Solar operations intelligence from the Solarvyew team
Most SCADA systems expose every warning. Here is how to reduce inverter telemetry to signals that help separate likely hardware failure from normal operating variance.
Compare expected and actual generation to locate the problem. Analyze gaps by string, inverter, and weather window to see where and when losses occur.
A 2% soiling layer on panel glass cuts output by 3-5% and often escapes monitoring dashboards. Here is how generation data can reveal it when a technician cannot visit.
Drone thermography finds hotspots but costs $15-30K per campaign and offers only a snapshot. Continuous ML monitoring catches degradation as it develops. The tools serve different maintenance phases.
Clipping starts when panel output surpasses inverter capacity, usually at peak irradiance. Proper DC/AC ratio and setpoint tuning can recover 1-3% of annual generation without hardware changes.
PR is the standard KPI for solar plant health, but weather effects can hide brief faults and dilute them in monthly averages.
Comparing plant output without adjusting for irradiance, temperature, and soiling can create misleading trends. Simple normalization makes month over month comparisons meaningful.
Solar O&M is often reactive: an alarm sounds, then a technician is dispatched. Predictive monitoring reverses that order and reduces emergency response costs.
Asset managers overseeing several solar plants often reconcile SCADA reports instead of making decisions. Portfolio insight needs less data than most assume.
A 50 MW plant with 1% undetected soiling and one missed inverter fault per quarter loses between $40K and $90K in annual generation revenue.
MPPT algorithms respond differently to partial shade. Learn which inverter patterns indicate algorithm behavior rather than faults.
