An interpretable AI framework for clear-sky detection in photovoltaics monitoring

Special Issue on ‘EU PVSEC 2025: State of the Art and Developments in Photovoltaics', edited by Robert Kenny and Carlos del Cañizo

Open Access

Table 2

Glossary of performance evaluation metrics used in this study. TP, TN, FP, and FN denote true positives, true negatives, false positives, and false negatives, respectively.

Metric	Description	Equation
Accuracy	Overall proportion of correctly classified samples among all samples.	$\frac{T P + T N}{T P + T N + F P + F N} \cdot 100 %$
Precision	Proportion of true positive cases among all samples predicted as positive.	$\frac{T P}{T P + F P} \cdot 100 %$
Recall	Proportion of true positive cases correctly identified among all actual positive samples.	$\frac{T P}{T P + F N} \cdot 100 %$
False Positive Rate (FPR)	Proportion of actual non-clear-sky instances incorrectly classified as clear-sky.	$\frac{F P}{F P + T N} \cdot 100 %$
False Negative Rate (FNR)	Proportion of actual clear-sky instances that the model failed to identify.	$\frac{F N}{F N + T P} \cdot 100 %$
False Positive Percentage	Number of false positives as a percentage of the total number of samples.	$\frac{F P}{T P + T N + F P + F N} \cdot 100 %$
False Negative Percentage	Number of false negatives as a percentage of the total number of samples.	$\frac{F N}{T P + T N + F P + F N} \cdot 100 %$

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.