Learn how the multi-step waterfall approach in GPM Horizon can transform your data processing for faster, more accurate results.
Effective loss calculation is crucial for renewable energy operators and asset managers who need accurate and timely insights to optimize performance. Wind turbines are particularly complex, because they operate in dynamic environments with fluctuating data quality, intermittent performance, and unexpected downtime. This poses challenges when allocating losses accurately based on available data.
To tackle this challenge, we are excited to introduce a game-changing feature to our GPM Horizon’s Logbook module: multi-step waterfall loss allocation. This feature equips users with unprecedented flexibility to create highly customized, automated loss calculation workflows tailored to the specific conditions of individual turbines. By using a multi-step (cascading) approach, users can define a series of calculation methods that adapt based on available data and turbine performance. This ensures the automatic selection of the the best possible allocation method before checking it in the Logbook and making the necessary modifications to fine-tune it.
This article explores the benefits of this feature, its real-world applications, and the steps to configure it to suit your operational requirements. By the conclusion, you’ll have a solid grasp of how the multi-step waterfall approach can enhance data processing efficiency, boost accuracy, and save valuable time.
What is the multi-step waterfall approach?
The waterfall approach provides a hierarchical method of selecting loss calculation methods. When enabled, it follows a pre-defined sequence, assessing conditions at each step. If the conditions are met (or unmet), the system can either stop at that calculation method or move on to the next one. This ensures that the most appropriate method, based on the current data and turbine status, is always used.
If the waterfall toggle is turned off, the system uses a single, predefined loss calculation method without falling back on alternative methods. This straightforward approach may be appropriate for turbines with stable data streams or situations where data issues are minimal. However, for more complex environments, turning on the waterfall ensures that the system has a backup plan if issues arise.
Key benefits
Flexibility across turbines and contractual compliance
Wind turbines often experience varying conditions due to differences in location, weather patterns, or operational statuses. The multi-step waterfall loss allocation feature allows users to define distinct loss calculation methods for each turbine. No two turbines need to have the same loss calculation workflow, enabling tailored data analysis that reflects the unique conditions of each turbine.
For example, contracts may require certain turbines to have different primary neighboring WTGs selected as the main reference, as well as measurements from the anemometer, etc. In this case, the user can select the correct reference assets for each turbine and prioritize loss calculation methods that rely on more consistent data sources.
Precision and efficiency
By setting up a multi-step process, the system dynamically selects the best calculation method based on user-defined conditions. This cascading effect ensures that users always have the most accurate loss data available without manual input. By automating this process, operators can significantly reduce the risk of human error or oversight. When certain conditions, such as turbine performance or missing data, are not met, the system automatically shifts to the next most appropriate method. This ensures a steady stream of reliable data, which is essential for long-term trend analysis, budgeting, and performance forecasting.
Minimizing data gaps and inaccuracies
Intermittent or missing data is a common issue when working with renewable energy assets, especially wind turbines. Data gaps can result from network failures, sensor malfunctions, or downtime during maintenance, among others. With the multi-step waterfall loss allocation, the system can automatically handle these gaps by jumping to a fallback method (e.g., from reference turbine production to power curve-based estimation) whenever a data issue is detected. This ensures that the loss calculations remain uninterrupted and as accurate as possible, even when primary data sources are unavailable.
A closer look at the interface
The interface for the multi-step waterfall loss allocation is intuitive and easy to navigate. It consists of two key areas:
Main table with method summary
This panel provides a clear overview of the current configuration for each turbine in your portfolio. The columns include:
- Asset: turbine identifier (e.g., DM1 01).
- Default method: initial loss calculation method applied when waterfall is turned off or as the first step in the cascade.
- Reference units: nearby turbines or data sources used in the calculation.
- Waterfall: indicates whether the waterfall approach is enabled (Yes/No).
- Last edited: tracks when the configuration was last modified.
Waterfall method customization panel
The panel presents details for the selected turbine, including the default loss calculation method, reference turbines, and conditions. Here, users can configure the loss calculation workflow for each turbine. Each step of the waterfall process is clearly outlined, with options to:
- Add new steps: users can insert additional steps into the waterfall by clicking the “+ add” button.
- Set conditions: define conditions for each step (e.g., “if all turbines are not at full performance for 99% of the time block” or “if wind speed data is missing for 95% of the time”)
- Toggle methods: switch between loss calculation methods, such as reference turbine production, power curve, neighboring wind park data, and forecast theoretical production.
- Rearrange steps: drag-and-drop to change the order of the steps in the waterfall.
- Delete steps: eliminate steps if needed (at least 1 must remain selected).
Real-world use cases:
Handling inconsistent data
In some wind farms, data from the SCADA system may not always be reliable. For example, DM1 01 might experience frequent network interruptions, resulting in gaps in the data. By using the multi-step waterfall loss allocation, the system could first check SCADA data for loss allocation. If SCADA data is missing, the system can then switch to other methods.
For example, if there is no reliable data coming in at all, an alternative could be the forecasted production for the given period. This process ensures that loss data remains as accurate as possible, even in less-than-ideal data conditions.
Managing turbine downtime
Wind turbines often undergo maintenance or experience downtime, during which performance data is unavailable. The waterfall approach ensures that the system doesn’t rely on unavailable data.
For example, if a turbine is offline, the system might automatically calculate losses using data from nearby reference turbines that are still operational.
Ensuring compliance with performance guarantees in Power Purchase Agreements
Many wind farms operate under strict Power Purchase Agreements (PPAs) that require turbines to meet specific production targets or uptime guarantees. These agreements often stipulate penalties if the turbines fail to perform at a certain percentage of availability during specified periods.
Using the multi-step waterfall loss allocation, operators can configure the system to ensure that loss calculations are based on the most accurate data available, reducing the risk of incorrect calculations that could trigger penalties.
Meeting regulatory reporting requirements for downtime and curtailment
In some regions, operators are required to submit detailed reports to regulatory bodies on turbine downtime and curtailment events. Regulatory requirements may stipulate that specific data sources must be used to calculate production losses during these periods.
Adhering to O&M contractual obligations
Operations and Maintenance (O&M) contracts often include specific conditions related to turbine availability and performance.
For example, service providers may be contractually obligated to keep turbines operating for a minimum number of hours each month, with penalties for failure to meet these targets. These contracts often include provisions where turbine availability is measured based on different calculation methods, especially in the event of conflicting data sources.
Other potential scenarios where the waterfall loss allocation functionality could prove highly beneficial include:
- Handling insurance claims for revenue losses due to unplanned downtime.
- Navigating grid imbalance penalties.
- Accurate loss calculations for production-based incentives.
- Managing environmental and seasonal factors for more accurate loss allocation.
- Mitigating losses from market pricing conditions in hybrid assets.
Conclusion
The multi-step waterfall loss allocation feature offers tremendous value to asset managers and operators who demand precision and flexibility in loss calculations. This feature empowers users to define customized workflows for each wind turbine and dynamically adapt to real-time conditions, significantly improving data accuracy and operational efficiency.
Whether you’re dealing with inconsistent data, turbine downtime, or the complexities of managing large wind parks, the waterfall approach provides a robust and adaptive solution. Take advantage of this feature—explore GPM Horizon wind features now and customize your loss calculation methods for smarter, more reliable performance analysis.
Do you want to meet us and talk to our renewable energy experts?
We look forward to connecting with you and sharing our expertise in monitoring, control of renewable assets for utility-scale plants. Fill in the form to request a meeting with our team who will be available to answer questions, provide demonstrations, and offer insights on best practices.
