Keeping models live
Just a heads-up—this section is still a work in progress! I’ll be revising and expanding it soon to make sure it’s as useful as possible. Curious about what’s already done or currently in the works? Check the changelog for updates.
A truly "live" model isn't just one that receives real-time data; it's one that intelligently uses that data to accurately reflect the current reality of your distribution system, which, as any operator knows, is constantly changing. This section explores the sophisticated techniques involved in state estimation and continuous model adaptation.
State estimation is the process where the model's current conditions — particularly parameters that aren't directly measured everywhere in real-time, like individual customer demands or effective pipe roughness — are automatically adjusted based on the incoming stream of live sensor data. The goal is to find the set of model inputs that makes the model's outputs (like calculated pressures and flows) best match the observed phenomena across your network.
Read more for free
Register or log in now to access additional chapters and unlock extra features.
By continuing, you agree to your personal information being collected under our privacy policy and accept our terms of use.