The optical performance of poly(methyl methacrylate) lenses from veteran concentrator photovoltaic modules was examined after the end of their service life. Lenses from the Martin-Marietta and Intersol module designs were examined from the “Solar Village” site near Riyadh, Saudi Arabia, as well as the Phoenix Sky Harbor airport, followed by the Arizona Public Service Solar Test and Research (APS-STaR) center in Tempe, Arizona. The various lens specimens were deployed for 20, 27, and 22 years, respectively. Optical characterizations included lens efficiency (Solar Simulator instrument), material transmittance and haze (of coupons cut from veteran lenses, then measured again after their faceted back surface was polished, and then measured again after the incident front surface was polished), and direct transmittance (as a function of detector's acceptance angle, using the Very Low Angular Beam Spread (“VLABS”) instrument). Lens efficiency measurements compared the central region to the entire lens, also using hot and cold mirror measurements to diagnose differences in performance. A series of subsequent characterizations was performed because a decrease in performance of greater than 10% was observed for some of the veteran lenses. The optimal focal distance of the lenses was quantified using the Solar Simulator, and then correlated to lens curvature using a recently developed measurement technique. Surface roughness was examined using atomic force microscopy and scanning electron microscopy. Facet geometry (tip and valley radius) was quantified on cross-sectioned specimens. Molecular weight was compared between the incident and faceted surfaces of the lenses
Anomaly detection is a critical problem in the manufacturing industry. In many applications, images of objects to be analyzed are captured from multiple perspectives which can be exploited to improve the robustness of anomaly detection. In this work, we build upon the deep support vector data description algorithm and address multi-perspective anomaly detection using three different fusion techniques, i.e., early fusion, late fusion, and late fusion with multiple decoders. We employ different augmentation techniques with a denoising process to deal with scarce one-class data, which further improves the performance (ROC AUC =80%). Furthermore, we introduce the dices dataset, which consists of over 2000 grayscale images of falling dices from multiple perspectives, with 5% of the images containing rare anomalies (e.g., drill holes, sawing, or scratches). We evaluate our approach on the new dices dataset using images from two different perspectives and also benchmark on the standard MNIST dataset. Extensive experiments demonstrate that our proposed multi-perspective approach exceeds the state-of-the-art single-perspective anomaly detection on both the MNIST and dices datasets. To the best of our knowledge, this is the first work that focuses on addressing multi-perspective anomaly detection in images by jointly using different perspectives together with one single objective function for anomaly detection.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.