Efficient Unsupervised Dimension Reduction for Streaming Multiview Data

With the installation and use of large-scale photovoltaic systems around the world, the detection of photovoltaic system operation and maintenance has become increasingly important. This research uses a convolutional neural network training model to detect and classify the infrared near-field images of photovoltaic modules from small-scale photovoltaic plants in the laboratory. This model classifies the images into two categories: with and without hot spots, with a classification accuracy of 96.58%. The experimental results show that the convolutional neural network training model has a good classification result

Section: Introductionmentioning

confidence: 99%

Hot Spot Detection of Photovoltaic Module Infrared Near-field Image based on Convolutional Neural Network

Fan

Zhang

et al. 2022

“…Since every photovoltaic module has a bypass diode in parallel [12], the output power curve of the photovoltaic array is multi-peak in the case of partial shadow occlusion, which leads to the local optimal value easily falling into the traditional MPPT method. Therefore, it is particularly important to study MPPT in the shadow state for the stable and efficient operation of photovoltaic power generation [13][14][15].…”

Section: Introductionmentioning

confidence: 99%

Maximum power point tracking of photovoltaic array based on improved Particle Swarm Optimization Algorithm

Lai

Zhang

et al. 2022

With the vigorous development of the photovoltaic industry, how to improve the efficiency of photovoltaic power generation has become an important issue, among which partial shadow occlusion is an important reason affecting the efficiency. The efficiency of photovoltaic power generation can be effectively improved by adopting the maximum power point tracking method (MPPT), but the traditional MPPT method is not ideal in the partial shadow occlusion of the photovoltaic array. To solve this problem, this paper proposes an improved particle swarm optimization method to effectively improve the tracking efficiency of MPPT when multiple peaks appear in the photovoltaic arrays power curve (P-V) under the partial shadow. The proposed method improves the learning factor of the traditional particle swarm optimization algorithm and designs the initial position of the particles according to the characteristics of the photovoltaic array. By adding the particle elimination mechanism, the number of particles changes dynamically, and the tracking speed of the algorithm for the maximum power of the photovoltaic array is improved. Through the result of the simulation, it is not difficult to get the conclusion that the improved particle swarm optimization algorithm can effectively improve the performance of the photovoltaic system under partial shadows.

“…In order to solve these problems, we designed a set of EL defect detection system based on photovoltaic power station modules, and carried the designed online defect detection algorithm [10] on the integrated device to run. The system applies the EL micro-cracks detection to the actual photovoltaic power station, which can detect the defects of photovoltaic modules in real time based on the string level, which improves the detection efficiency and saves time and cost.…”

Section: Introductionmentioning

confidence: 99%

Design of EL defect detection system for photovoltaic power station modules

Fan

Yan

et al. 2022

In recent years, the photovoltaic power generation industry has been vigorously promoted and developed, while the solar cell as its core component may have micro-crack defects, which directly affect the power generation efficiency and stability of the photovoltaic system. Therefore, it is necessary to adopt a low-cost, efficient and flexible method to detect defects in solar cells. At present, most of the existing micro-cracks detection is carried out in the laboratory, and there is no EL micro-cracks detection in the actual photovoltaic power station. The main purpose of this paper is to design a set of EL defect detection system that can be used for actual photovoltaic power station modules, which is different from the traditional laboratory-level or module-level defect detection, and to carry the designed detection algorithm to the integrated device to operate. Experiments have shown that the system can perform real-time detection of photovoltaic module defects based on the string level, which improves the detection efficiency while saving time and cost, and shows good performance.