into digital has yet to come.
Wind energy platforms, such as the European Technology and Innovation Platform on wind energy (ETIPWind), consisting of political stakeholders, wind energy industry and research institutions, as well as, several initiatives, as the first ever “hackatlon”, a competition organized by WindEurope which allowed innovators from different backgrounds to come up with digital solutions for the wind industry, help to support and foster innovation towards a digital wind energy revolution.
Many questions still arise around digital solutions, and the answer is blowin’ in the wind.
References Renewables 2018. Global Status Report. Technical report. Renewable Energy Policy Network for the 21st Century (REN21), 2018.  Wind in Power 2017. Technical Report. Wind Europe, February 2018.  Digital Wind Farm. The Next Evolution of Wind Energy. Technical Report. GE Renewable Energy, May 2015.  https://www.utopusinsights.com/press-release-vestas-acquires-utopus-insights  https://blogs.sap.com/2016/09/21/digital-twin-helps-solve-arctic-challenges-at-remote-wind-farm/  https://www.windpowerengineering.com/business-news-projects/windtwin-digital-technology-aims-support-lower-wind-turbine-om-costs/
[i] The term “Industrial Internet” was coined by GE to refer to the combination of Big Data, analytical tools and wireless connections with physical and industrial equipment.