The Article titled "PrOuD: Probabilistic Outlier Detection Solution for Time Series Analysis on Real-world Photovoltaic Inverters" by Yujiang He, Zhixin Huang, Stephan Vogt and Bernhard Sick was published in the journal Energies (MDPI). This publication addresses the opacity of anomaly detection methods in time series analysis, especially in real-world applications, and proposes a mathematical analysis of anomalies and innovations in multivariate time series. The study introduces PrOuD, a solution developed for interpretable detection results that uses Monte Carlo estimation to convert prediction uncertainty into estimated outlier uncertainty. Experimental results with artificial and real photovoltaic inverter data demonstrate the accuracy of PrOuD in detecting emerging anomalies and provide professionals with a reliable tool for efficient time series diagnosis and clustering of anomalous patterns.

The Article titled "PrOuD: Probabilistic Outlier Detection Solution for Time Series Analysis on Real-world Photovoltaic Inverters" by Yujiang He, Zhixin Huang, Stephan Vogt and Bernhard Sick was published in the journal Energies (MDPI). This publication addresses the opacity of anomaly detection methods in time series analysis, especially in real-world applications, and proposes a mathematical analysis of anomalies and innovations in multivariate time series. The study introduces PrOuD, a solution developed for interpretable detection results that uses Monte Carlo estimation to convert prediction uncertainty into estimated outlier uncertainty. Experimental results with artificial and real photovoltaic inverter data demonstrate the accuracy of PrOuD in detecting emerging anomalies and provide professionals with a reliable tool for efficient time series diagnosis and clustering of anomalous patterns.