Search

Power systems protection has become more vital in recent years to ensure stability, reliability, security, and power quality due to the exponential growth of grid-connected photovoltaic (GCPV) systems. As a result, several nations have set new grid codes for the grid integration of PV plant installations to overcome these concerns. Investigating Fault Ride Through (FRT) capacity is one of the primary criteria for grid codes. For 3-phase GCPV systems, fault detection and classification approaches are proposed in this study. Firstly, different faults that occurred in the GCPV system are categorized and compared, with the critical and analytical evaluation of grid codes, particularly FRT requirements such as Low Voltage Ride Through (LVRT) and High Voltage Ride Through (HVRT) for different nations. Further, a detailed classification and a comparison of the existing FRT techniques are presented for better control methods based on system complexity, detection accuracy, and other evolutionary criteria. To ensure smooth grid operation, accurate fault detection and condition monitoring of the systems are required. This paper discusses a machine learning (ML) based technique for detecting faults in 3-phase GCPV systems. Multi-peak phenomena caused by FRT capabilities and anti-islanding detection are the main problems experienced while integrating a PV system into the local grid. The fault classification technique is developed using weighted K-nearest neighbor (WKNN) and fine Gaussian support vector machine (FGSVM) based ML approaches utilizing Wavelet Transform. The proposed ML-based findings show that the fault detection algorithm-based classification accuracy has significantly improved.

The performance status of a grid-connected photovoltaic (GCPV) system is denoted by performance indices, namely performance ratio, capacity factor, and even through power acceptance ratio (AR), as documented in Malaysia Standard (MS) procedures for acceptance test of GCPV testing and commissioning (TNC). Even though AR analysis can be either on the DC or AC side, the MS TNC procedures implemented analysis on the AC side. Therefore, the question arises whether there is any significant difference when using AC AR analysis compared to DC AR analysis in evaluating the system performance. Thus, this paper evaluates the differences between applying DC AR analysis and AC AR analysis in accessing the performance of the ten kWp GCPV system in Malaysia. The AR analytical analysis employed the 2019 one-year historical data of solar irradiance, module temperature, DC power, and AC power. The results demonstrated that the monthly AC AR were consistently lower than DC AR with a percentage difference of approximately 3%. The percentage discrepancy was due to the variation of actual inverter efficiencies compared to the declared constant value by the manufacturer used in the AR prediction model. These findings have verified a significant difference between DC AR analysis and AC AR analysis. Most importantly, this study has highlighted the significance of AC AR analysis compared to DC AR analysis as a tool to evaluate GCPV system performance because AC AR has taken an additional factor into consideration, which is the inverter efficiency variation. [ABSTRACT FROM AUTHOR]

In order to meet the increasing energy demands of the modern-day world, the shift towards renewable energy has proved to be a viable alternative, as it has significantly reduced reliance on conventional energy resources. Solar energy, in particular, remains one of the best available renewable energy options, as it is abundant, clean and reliable. Owing to the susceptibility of grid-connected photovoltaic (GCPV) system against grid faults, conventionally the PV inverter would disengage from the power grid by utilizing an anti-islanding technique. Nevertheless, many countries have now implemented grid codes (GCs) to secure and regulate the operation of the GCPV systems from exposure to grid faults. Modern advanced technologies have equipped the PV inverters with the low-voltage ride-through (LVRT) capability to address the issue of grid faults and prevent equipment failure, as it develops compliance of the GCPV systems with the GCs. This paper presents an overview of inverter topologies which include the power processing stage, transformer and transformerless inverter, multilevel inverter, soft and hard switching, bidirectional and hybrid inverters. Thereafter, current injection techniques with LVRT control strategy are discussed and examined. Such techniques have been classified under four main categories: the current compensation controls, reactive current injection (RCI), linear current controls and non-linear current controls. In addition, a comparative analysis has been presented, by thoroughly discussing the merits and demerits of these control schemes, based on the performance parameters of complexity, stability, robustness and power quality assessment. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents an intelligent sizing technique for sizing grid-connected photovoltaic (GCPV) system using evolutionary programming (EP). EP was used to select the optimal set of photovoltaic (PV) module and inverter for the system such that the technical or economic performance of the system could be optimized. The decision variables for the optimization process are the PV module and inverter which had been encoded as specific integers in the respective database. On the other hand, the objective function of the optimization task was set to be either to optimize the technical performance or the economic performance of the system. Before implementing the intelligent-based sizing algorithm, a conventional sizing model had been presented which later led to the development of an iterative-based sizing algorithm, known as ISA. As the ISA tested all available combinations of PV modules and inverters to be considered for the system, the overall sizing process became time consuming and tedious. Therefore, the proposed EP-based sizing algorithm, known as EPSA, was developed to accelerate the sizing process. During the development of EPSA, different EP models had been tested with a non-linear scaling factor being introduced to improve the performance of these models. Results showed that the EPSA had outperformed ISA in terms of producing lower computation time. Besides that, the incorporation of non-linear scaling factor had also improved the performance of all EP models under investigation. In addition, EPSA had also shown the best optimization performance when compared with other intelligent-based sizing algorithms using different types of Computational Intelligence. [Copyright &y& Elsevier]

Abstract: This paper presents a Hybrid Multi-Layer Feedforward Neural Network (HMLFNN) technique for predicting the output from a grid-connected photovoltaic (GCPV) system. In the proposed HMLFNN, the Artificial Immune System (AIS) was selected as the optimizer for the training process of the Multi-Layer Feedforward Neural Network (MLFNN). AIS was used to optimize the number of neurons in the hidden layer, the learning rate, the momentum rate, the type of activation function and the learning algorithm. In addition, the MLFNN utilized solar irradiance (SI) and module temperature (MT) as its inputs and kWh energy as its output. When compared with the classically trained MLFNN, the proposed HMLFNN was found to be superior in terms of having shortest computation time and lower prediction error. [Copyright &y& Elsevier]

This paper presents a multi-objective-evolutionary programming (MOEP) algorithm for sizing grid-connected photovoltaic (GCPV) system. Unlike previous studies which focused on the sizing of GCPV system based on a single objective optimization problem, the sizing of GCPV system in this study was initially converted into a three different bi-objective optimization problems. The objectives were formulated based on the aim to maximize the expected technical and economic performance indicators for the system design. Apart from that, in each bi-objective optimization problem, the proposed MOEP was used to select the optimal combinations of photovoltaic (PV) module and inverter such that the optimal values of the two performance indicators could be achieved in a single run. The performance of the proposed MOEP-based sizing algorithm was later compared with the performance of the weighted sum method (WSM)- based sizing algorithm. In addition, when illustrating the effectiveness of the proposed MOEP in approximating the Pareto front, the non-dominated solutions from MOEP were compared with a benchmark algorithm which produced all possible design solutions. The proposed MOEP-based algorithm was found to outperform WSM in providing faster computations and better approximation of the Pareto front. [ABSTRACT FROM AUTHOR]

This paper presents a new intelligent algorithm for sizing grid-connected photovoltaic (GCPV) systems. The sizing process involves the selection of PV module and inverter as well as the estimation of the technical and economic output performance of the system. Although many existing sizing software have been beneficial in assisting the GCPV system design, the application of these software might be limited due to the rigid algorithm embedded in the software. On the other hand, the manual calculation-based design approach could be another viable option but the design process could be extremely tedious especially when various design options need to be evaluated. In this study, the PV module and the inverter were selected from the respective databases whereas the sizing algorithm was developed evolutionary programming (EP). Besides selecting the PV module and inverter, the new algorithm, known as EP-based sizing algorithm (EPSA) was used to determine the techno-economic performance of the system. The EPSA was tested using the classical EP and meta-EP models before the best model was selected for EPSA. The proposed EPSA with meta-EP model was found to outperform a proposed conventional iterative sizing algorithm (ISA) in terms of computation time. [ABSTRACT FROM AUTHOR]