As financial support for wind energy started falling over the past few years, various software applications have been created to make more use of the potential of digitalisation and big data to improve efficiency and reduce costs.
Digitalisation started making its presence felt roughly a decade ago, as sensors were built into wind turbines and ever increasing volumes of data were collected to be analysed for condition monitoring of components. Predictive maintenance and improved wind-farm design are just two of the benefits it has provided.
Turbine makers, asset owners and operators, and independent service providers (ISPs) around the world are testing the latest software products on offer, with a keen eye to keeping up with the competition.
Now another buzz concept with a whole new level of activity has entered the frame: the industrialised internet of things (IIoT).
The IIoT is another gamechanger; developments are unfolding fast, and in various directions. The information technology business has discovered energy and the wind sector, and finds it extremely promising.
Intelligence is key
"Digitalisation, big data and the IIoT will become much more significant than technical questions like whether a turbine has a gearbox or not," predicts Andreas Reuter, head of the north-west division of the Fraunhofer IWES wind energy and energy systems technologies institute. "Intelligence of the wind-power station will become the overridingly important aspect."
Broadly speaking, the IIoT means connecting the digital world of the internet with conventional processes and services, with a focus on digitalisation, communication, data management, industrial analytics and interoperability.
The IIoT is different from the consumer internet of things, where your fridge informs the supermarket that you need to stock up on milk, and the living room blinds close when the TV goes on. But further down the line, interlinking the two will not be difficult.
The IIoT goes under a number of names, depending on country, and the emphasis on application varies. In the US for example, the main focus is on integration across domains, such as integrating wind-farm operation with electricity markets.
Germany's equivalent, known as Industry 4.0, is more manufacturing-oriented. Similar efforts around Europe are the French usine du futur, the Netherlands' smart industry, Sweden's Produktion 2030 and the UK's industrial strategy, according to the Fraunhofer Institute for Embedded Systems and Communication Technologies in its April 2016 publication, Industrial Internet of things: reference architecture for the communication.
Traditional processes and services are being absorbed by the IIoT. These include practices and know-how for condition monitoring that have existed in the wind sector for quite a while.
"In the offshore wind business, condition-monitoring systems utilising sensors and analysis of sensor data have been obligatory for over ten years," says Eddie Monch, an expert for industrial analytics at software provider Empolis Information Management.
"The driving force behind these systems was the 'Allianz law', criteria established by insurance giant Allianz and based on the guidelines first developed by Germanischer Lloyd (now DNV GL). These criteria must be fulfilled before a project can be insured.
"These requirements have extended to onshore turbines. With the capability of collecting and evaluating all turbine data, the wind-power industry has attained a unique position, placing itself ahead of all other industrial sectors," says Monch.
"In classic industry sectors, such data is not there, it has not been collected," agrees Reuter. "Thanks to condition monitoring, the wind sector has all the data, and the turbine is already on its way to becoming 'intelligent'. The data volumes are there in enormous amounts, but their use is still rudimentary," he says.
Condition monitoring from data collected from the 40-50 sensors installed in turbines today is still at an early stage. Trends are analysed, "but this is far behind what we imagine can be derived out of big data and put to intelligent use", Reuter says.
The transition from digitalisation and big-data usage into the IIoT is seamless. Assuming good interconnectivity, data from a wide variety of sources - condition monitoring, wind and weather, technician-service schedules, "virtual" power stations, electricity markets - are brought together in the IIoT environment for use in new, useful and lucrative ways.
Combining digitalisation and big-data competences with other domains of activity in the IIoT environment goes further than pushing the advance from basic condition monitoring to smart predictive maintenance.
There are many other advantages: optimising service schedules; control over individual turbines and their components; adapting asset output to optimise earnings in the electricity markets; running wind farms within microgrids; integrating with industrial manufacturing processes; playing a role in the power-to-gas and power-to-heat sectors. The list is long.
While companies forge ahead with ideas for new applications in the IIoT world, some are making efforts to bring order to the unsystematic jungle of developments marking this early phase of progress. Without good interconnectivity and agreed standards, the IIoT could break into islands of activity, controlled by a few large companies.
The US-based Industrial Internet Consortium, which includes companies such as AT&T, Bosch, Cisco, GE, IBM, Intel and SAP, presented a first draft for an overall IIoT architecture - the industrial internet reference architecture - in mid 2015.
This followed close on the heels of an offering entitled "reference architecture for Industrie 4.0 (RAMI 4.0)" from Germany's electrical and electronic federation ZVEI, before the two bodies decided in March 2016 to work together.
One of the aims is to have industrial internet standards set to ensure that IIoT platforms can communicate easily with each other, at least at a basic level.
They need to work fast because major IIoT players, including names such as Amazon, Cisco, IBM, Microsoft and SAP, are building their own industrial internet platforms into which energy and wind-power will be integrated.
Among their number are industrial giants like GE and Siemens with hands-on experience of industry technology and engineering, which are linking up with software companies for specific application areas such as energy and, as a subdivision, wind power.
GE has teamed up with technology provider PTC's Thingworx platform for its digital-wind-farm portfolio. Siemens is working with Atos for the digital services of its Sinalytics platform with wind-power applications, and with RTI's Connext application for turning a wind farm into a "smart, distributed machine".
They face customers who may want the whole package of turbines, plus an analytics portfolio from their OEM, or who prefer to compile their own analytics portfolio for their wind projects.
Independent supplier Empolis offers an industrial analytics platform that provides special service-diagnostics applications for wind energy.
Its box of tricks includes root-cause analysis, error prevention, data aggregation from multiple sources, text mining to interpret technicians' service reports in different languages, visual interaction for further ease of use, and continual optimisation in turbine-service processes.
But focusing digitalisation and big-data applications solely on the turbines and wind farm to raise output and lower costs is fast becoming outdated. As wind power starts to become a player in wholesale electricity markets, the economic view on generation costs is changing.
"A kilowatt hour does not always have the same value. When it is very windy you hardly get any money for your kilowatt hour, and vice versa," says Reuter.
"One outcome is that the standard service strategy of performing maintenance in low or no wind periods is being re-examined," he says.
Within the IIoT, this aspect of wind-farm operation will be linked to many other factors, including other renewables generation, battery storage and electricity demand. As turbine manufacturers continue on their cost-cutting path, wind turbines will become more similar and other factors gain in importance.
"Turbine hardware will tend to become a commodity - what will be important is the price per kilowatt hour and the intelligence of the system and, with it, the possibility to maximise profits," says Reuter.
Coordination within the IIoT will then be extended to interact with the consumer side of things. Tools such as blockchain will enable the electricity demand side to fetch and pay for its supply in real time - including from wind turbines - through the internet.
Some companies are already taking advanced analytics of turbines and wind farms to a new level of integration across domains. In September 2016 Chinese company Envision Energy unveiled its EnSight energy analytics platform for the wind industry, and its EnOS system.
Describing EnOS as a "smart, scalable and open platform that is enabling the IIoT for energy", Envision says it "orchestrate(s) all types of energy infrastructure, connecting wind turbines, PV panels, energy storage, electric vehicles, smart grids, smart meters and home appliances, and other energy devices and machines".
The global rollout of EnOS will enable homes, neighbourhoods and cities to become self-sustained sustainable power plants, claims the company.
The possibilities of technology may be wide-ranging, but it seems nature can still put it in its place. "Despite the huge potential offered by digitalisation, big data and the IIoT for wind energy, it can't completely replace the technicians in the field. There is an 'end to the vision'," points out Empolis's Monch.
"Take a look at the rapeseed fields in Germany, where many wind turbines are located. When they are in full bloom in the spring, the enormous amount of pollen can quickly block the filters, the generator overheats and the turbine switches off. Cleaning the filters can only be carried out by on-site technicians."