The following is from the lab notes for the hands-on lab "Operational Forecasting" at OSIsoft Users Conference 2017, San Francisco, CA. Lab VM is available via OSIsoft Learning
The Lab manual is attached; the manual is intended for an instructor led interactive workshop in a classroom setting.
The lab's objective is to step through an end-to-end data science/machine learning task - collect data, publish historical data, develop a predictive model and deploy the model in real-time for wind turbine operations .
The predictive model is to forecast power generation for each turbine in our fleet as shown below
Figure shows a graph of Active Power vs. Time - actual power in purple and forecasted power in yellow.
The predictive model is based on forecasted wind speed and air temperature.
The tools used are:
- PI Integrator - publish historical turbine operations data to a SQL endpoint
- Power BI and its built-in support for R scripts - data munging, data diagnostics and exploring the features
- Azure ML - develop and deploy the model (as web services)
- Windows script (or, alternatively a .Net C# code via AF-SDK) is used to read/write forecast data to PI
Figure shows a graph of Active Power vs. Wind Speed from operations data.
For additional details, please see the Lab Manual.