Table 4 Detection results of non-weather-related anomalies in daily PV operation.

From: A novel hybrid model integrating CEEMDAN decomposition, dispersion entropy and LSTM for photovoltaic power forecasting and anomaly detection

Anomaly Type

Artificially Induced Anomalies

Detected Anomalies

Detection Rate

Line Failure

30

30

100.0%

Partial Shading

30

19

63.33%

PID Effect

30

27

90.00%

Back to article page