Behind-the-Meter Solar PV Forecasting in West Australia

05 June 2017

Behind the meter solar PV forecasting

Behind the meter solar PV power forecasting is an essential tool for accurate supply and demand balance for grid operators. And with rooftop PV installations in Western Australia once again surging for a second time during the past year or two (following first surge 2011/12), load forecasting is becoming increasingly difficult for our partner Western Power.

As our CTO has previously written about on the Western Power blog, Solcast is actively applying our rapid update solar forecasting systems to the behind-the-meter solar challenge in West Australia, through our Grid Aggregations tools.

I was interested to note our live SWIS (Southwest Interconnected System, the main WA electricity grid) grid-aggregate output estimates touching close to 500MW on sunny days last weekend, despite it being close to the winter solstice...

western power behind the meter solar PV forecasting Solcast.png

On these graphs we track the recent accuracy of our forecasts in the nowcasting range (minutes to a few hours ahead). These nowcasts track and model current cloud and irradiance every 10 minutes at 1km resolution. As you can see, correctly identifying a cloudy day can save about 300-400MW of error in the load forecast. Correctly identifying the timing and magnitude of cloud features on-the-day can save about 150-200MW of error.

Behind the Meter Solar PV Power Forecasting in Practice

This is an example of the Grid Aggregations tool in action for rooftop solar PV in West Australia. This tool is also useful in many other contexts, and has been put to use to:

  • Train and improve load forecasting models to include the impacts of behind-the-meter solar generation
  • Manage low voltage network demand forecasting, including planning for periods of minimum demand
  • Validate and tune models based on upscaling techniques or other surface data measurements
  • Manage spinning reserve or backup power supplies in tight demand situations or for remote grid applications
  • Dynamically control virtual power plants, including for the orchestration of low voltage network support services
  • Avoid the complexities of building in-house solutions, or to benchmark or improve existing in-house solutions

Setup a Grid Aggregation for Behind the Meter Solar PV in your network

You can get started on your first grid aggregation quickly and trial it by privately sharing very basic PV installation metadata (have you signed up for our API?). Get in touch with us for a sample upload file format, and to discuss your use case in more detail. Once your aggregation goes live in our system (usually within 2-3 days), we’ll provide you with unique forecasting and estimated actuals API endpoints for your evaluation.

James Luffman

James Luffman

Solcast CEO • Author

James is a former operational meteorologist, and has worked as a senior manager in the weather industry. James has designed, built and operated real-time modelling systems for industrial applications, bringing key expertise to the deployment of our operational products and services.

James sees the integration of increasing solar and storage as a singularly critical technology challenge of the next 10 years.