OpenOA: Open Operational Assessment
NREL's Open Operational Assessment (OpenOA) software helps the wind energy community assess the performance of operating wind plants.
The wind energy industry is growing as demand for renewable energy increases. But methods for assessing operational wind power plant performance are not standardized, which leads to inefficiencies and irregularity in operations and financial transactions. Efforts are duplicated, and a lack of transparency regarding how key performance metrics are measured creates uncertainty in those metrics, potentially resulting in disagreement between stakeholders.
To share resources and increase consistency across the industry, the OpenOA software, which launched in 2019, was developed to act as an open-source library for assessing operational wind plant performance by providing reference implementations of important operational data structures and analytics methods.
OpenOA, which is available in Python on GitHub, grew out of feedback from wind plant owners and consultants gathered through the U.S. Department of Energy's (DOE's) Wind Plant Performance Prediction Benchmark initiative and DOE's Energy I-Corps, a national laboratory training program. The OpenOA project has also received funding from the DOE Technology Commercialization Fund to integrate with the ENTR Foundation .
A first-of-its-kind resource, OpenOA is intended to provide a central hub for knowledge sharing and collaboration among wind energy industry researchers, wind plant owners and operators, and wind energy data analysts, helping to standardize operational analyses of wind power plants.
The tool acts as a central repository that contains a number of analysis methods, many of which implement International Electrotechnical Commission standards. With OpenOA, stakeholders can crowd source best practices and optimization strategies for wind farm research and development, which lowers the investment risk and encourages installation of more wind energy in the United States.
OpenOA consists of modules for organizing different types of data, low-level data analysis (e.g., filtering, power curve fitting) toolkits, and high-level operational assessment methods. Users can calculate the expected annual energy production over the lifetime of a wind farm, which, in the long run, can help wind plant operators identify and analyze the factors that drive wind plant performance, determine the accuracy of preconstruction energy yield estimates, and assist with financial transactions involving wind plants. The tool also enables users to estimate electrical losses, availability losses (losses caused when a turbine is not in operation, such as during maintenance), and the ideal energy production if all turbines are operating normally with built-in methods to help understand sources of gaps between preconstruction energy yield estimates and operational energy production. Examples that demonstrate these features are included in the GitHub repository.
Version 2.3, which was released in 2022, includes an improved data-quality-control toolkit, hourly energy production calculation, and automated date selection for long-term reference data. Another new feature, which was developed with the help of a wind plant owner, allows users to create interactive maps of wind farms.
With funding from DOE's Technology Commercialization Fund, the OpenOA team is working to create a data analysis software package that combines OpenOA with ENTR, which is an open-source wind plant data model and data engineering environment.
In the future, the development team plans to continue to add to OpenOA's capabilities, including allowing users to assess individual wind turbine performance and explore methods for quantifying the impact of wind turbine performance upgrades. Further improvements aim to add a wake-loss analysis feature and improve user experience by simplifying the data-importing process and allowing more data-source options.
More than 40 developers have modified OpenOA's code for their own purposes. DOE's Wind Plant Performance Prediction Benchmark initiative led to the development of OpenOA. The benchmark provided user feedback from eight major wind power plant owners and 10 third-party consultants, resulting in the Wind Plant Performance Prediction Benchmark Phase 1 technical report.
By integrating weather data from Intertrust's Planet OS API into Version 2.3 for a period of time in the spring of 2022, OpenOA achieved its first third-party collaboration on the NREL-developed platform.
The OpenOA team invites potential partners and users to contact them with feedback or suggestions or to contribute to the software through GitHub.
OpenOA: An Open-Source Codebase For Operational Analysis of Wind Farms, Journal of Open Source Software (2021)
Wind Plant Performance Prediction Benchmark Phase 1, Technical Report (2021)
Lowering Post-Construction Yield Assessment Uncertainty Through Better Wind Plant Power Curves, Wind Energy (2021)
Operational-Based Annual Energy Production Uncertainty: Are Its Components Actually Uncorrelated?, Wind Energy Science (2020)
Wind Energy EngineerMichael.Fields@nrel.gov