Control
Mohsen Hamzeh; Sepehr Shakibi; Amir Mohammad Farahani
Abstract
Photovoltaic (PV) systems are a backbone of the infrastructure of renewable energy with its usage growing significantly. Early fault detection of these systems being essential to enhance their reliability and efficiency. Despite the development of fault diagnosis methods of PV system promoted by ...
Read More
Photovoltaic (PV) systems are a backbone of the infrastructure of renewable energy with its usage growing significantly. Early fault detection of these systems being essential to enhance their reliability and efficiency. Despite the development of fault diagnosis methods of PV system promoted by machine learning models such as ensemble learning, support vector machine and neural networks, challenges in achieving high accuracy and generalization persist. In this paper, propose a deep learning method based on a ResNet architecture for reliable and efficient fault detection, including the following categories: Normal Operation, Short-Circuit, Degradation, Open Circuit, and Shadowing. Also devise a new learning rate schedule(LRS), which considerably improves the training dynamics and enables a 63% improvement in model performance. The suggested method has excellent performance achieves 99.8% accuracy throughout the training, validation and testing phases. The results obtained showcase the potential of ResNet-based architectures, in addition to prowess in adaptive learning rate strategies, at enhancing the reliability of photovoltaic systems through scalable and precise fault diagnosis.
Optimization
Javad Farzaneh; Ali Karsaz
Abstract
Maximum Power Point Tracking (MPPT) is an important concept for both uniform solar irradiance and Partial Shading Conditions (PSCs). The paper presents an Improved Salp Swarm Algorithm (ISSA) for MPPT under PSCs. The proposed method benefits a fast convergence speed in tracking the Maximum Power Point ...
Read More
Maximum Power Point Tracking (MPPT) is an important concept for both uniform solar irradiance and Partial Shading Conditions (PSCs). The paper presents an Improved Salp Swarm Algorithm (ISSA) for MPPT under PSCs. The proposed method benefits a fast convergence speed in tracking the Maximum Power Point (MPP), in addition to overcoming the problems of conventional MPPT methods, such as failure to detect the Global MPP (GMPP) under PSCs, getting trapped in the local optima, and oscillations around the MPP. The proposed method is compared with original algorithms such as Perturbation and Observation (P&O) method (which is widely employed in MPPT applications), Differential Evolutionary (DE) algorithm, Particle Swarm Optimization (PSO), and Firefly Algorithm (FA). The obtained results show that the proposed method can detect and track the MPP in a very short time, and its accuracy outperforms the other methods in terms of detecting the GMPP. The proposed ISSA algorithm has a higher speed and the convergence rate than the other traditional algorithms.
Optimization
Javad Farzaneh; Reza Keypour; Ali Karsaz
Abstract
It is highly expected that partially shaded condition (PSC) occurs due to the moving clouds in a large photovoltaic (PV) generation system (PGS). Several peaks can be seen in the P-V curve of a PGS under such PSC which decreases the efficiency of conventional maximum power point tracking (MPPT) methods. ...
Read More
It is highly expected that partially shaded condition (PSC) occurs due to the moving clouds in a large photovoltaic (PV) generation system (PGS). Several peaks can be seen in the P-V curve of a PGS under such PSC which decreases the efficiency of conventional maximum power point tracking (MPPT) methods. In this paper, an adaptive neuro-fuzzy inference system (ANFIS) is proposed based on particle swarm optimization (PSO) for MPPT of PV modules. After tuning the parameters of the fuzzy system, including membership function parameters and consequent part parameters, to obtain maximum power point (MPP), a DC/DC boost converter connects the PV array to a resistive load. ANFIS reference model is used to control duty cycle of the DC/DC boost converter, so that maximum power is transferred to the resistive load. Comparing the proposed method with PSO alone method and firefly algorithm (FA) alone shows its efficacy and high speed tracking of MPP under PSC. Due to the fact that these optimization algorithms have online applications, the convergence time of the algorithms is very important. The simulation results show that the convergence time for the proposed ANFIS-based method is lower than 0.15 second, while it is nearly three second for PSO and FA methods.
