Search

This research proposes a low-cost system consisting of a hardware setup and a deep learning-based model to estimate broiler chickens' feed intake, utilizing audio signals captured by piezoelectric sensors. The signals were recorded 24/7 for 19 consecutive days. A subset of the raw data was chosen, and events were labeled in two classes, feed-pecking and non-pecking (including singing, anomaly, and silence samples). Next, the labeled data were preprocessed through a noise removal algorithm and a band-pass filter. Then, the spectrogram and the signal envelope were extracted from each signal and fed as inputs to a VGG-16-based convolutional neural network (CNN) with two branches for 1D and 2D feature extraction followed by a binary classification head to classify feed-pecking and non-pecking events. The model achieved 92% accuracy in feed-pecking vs. non-pecking events classification with an f1-score of 91%. Finally, the entire raw dataset was processed utilizing the developed model, and the resulting feed intake estimation was compared with the ground truth data from scale measures. The estimated feed consumption showed an 8 ± 7% mean percent error on daily feed intake estimation with a 71% R2 score and 85% Pearson product moment correlation coefficient (PPMCC) on hourly intake estimation. The results demonstrate that the proposed system estimates broiler feed intake at each feeder and has the potential to be implemented in commercial farms. [ABSTRACT FROM AUTHOR]

Re is often used as a nonradioactive surrogate for Tc in borosilicate melts because both elements have similar valences, ionic radii, and coordination environments. Re solubility in high sodium borosilicate waste glasses is sensitive to melter redox conditions; at 1150°C the solubility is ∼4130 ppm under oxidizing atmospheres. Raman, X‐ray absorption spectroscopy, and X‐ray diffraction show that Re dissolved in waste glasses under oxidizing conditions is in perrhenate tetrahedra; under more reducing conditions, Re0 separates from the melt. From air to PO2$ {\rm P}_{{\rm O}_2} $∼ 10−4 atm, both Re7+ and Tc7+ are dominant species in borosilicate melts. However, Tc and Re behaviors deviate between PO2$ {\rm P}_{{\rm O}_2} $< 10−4 atm and PO2$ {\rm P}_{{\rm O}_2} $> 10−8 atm, where Tc4+ becomes dominant, whereas Re7+ remains stable. Unlike Tc4+, Re4+ is not observed in borosilicate melts at 1150°C and associated glasses. Near PO2$ {\rm P}_{{\rm O}_2} $= 10−8 atm, Tc4+ reduces to Tc0, whereas below PO2$ {\rm P}_{{\rm O}_2} $= 10−8 atm, Re7+ reduces to Re0. Re is found to be a good surrogate for Tc under oxidizing conditions (PO2$ {\rm P}_{{\rm O}_2} $> 10−4 atm) and under highly reducing conditions (PO2$ {\rm P}_{{\rm O}_2} $< 10−9 atm). Redox and solubility behaviors of Tc and Re are also sensitive to glass chemistry, such that the fair assessment of Re modeling Tc behavior should be glass specific. [ABSTRACT FROM AUTHOR]