Peijie Lin

A Density Peak-Based Clustering Approach for Fault Diagnosis of Photovoltaic Arrays

Fault diagnosis of photovoltaic (PV) arrays plays a significant role in safe and reliable operation of PV systems. In this paper, the distribution of the PV systems’ daily operating data under different operating conditions is analyzed. The results show that the data distribution features significant nonspherical clustering, the cluster center has a relatively large distance from any points with a higher local density, and the cluster number cannot be predetermined. Based on these features, a density peak-based clustering approach is then proposed to automatically cluster the PV data. And then, a set of labeled data with various conditions are employed to compute the minimum distance vector between each cluster and the reference data. According to the distance vector, the clusters can be identified and categorized into various conditions and/or faults. Simulation results demonstrate the feasibility of the proposed method in the diagnosis of certain faults occurring in a PV array. Moreover, a 1.8 kW grid-connected PV system with array is established and experimentally tested to investigate the performance of the developed method.

A Population Classification Evolution Algorithm for the Parameter Extraction of Solar Cell Models

To quickly and precisely extract the parameters for solar cell models, inspired by simplified bird mating optimizer (SBMO), a new optimization technology referred to as population classification evolution (PCE) is proposed. PCE divides the population into two groups, elite and ordinary, to reach a better compromise between exploitation and exploration. For the evolution of elite individuals, we adopt the idea of parthenogenesis in nature to afford a fast exploitation. For the evolution of ordinary individuals, we adopt an effective differential evolution strategy and a random movement of small probability is added to strengthen the ability to jump out of a local optimum, which affords a fast exploration. The proposed PCE is first estimated on 13 classic benchmark functions. The experimental results demonstrate that PCE yields the best results on 11 functions by comparing it with six evolutional algorithms. Then, PCE is applied to extract the parameters for solar cell models, that is, the single diode and the double diode. The experimental analyses demonstrate that the proposed PCE is superior when comparing it with other optimization algorithms for parameter identification. Moreover, PCE is tested using three different sources of data with good accuracy.

Motion Detection System Based on Improved LBP Operator

A fast and reliable motion detection algorithm is very important to an intelligent surveillance system. Local Binary Pattern (LBP) is one of powerful texture description and comparison mechanisms, but in contrast consumes a large portion of computational time in a CPU based system. In this paper, we propose a moving object detection algorithm based on the improved LBP operator which is tolerant against pixel noise. Combining the background subtraction algorithm and the frame difference algorithm, the automatic refreshing of the background is realized. The moving object detection system which can achieve real time processing of a 1024 × 768/60 Hz VGA signal is designed on a PFGA chip and all the algorithms are mapped to hardware logic. ROC curves of the experiments demonstrate that in the condition with shifty illumination, the algorithm based on LBP operator has a better performance than the algorithm based on grayscales.

Orderless and Blurred Visual Tracking via Spatio-temporal Context

In this paper, a novel and robust method which exploits the spatiotemporal context for orderless and blurred visual tracking is presented. This lets the tracker adapt to both rigid and deformable objects on-line even if the image is blurred. We observe that a RGB vector of an image which is resized into a small fixed size can keep enough useful information. Based on this observation and computational reasons, we propose to resize the windows of both template and candidate target images into 2×2 and use Euclidean Distance to compute the similarity between these two RGB image vectors for the preliminary screening. We then apply spatio-temporal context based on Bayesian framework to further compute a confidence map for obtaining the best target location. Experimental results on challenging video sequences in MATLAB without code optimization show the proposed tracking method outperforms eight state-of-the-art methods.

On-line Monitoring and Fault Diagnosis of PV Array Based on BP Neural Network Optimized by Genetic Algorithm

The vast majority of photo voltaic (PV) arrays often work in harsh outdoor environment, and undergo various fault, such as local material aging, shading, open circuit, short circuit and so on. The generation of these fault will reduce the power generation efficiency, and even lead to fire disaster which threaten the safety of social property. In this paper, an on-line distributed monitoring system based on ZigBee wireless sensors network is designed to monitor the output current, voltage and irradiate of each PV module, and the temperature and the irradiate of the environment. A simulation PV module model is established, based on which some common faults are simulated and fault training samples are obtained. Finally, a genetic algorithm optimized Back Propagation (BP) neural network fault diagnosis model is built and trained by the fault samples data. Experiment result shows that the system can detect the common faults of PV array with high accuracy.

Optimal Data Collection of MP-MR-MC Wireless Sensors Network for Structural Monitoring

Structural health monitoring (SHM) is a kind of data-intensive applications for wireless sensors networks, which usually requires high network bandwidth. However, the bandwidth of traditional single-radio single-channel (SR-SC) WSN is quite limited. In order to meet the requirement of structural monitoring, multi-radio multi-channel (MR-MC) WSN is emerging. In this paper, we address the optimal data collection problem in MR-MC WSN by modelling it as an integer linear programming problem. Combining the advantages of the particle swarm optimization (PSO) algorithm and flower pollination optimization (FPA) algorithm, we propose a new hybrid algorithm BFPA-PSO to solve the optimization problem under the constraint of time slot and multi-power multi-radio multi-channel (MP-MR-MC). Theoretical analysis and simulation experiments are carried out and the results show that the proposed method has good performance in improving network capacity as well as reducing energy consumption.

Automated Pixel-Level Surface Crack Detection Using U-Net

Crack detection is significant for the inspection and diagnosis of concrete structures. Various automated approaches have been developed to replace human-conducted inspection, many of which are not adaptive to various conditions and unable to provide localization information. In this paper, an end-to-end semantic segmentation neural network based on U-net is employed to detect crack. Due to the limited number of available annotated samples, data augmentation is employed to avoid overfitting. The adopted network is trained by only 200 images of 512 \(\times \) 512 pixels resolutions and achieves a satisfactory accuracy of 99.56% after 37 epochs. The output is an image of the same size as the input image where each pixel is assigned a class label, i.e. crack or not crack. It takes about 7 s to process an image of designed size on CPU. Combined with sliding window technique, our model can cope with any image of larger size. Comparative experiment results show that our model outperforms traditional Canny and Sobel edge detection methods in a variety of complex environment without extracting features manually.

One-Dimensional Camera Calibration Based on PSO Algorithm

Camera calibration is an essential process in visual measurement. 1D target based camera calibration can great facilitate the operating procedure especially when multiple vision sensors should be calibrated. However, the current one-dimensional calibration algorithm is still imprecision in practice. In this work, the PSO algorithm is employed to improve the precision of one-dimensional camera calibration. Since the swarm intelligence algorithm is initial value sensitive, in this work, a data cluster algorithm is proposed to get a better initial value. To overcome the over optimizing problem accounted in swarm intelligence algorithm, prior knowledge, such as the picture’s size, is employed to make sure the parameters will converge toward the true values.

On-line monitoring system of PV array based on internet of things technology

The Internet of Things (IoT) Technology is used to inspect photovoltaic (PV) array which can greatly improve the monitoring, performance and maintenance of the PV array. In order to efficiently realize the remote monitoring of PV operating environment, an on-line monitoring system of PV array based on IoT is designed in this paper. The system includes data acquisition, data gateway and PV monitoring centre (PVMC) website. Firstly, the DSP-TMS320F28335 is applied to collect indicators of PV array using sensors, then the data are transmitted to data gateway through ZigBee network. Secondly, the data gateway receives the data from data acquisition part, obtains geographic information via GPS module, and captures the scenes around PV array via USB camera, then uploads them to PVMC website. Finally, the PVMC website based on Laravel framework receives all data from data gateway and displays them with abundant charts. Moreover, a fault diagnosis approach for PV array based on Extreme Learning Machine (ELM) is applied in PVMC. Once fault occurs, a user alert can be sent via E-mail. The designed system enables users to browse the operating conditions of PV array on PVMC website, including electrical, environmental parameters and video. Experimental results show that the presented monitoring system can efficiently real-time monitor the PV array, and the fault diagnosis approach reaches a high accuracy of 97.5%.

A grid-connected single-phase photovoltaic micro inverter

In this paper, the topology of a single-phase grid-connected photovoltaic (PV) micro-inverter is proposed. The PV micro-inverter consists of DC-DC stage with high voltage gain boost and DC-AC conversion stage. In the first stage, we apply the active clamp circuit and two voltage multipliers to achieve soft switching technology and high voltage gain. In addition, the flower pollination algorithm (FPA) is employed for the maximum power point tracking (MPPT) in the PV module in this stage. The second stage cascades a H-bridge inverter and LCL filter. To feed high quality sinusoidal power into the grid, the software phase lock, outer voltage loop and inner current loop control method are adopted as the control strategy. The performance of the proposed topology is tested by Matlab/Simulink. A PV module with maximum power 300W and maximum power point voltage 40V is applied as the input source. The simulation results indicate that the proposed topology and the control strategy are feasible.