While wind turbine manufacturers and wind farm operators have developed suites of advanced applications to help diagnose operating anomalies on wind turbines, many are still taking a “trip first, analyze later” approach.
This approach made sense when the application of near real-time analytics was not cost effective. However today, near real-time analytics can be applied seamlessly within the operational workflow of existing monitoring operations, resulting in several key benefits.
First, real-time analytics enable operators to avoid tripping turbines needlessly and reset tripped turbines faster, resulting in more power output due to higher availability. This application of automated analytics can also result in safer asset operation of turbines with consistent, automatic review of event data to drive decisions.
Second, complex analytics that are currently being processed in batch mode by diagnostic engineers can now happen automatically in near real time. This closes the gap between insights showing what is happening on the turbine and the ability to respond quickly, resulting in better wind turbine performance.
Lastly, the use of real-time analytics makes monitoring operations vastly more efficient, as monitoring personnel and diagnostic engineers are freed from routine tasks and can focus on increasing asset performance and uptime.
While some companies may have shied away from near real-time analytics in the past, it is now easier than ever to deploy this type of monitoring operation that has low risk and high impact on wind assets.
In order to a create successful digital transformation using real-time analytics, companies must implement a solution that:
- Incorporates your Subject Matter Experts’ knowledge
- Enables engineers to create analytics within the platform
- Provides automation of key business processes (E.g. alarm diagnostic)
- Builds a foundation for more complex analytics (E.g. machine learning)
Having software built on your Subject Matter Experts’ knowledge is the best starting point.
Your Subject Matter Experts (SMEs) and engineers have been building and maintaining your critical assets for years. Their knowledge is the best available expertise on how your equipment should be operated, maintained, and evaluated. Incorporating their knowledge about how to best evaluate data from the equipment is the ideal starting point for the application of real time analytics.
Analytic platforms using purely Machine Learning or Artificial Intelligence can lack insight on the data. Without human interpretation, more complex analytics, such as Machine Learning, have a difficult time achieving the desired outcome.
Using software that allows engineers to create analytics within the platform helps with the adoption of analytics.
Adopting new analytics and data driven business models is about changing the way business has been done for many years. Having software where your SMEs can interact and engage, without needing assistance from a data scientist or a developer, allows users to impact business outcomes and drive higher adoption.
Implementing software that automates current processes creates both short and long-term significant value.
The software must be able to interpret data, generate insights, and provide recommended outcomes. Otherwise, it’s just another way to visualize and explore data.
Having a system first built on your proven analytic methods, and then adding a layer of more advanced analytics, such as machine learning predictions, is the best route to a highly accurate, automated platform.
In order to gain value from large datasets and sensor data, a solution must automate part of the process to create scalable value.
Building a platform with a solid foundation of your experts’ knowledge is the best way to approach implementing an entire suite of analytics.
Using software that can be configured by your own SMEs creates the optimal foundation for an entire range of analytics. Your experts can provide knowledge about significant areas such as:
- Decisions relative to operational sensor data
- How sensors are related to each other
- What constitutes an actionable event & false alarm
- Exceptions to the rule
Once your experts’ knowledge is embedded in an automated system, like NarrativeWave, adding a full range of analytics becomes exponentially more impactful. For example, knowledge of what determines a false alarm can lead to a business insight, describing what turned a false alarm into a valid alarm. In contrast, an approach that solely tries to use Machine Learning or AI techniques without these key understandings can struggle with accuracy, dealing with exceptions, and delivering significant value to the business.
Contact us to understand how you can effectively build upon your engineers’ existing knowledge to provide automation, productivity savings, and better asset performance.