Dynamically Typed

SunDown

SunDown is a “a sensorless approach designed to detect per-panel faults in residential solar arrays” by Feng et al. (2020). Trained on years of solar generation data from homes, it “leverages correlations between the power produced by adjacent panels to detect deviations from expected behavior,” detecting faults and electrical failures with > 99% accuracy. This also sounds like an app waiting to happen!