Big Data analytics: Breeding the next-gen smart and efficient wind farms

Determined to drive down operating costs, wind operators are turning to Big Data to leverage the data, and to simplify and streamline operations and maintenance across projects. Here, we do an analysis on how Big Data and predictive analysis made the wind turbines grow progressively bigger, and more powerful.

Over the coming decades, digital technologies are set to make energy systems around the world more connected, intelligent, efficient, reliable and sustainable. The renewable sector already contributing around 20 per cent of India’s total installed power capacity is mainly driven by over 34 GW of wind power. The government’s target of reaching 175 GW has now been revised to 227 GW by 2022.

The wind industry, on the other hand, is going through a change. The industry is witnessing a shift from feed-in tariffs to auctions to determine the cost of electricity. With a reduction in per unit cost, it is necessary to identify methods to maximise the generation from wind turbines. The wind farm operators understand and manage the performance analysis of their farms to achieve desired production and revenue goals to stand par with the economically competitive wind energy industry.

The competence of Big Data
The wind industry has boomed over the last decade, growing more quickly than what many energy optimists predicted. Today, a virtuous cycle is underway, seeded by public subsidies and investments; driven by the need for energy security and the goal to reduce emissions; and sustained by a familiar process where the scaling up of manufacturing capacity lowers per-unit costs, increases competitiveness with conventional fuels and drives further innovation.

Wind energy systems stand out from other complex technical systems because of the combination of large levels of wind uncertainty and high levels of interaction of wind farm physics. Big Data analytics techniques can significantly improve wind farm performance and reduce costs.

Given the uncertainty and complexity associated with wind energy systems, there is huge potential for these techniques to significantly improve the performance and reduce the costs of wind energy systems. Big Data is boosting power production, reducing downtime across wind fleets.

Duncan Koerbel, Chief Technology Officer, Suzlon Group, opines,” To meet the government’s demand of 227 GW, the wind energy industry must focus on employing cutting-edge technologies, leveraging Big Data analytics to cater to bigger and increasingly reliable turbines, improving supply chain, enabling grid integration and improving the service standards to enhance the turbine’s reliability and performance.”

Vibhav Gupta, Business Development Manager, Algo Engines says,” Big Data analytics provide a unique proposition to analyse the tons of data generated by a wind turbine to present actionable insights. On a simplistic level, data analytics can be leveraged to identify under-performing wind turbines based on power curve analysis.”

Historical data can be leveraged to create power curve of the turbines for two different seasons and compare deviations. Power curve of two adjoining turbines can also be constructed to visualise how similar wind speeds can impact performance. Power curve analysis can also identify deviations due to derating, under-performance, firmware upgradation etc.

Predictive analysis: Less downtime, more wind power
The ability to spot and stop problems months before they start was once thought to be years away. But today, it is possible due to specific advances in technology during the last five to ten years.

Predictive analysis identifies turbine failure times and optimise repair and replacement to extend asset life and minimise downtime.

According to Vibhav Gupta of Algo Engines, “The biggest gain from data analytics for a wind farm lies in predictive analytics. Predictive modelling can be used to identify likelihood of component failure and plan maintenance activities. Machine learning models can be leveraged to predict energy generation and plan maintenance activities in periods of low wind.

A gearbox failure when not predicted can cause a turbine to be unavailable for a month in case of spare unavailability. However, the same downtime can be minimised considerably if the failure is predicted beforehand, giving the site team sufficient time to plan the maintenance activity. “

Duncan Koerbel, of Suzlon Group, says,” Suzlon was employing the ‘Internet of Things’ concept long before it became a buzz word. Predictive analytics plays a key role, as it allows to better predict or forecast the power generation. Predicting turbine operations and energy output with higher certainty 12-48 hours ahead of time will enable greater penetration of wind into the grid as a reliable energy source.”

Better analytics simultaneously assists in identifying areas for enhancing the wind turbine generation and optimising the operation and maintenance costs by proactively identifying failures and taking action before the failures occur.

Breeding the next gen smart and efficient wind farms
In future, the introduction of augmented and virtual reality based solutions will be effectively utilised for training engineers, in various health & safety scenarios, thereby, reducing the chance of human and material losses.

Duncan Koerbel of Suzlon Group opines,” In order to mitigate the effects of reduction in margins, the employment of predictive/ prognostics analytics based maintenance, condition monitoring system and inventory optimisation will play a significant role. This transition to automation and digitalisation of systems and processes, now form the DNA of Suzlon’s OMS operations.”

Vibhav Gupta of Algo Engines, while talking about the future of Big Data in wind farm said,” There are multiple areas in which data analytics can be leveraged to identify areas of improvement for a wind farm; however, it will only be effective if an organisation takes a proactive approach on the insights. We at Algo Engines help wind asset owners focus on critical aspects of plant performance and thereby, improve their bottom-line. “

Demand for renewable energy will be higher than in the past due to state laws and consumer awareness of how energy consumption affects climate change. Thanks to engineering advancements, turbines have captured more energy from the wind during the last four decades. However, the biggest opportunity for the wind power lies right in front of the industry: leveraging the data it already has in abundance.

Wind farm operators can then better understand the current scenario in the field, plan the timeline, and accurately predict the operating life, resulting in minimising downtime and maintenance costs with improved performance. Soon, stopping potential turbine problems before they occur will be typical; wherever the wind blows, there will be a turbine to capture its energy.

“The transition to automation and digitalisation of systems and processes, now form the DNA of Suzlon’s OMS operations.”
Duncan Koerbel Chief Technology Officer, Suzlon Group