skip to Main Content

How does your predictive maintenance solution adapt and improve over time?

A continually developing industry

The wind industry doesn’t stand still as it continually drives towards greater turbine and wind farm capacity. And when it comes to predictive maintenance programmes for the industry (PdM) – involving advanced technologies, such as big data, artificial intelligence (AI) and machine learning (ML) – adaptation and improvement is at the heart of the discipline.

Using the predictive power that these technologies unlock creates a virtuous circle for owners, allowing them to continue learning about their machinery and improving operational practices. But the technologies that underpin a predictive maintenance solution are themselves being continually adapted and improved by the experts. It is essential then that owners and operators considering working with a PdM supplier, seek to understand how the service provider maintains the competitive advantages of their solution over time, how they update it, adapt it to new demands and, ultimately, continually improve it to be fit for the future.

Fit for the future

Predictive maintenance solutions can be developed, improved and disrupted on three broad fronts – the hardware used to collect the data, the digital platform used to collate and analyse the data, and the working relationship between the ISP and the asset owners

Turbine owners relying on predictive maintenance solutions to improve their operations, want to be sure the technology and the programme it supports remains fit for purpose as their business and assets adapt to a post-subsidy world with increasing competition pressures.

Improving data collection and quality

For data collection, adaptation and improvement can start with the quantity and quality of the data collected. Condition monitoring systems (CMS) continuously collect data from a wind turbine and digitise it. ecoCMS, a CMS unit from ONXY InSight uses microelectromechanical technology (MEMs) to collect data. A wide range of industrial uses, especially in mobile phones, means MEMs technology is not only highly reliable, there is also substantial on ongoing development to keep improving the technology.

Much field data continues to be collected using manual processes and recorded using pen and paper, which limits its contribution to effective predictive maintenance. Digitalising the collection process through mobile applications not only allows this type of data to be used in PdM models more effectively but means data capture can be more easily improved and adapted in the future.

Monitoring more, seeing more

The expanding capabilities of data collection hardware and growing turbine portfolios means analysts and the tools they use, need to handle rapidly increasing amounts of data. Many PdM monitoring platforms are limited in the types of data they can incorporate and analyse, and while they can handle CMS or SCADA data struggle to accommodate digitised field data such as oil analysis, inspection reports and images.

Top-performing platforms will have an exhaustive list of adapters that allow them to handle all types of data, either from different types of turbines or from different data collection processes, increasing the power and value of the insight it can offer to a monitoring team. Going further, the latest monitoring platforms adhere to the highest security profiles, ensuring data is safe, but not hidden. Platforms should not limit your access to data streams through encryption, partial reporting or format limitations. Only with full data disclosure can platforms provide the best reporting and recommendations for engineering teams to use.

AI meets real-world

AI technology is critical for the scaling up of the predictive maintenance solutions. When AI and ML technologies are utilised in conjunction with a strong foundation in physical engineering principles, a conscientious approach to model evolution boosts the accuracy of analysis with much reduced  uncertainty.

Only through this approach, an effective predictive maintenance system could evolve over time as the machinery it monitors ages. The recommendations for maintenance and repair generated through AI technology could remain relevant and actionable as an owner’s fleet ages and develops.

Adapting to your business needs

Monitoring platforms can be built modularly, allowing them to adapt as users’ needs evolve. Beginning with a top-level dashboard to show trends and feature alarms, to a more in-depth tool for visualising and analysing technical data, such as drive train vibration raw signals. As operators’ operations and maintenance teams accelerate through the digital transformation, increasing their functions and responsibilities, more modules can be added to increase and support the effectiveness of teams.

As the expertise of an in-house team develops, a PdM provider should be able to adapt its offering, moving on from an all-inclusive monitoring solution with a high-level of technical and consulting support to a more on-demand consultative basis to build a relationship where the in-house expertise has overall control and calls on the consultant for their expertise when needed.

Getting the predictive maintenance strategy right can provide significant results for an operations and maintenance team. In wind, PdM helps cut back nearly 60% of operations and maintenance costs caused by unscheduled maintenance. Better understanding of turbine health and a more responsive and efficient maintenance programme that leads to reduced OPEX costs, can make turbine fleets more resilient to the changing market dynamics that owners are facing in a world with increasing merchant risk.

Back To Top