No abstract

In this paper, a fault detection algorithm for photovoltaic systems based on artificial neural networks (ANN) is proposed. Numerous literatures can be found on the topic of PV fault detection through the implementation of artificial intelligence. The novel part of this research is the successful development, deployment and validation of a fault detection PV system using radial basis function (RBF), requiring only two parameters as the input to the ANN (solar irradiance and output power). The results obtained through the testing of the developed ANN on a PV installation of 2.2 kW capacity, provided an accuracy of 97.9%. To endorse the accuracy of the newly developed algorithm, the ANN was tested on another PV system, installed at a remote location.The total capacity of the new system was significantly higher, 4.16 kW. A vital part of the test was to see how the proposed ANN would perform with 'scaled-up' input data, during normal operation as well as partial shading scenarios. The validation process provided an overall fault detection accuracy of above 97%. The decrease in accuracy was due to the varying nature of the two systems in terms of total capacity, number of samples and type of faults.

The first processing step in synchrotron-based micro-tomography is the normalization of the projection images against the background, also referred to as a white field. Owing to time-dependent variations in illumination and defects in detection sensitivity, the white field is different from the projection background. In this case standard normalization methods introduce ring and wave artefacts into the resulting three-dimensional reconstruction. In this paper the authors propose a new adaptive technique accounting for these variations and allowing one to obtain cleaner normalized data and to suppress ring and wave artefacts. The background is modelled by the product of two timedependent terms representing the illumination and detection stages. These terms are written as unknown functions, one scaled and shifted along a fixed direction (describing the illumination term) and one translated by an unknown two-dimensional vector (describing the detection term). The proposed method is applied to two sets (a stem Salix variegata and a zebrafish Danio rerio) acquired at the parallel beam of the micro-tomography station 2-BM at the Advanced Photon Source showing significant reductions in both ring and wave artefacts. In principle the method could be used to correct for time-dependent phenomena that affect other tomographic imaging geometries such as cone beam laboratory X-ray computed tomography.

The Lost City hydrothermal field (LCHF) is hosted in serpentinite at the crest of the Atlantis Massif, an oceanic core complex close to the mid-Atlantic Ridge. It is remarkable for its longevity and for venting low-temperature (40-91°C) alkaline fluids rich in hydrogen and methane. IODP Hole U1309D, 5 km north of the LCHF, penetrated 1415 m of gabbroic rocks and contains a near-conductive thermal gradient close to 100°C km À1 . This is remarkable so close to an active hydrothermal field. We present hydrothermal modelling using a topographic profile through the vent field and IODP site U1309. Long-lived circulation with vent temperatures similar to the LCHF can be sustained at moderate permeabilities of 10 À14 to 10 À15 m 2 with a basal heatflow of 0.22 W m À2 .Seafloor topography is an important control, with vents tending to form and remain in higher topography. Models with a uniform permeability throughout the Massif cannot simultaneously maintain circulation at the LCHF and the near-conductive gradient in the borehole, where permeabilities <10 À16 m 2 are required. A steeply dipping permeability discontinuity between the LCHF and the drill hole is required to stabilize venting at the summit of the massif by creating a lateral conductive boundary layer. The discontinuity needs to be close to the vent site, supporting previous inferences that high permeability is most likely produced by faulting related to the transform fault. Rapid increases in modelled fluid temperatures with depth beneath the vent agree with previous estimates of reaction temperature based on geochemical modelling.

Heart disease, caused by low heart rate, is one of the most significant causes of mortality in the world today. Therefore, it is critical to monitor heart health by identifying the deviation in the heart rate very early, which makes it easier to detect and manage the heart’s function irregularities at a very early stage. The fast-growing use of advanced technology such as the Internet of Things (IoT), wearable monitoring systems and artificial intelligence (AI) in the healthcare systems has continued to play a vital role in the analysis of huge amounts of health-based data for early and accurate disease detection and diagnosis for personalized treatment and prognosis evaluation. It is then important to analyze the effectiveness of using data analytics and machine learning to monitor and predict heart rates using wearable device (accelerometer)-generated data. Hence, in this study, we explored a number of powerful data-driven models including the autoregressive integrated moving average (ARIMA) model, linear regression, support vector regression (SVR), k-nearest neighbor (KNN) regressor, decision tree regressor, random forest regressor and long short-term memory (LSTM) recurrent neural network algorithm for the analysis of accelerometer data to make future HR predictions from the accelerometer’s univariant HR time-series data from healthy people. The performances of the models were evaluated under different durations. Evaluated on a very recently created data set, our experimental results demonstrate the effectiveness of using an ARIMA model with a walk-forward validation and linear regression for predicting heart rate under all durations and other models for durations longer than 1 min. The results of this study show that employing these data analytics techniques can be used to predict future HR more accurately using accelerometers.