ABSTRACT
To achieve automated rock classification and improve classification accuracy, this work discusses an investigation of the combination of laser-induced breakdown spectroscopy (LIBS) and the use of one-dimensional convolutional neural networks (1DCNNs). As a result, in this paper, an improved Bayesian optimization (BO) algorithm has been proposed where the algorithm has been applied to automatic rock classification, using LIBS and 1DCNN to improve the efficiency of rock structure analysis being carried out. Compared to other algorithms, the improved BO method discussed here allows for a reduction of the modeling time by about 65% and can achieve 99.33% and 99.00% for the validation and test sets of 1DCNN.
ABSTRACT
Photovoltaic module measurements are predominantly taken by using pulsed solar simulators. However, significant errors can be generated when the existing simulators are applied to current high-efficiency crystalline silicon photovoltaic modules. This paper presents the design and implementation of a novel solar simulator featuring reflection-type light source and single long-pulse flash. The analysis and experimental study of the capacitance effect and the technical details of the simulator including reflection-type lamp house, xenon flash lamp power supply, and source-measure unit are introduced. The results show that the complete system achieves Class AAA performance in accordance with the international standard. The proposed simulator outperforms other similar products on the market and has been adopted by some well-known photovoltaic module manufacturers. The practical application demonstrates that this high-performance and cost-effective simulator is quite suitable for photovoltaic module production line.
