Examples

All notebooks are located at /examples in the OpenOA repository, and can be modified and run on Binder.

Important

Be sure to install OpenOA using the examples modifier (pip install "openoa[examples]"). This will install all of the dependencies used by the additional methods contained within the examples notebooks.

Overview

In each of the following examples we’ll be providing more insight into the different functionalities of OpenOA. In each notebook, OpenOA is demonstrated using two years of operational data for the La Haute Borne wind power plant from the ENGIE open data set (https://opendata-renewables.engie.com). The examples start by introducing the openoa.plant.PlantData class and quality assurance methods in the openoa.utils.qa utils module, illustrating how a PlantData object is created for the La Haute Borne data set (the rest of the examples use this PlantData object to demonstrate OpenOA analysis and utils methods). Next, several utils module use cases are demonstrated, such as power curve fitting and plotting. Three analysis methods are then demonstrated: MonteCarloAEP (long-term AEP analysis), TurbineLongTermGrossEnergy (turbine ideal energy), and ElectricalLosses. Next, the EYAGapAnalysis class is used to perform a gap analysis using the estimated operational long-term AEP, turbine ideal energy, electrical losses, and availability losses together with corresponding example pre-construction estimates. Two additional analysis methods are then demonstrated: the WakeLosses method for estimating operational wind plant and wind turbine-level wake losses, and the StaticYawMisalignment method for estimating static yaw misalignment for individual wind turbines as a function of wind speed.

Intro to OpenOA PlantData and the QA Methods [link]

• Reiterates some of the essential concepts of the of the openoa.plant.PlantData and openoa.plant.PlantMetaData classes

• Shows how we formulated the examples/project_ENGIE.py data cleaning and loading scripts

• Highlights the QA methods available in openoa.utils.qa and how they work on real data

Demonstrating the Utils With the ENGIE Open Data [link]

• Walks through the use of various plotting and analysis methods

• Introduces some of the building blocks of how analyses are composed

Gap Analysis Step 1a: Estimate the AEP and Its Uncertainty [link]

• Introduces the annual energy production (AEP) class, and how to estimate the uncertainty using a Monte Carlo approach

• Demonstrates some of the supplementary tools to analysis, such as the plotting routines, and how to customize them, and how to use them through the openoa.utils.plot interface

Gap Analysis Step 1b: Estimate the AEP and Its Uncertainty Using Cubico Open Data [link]

Important

Be sure to install OpenOA using the examples and reanalysis modifiers for this notebook (pip install "openoa[examples,reanalysis]").

• Introduces the annual energy production (AEP) class, and how to estimate the uncertainty using a Monte Carlo approach

• Demonstrates some of the supplementary tools to analysis, such as the plotting routines, and how to customize them, and how to use them through the openoa.utils.plot interface

Gap Analysis Step 1c: Alternative Methods for Calculating the AEP [link]

• Building from the previous example, the augmented capabilities for calculating AEP using a Monte Carlo framework for calculating the AEP are demonstrated

• Demonstrates how to change the regression model and additional variables that can be considered for analyses

Gap Analysis Step 2: Calculate the Turbine Ideal Energy [link]

• Introduces the turbine long term gross energy estimation workflow

• Demonstrates some of the supplementary tools to analysis, such as the plotting routines, and how to customize them, and how to use them through the openoa.utils.plot interface

Gap Analysis Step 3: Estimate Electrical Losses [link]

• Introduces the electrical losses analysis workflow

• Demonstrates some of the supplementary tools to analysis, such as the plotting routines, and how to customize them, and how to use them through the openoa.utils.plot interface

Gap Analysis Step 4: Compare the Energy Yield Assessment to the Operational Assessment (Gap Analysis) [link]

• Ties together the previous examples to estimate energy production and potential losses

• Demonstrates some of the supplementary tools to analysis, such as the plotting routines, and how to customize them, and how to use them through the openoa.utils.plot interface

Estimate Operational Wake Losses [link]

• Introduces the operational wake loss estimation class and workflow

• Demonstrates the estimation of wake losses based on turbine-level SCADA data during the period of record as well as the long-term corrected wake losses incorporating historical reanalysis wind resource data

• Illustrates the estimation of wake losses at the wind plant level as well as for each wind turbine with and without uncertainty quantification

• Demonstrates methods for plotting wake losses as a function of wind direction and wind speed

Estimate Yaw Misalignment [link]

• Introduces the static yaw misalignment estimation class and workflow

• Demonstrates the estimation of yaw misalignment based on turbine-level SCADA data during the period of record

• Illustrates the estimation of wake losses for each wind turbine with and without uncertainty quantification

• Demonstrates methods for plotting yaw misalignment as a function of wind vane angle and normalized power

Table of Contents

• Intro to the OpenOA PlantData and QA Methods
  • The PlantData object
    • What’s changed in v3?
    • PlantData and PlantMetaData 101
  • Imports
  • QA’ing ENGIE’s open data set
    • Step 1: Load the SCADA data
    • Step 2: Convert the timestamps to proper timestamp data objects
    • Step 3: Dive into the data
      • Inspecting the distributions of each column of numerical data
      • Checking the power curve distributions
    • Step 4: Inspecting the timestamps for DST gaps and duplications
      • Timezone-unaware data
      • Timezone-aware data
  • Summarizing the QA process into a reproducible workflow
  • PlantData demonstration
    • PlantData documentation
    • PlantMetaData Documentation
    • PlantData validation
    • PlantData string and markdown representations
• Demonstrating the Utils With the ENGIE Open Data
  • Import the data
  • Range filter
  • Window range filter
  • Bin filter
  • Unresponsive Filter
  • Power curve fitting
• Gap Analysis Step 1a: Estimate the AEP and Its Uncertainty
  • Step 1: Import the data
  • Step 2: Review the data
  • Step 3: Process the data into monthly averages and sums
  • Step 4: Review reanalysis data
  • Step 5: Review energy and loss data
  • Step 6: Specify availabilty and curtailment data not represenative of actual plant performance
  • Step 7: Select reanalysis products to use
  • Step 8: Set up Monte Carlo inputs
  • Step 9: Run the OA code
  • Step 10: Post-analysis visualization
• Gap Analysis Step 1b: Estimate the AEP and Its Uncertainty Using Cubico Open Data
  • Step 1: Import the data
  • Step 2: Review the data
  • Step 3: Process the data into monthly averages and sums
  • Step 4: Review reanalysis data
  • Step 5: Review energy and loss data
  • Step 6: Specify availabilty and curtailment data not represenative of actual plant performance
  • Step 7: Select reanalysis products to use
  • Step 8: Set up Monte Carlo inputs
  • Step 9: Run the OA code
  • Step 10: Post-analysis visualization
• Gap Analysis Step 1c: Alternative Methods for Calculating the AEP
  • Comparison 1: AEP calculation using different regression models and different time resolution
    • Results Stability and Number of Simulations
  • Comparison 2: AEP calculation using various input variables
• Gap Analysis Step 2: Calculate the Turbine Ideal Energy
  • TIE calculation without uncertainty quantification
  • TIE calculation including uncertainty quantification
• Gap Analysis Step 3: Estimate Electrical Losses
  • Electrical loss estimation without uncertainty quantification
  • Electrical loss estimation including uncertainty quantification
• Gap Analysis Step 4: Compare the Energy Yield Assessment to the Operational Assessment (Gap Analysis)
  • Calculate AEP, TIE, and electrical losses from operational data
  • Summarize operational results and create EYA data for comparison
  • Perform gap analysis
• Estimate Operational Wake Losses based on SCADA Data
  • Import the data
  • Investigate and correct wind direction data
    • Identify changes in wind direction northing offset for each turbine
    • Identify any wind direction biases between turbine pairs
    • Determine the northing offset and correct wind direction signals for each turbine
  • Wake loss estimation without uncertainty quantification
    • Analysis of wind plant wake losses
    • Analysis of turbine-level wake losses
  • Wake loss estimation with uncertainty quantification
    • Analysis of wind plant wake losses
    • Analysis of turbine-level wake losses
• Estimate Static Yaw Misalignment based on SCADA Data
  • Import the data
  • Examine Power Curve Filtering Parameters
  • Yaw misalignment detection without uncertainty quantification
  • Yaw misalignment detection with uncertainty quantification