Fault detection in photovoltaic systems using machine learning technique

This paper deals with a key solution for power outages problem experienced by many African countries and this through grid-connected photovoltaic (PV) systems with batteries storage. African grids are characterized by an insufficient power supply and frequent interruptions. Due to this fact, users who especially use classical grid-connected photovoltaic systems are unable to profit from their installation even if there is sun. In this study, we suggest the using of a grid-connected photovoltaic system with batteries storage as a solution to these problems. This photovoltaic system works by injecting the surplus of electricity production into grid and can also deliver electricity as a stand-alone system with all security needed. To achieve our study objectives, firstly we conducted a survey of a real situation of one African electrical grid, the case of Burkina Faso (SONABEL: National Electricity Company of Burkina). Secondly, as study case, we undertake a sizing, a modeling and a simulation of a grid-connected PV system with batteries storage for the LAME laboratory at the University of Ouagadougou. The simulation shows that the proposed grid-connected system allows users to profit from their photovoltaic installation at any time even if the public electrical grid has some failures either during the day or at night.

The rapid revolution in the solar industry over the last several years has increased the significance of photovoltaic (PV) systems. Power photovoltaic generation systems work in various outdoor climate conditions; therefore, faults may occur within the PV arrays in the power system. Fault detection is a fundamental task needed to improve the reliability, efficiency, and safety of PV systems, and, if not detected, the cost associated with the loss of power generated from PV modules will be quite high. Moreover, maintenance staff will take more time and effort to fix undetermined faults. Due to the current-limiting nature and nonlinear output characteristics of PV arrays, fault detection is not that easy and the application of artificial intelligence is proposed for the sake of fault detection in PV systems. The idea behind this approach is to compare the faulty PV module with its accurate model (factory fingerprint) by checking every PV array’s I–V and P–V curves using the Artificial Neural Network (ANN) logarithm as a subsection of the Artificial Intelligence’s (AI) techniques. This proposed approach achieves a high performance of fault detection and gives the advantage of determining what type of fault has occurred. The results confirm that the proposed logarithm performance becomes better as the number of distinguishing points extend, providing great value to the Solar PV industry.

-The rapid integration of photovoltaic (PV) systems into modern power networks has intensified the need for efficient control strategies to ensure maximum power extraction, grid stability, and power quality. Variations in solar irradiance, temperature, and grid conditions significantly affect PV system performance, necessitating advanced Maximum Power Point Tracking (MPPT) and inverter control methods. This paper presents a comprehensive review of conventional, intelligent, and optimization-based control techniques employed in grid-connected PV and hybrid microgrid systems. Various MPPT algorithms, including classical methods, soft computing approaches, and metaheuristic optimization techniques, are critically analyzed with respect to tracking accuracy, convergence speed, computational complexity, and robustness under dynamic conditions. Furthermore, inverter control strategies, synchronization methods, and hierarchical control architectures for microgrids are discussed. A generalized control block diagram is presented to illustrate system integration. Finally, key challenges, research gaps, and future directions are highlighted to guide researchers toward next-generation intelligent and resilient PV-based energy systems. Key Words: Photovoltaic System; Maximum Power Point Tracking (MPPT); Grid-Connected Inverter; Microgrid Control; DC–DC Converter; Intelligent Control; Renewable Energy

Grid-connected photovoltaic (PV) systems require a power converter to extract maximum power and deliver high-quality electricity to the grid. Traditional control methods, such as proportional-integral (PI) control for DC-link voltage regulation, often struggle under abnormal operating conditions, resulting in voltage fluctuations and instability in the maximum power point tracker (MPPT). This paper proposes a synergistic control strategy that combines a musical chairs algorithm (MCA) MPPT with sliding mode control (SMC) together for the boost converter DC-link control. This approach enhances DC-link voltage stability by switching the MPPT to SMC of the boost converter when the DC-link voltage exceeds the predefined limit. This strategy enhances the stability of the DC-link voltage and allows for a smaller DC-link capacitor, thereby reducing system cost and improving the power quality of PV systems. A phase-locked loop (PLL) further ensures effective grid synchronization. The reduction in DC-link voltage overshoot (from 570 V to 522 V) improved stability under varying irradiance conditions. Moreover, a 48 V reduction in overshoot voltage and a 66% decrease in DC-link voltage ripple (standard deviation from 17.93 V to 5.92 V) occurred. Simulation and experimental results demonstrate the superiority of the proposed strategy compared to the case without coordination between the DC-DC converter and inverter controllers, particularly under challenging conditions.

Grid-connected photovoltaic (PV) systems with relatively high capacity effectively reduce peak load, but because of their reverse power flow, they can cause overvoltage along a feeder that can exceed five percent of the rated voltage. Modern PV systems with the capability of Volt/Var control can mitigate overvoltage by either injecting or absorbing reactive power. Thus, the objectives of this study are to (1) model a distribution network (feeder J1 located in the northeastern United States and enhanced by thirteen PV systems) using OpenDSS; (2) use actual collected load profile data as input data to the simulation model and PV generation output data at 15-minute intervals; and (3) calculate the time-series steady-state power flow of the distribution network when PV systems generate or absorb reactive power. Finally, it addresses overvoltage resulting from the deployment of high-capacity grid-connected renewable PV systems and mitigation techniques for such overvoltage issues, particularly Volt/Var management and control.

In this paper, a complete simulation model of a single phase grid-connected photovoltaic (PV) system with associated controllers is presented. The simulation model is developed in MATLAB/SIMULINK tool. The main component of the single phase grid-connected PV system are, a PV array, a dc-dc boost converter, a PWM based voltage source inverter and filter. For high efficiency of the PV system used maximum power point tracking (MPPT) algorithm. Simulation studies shows that the sun irradiance are increase so increase the PV array output power and also increase the PQ. PV energy is large than load demand; the excess energy is will exported to the utility.

Control of multi-functional grid-connected PV systems with load compensation under distorted and unbalanced grid voltages

Nowadays, due to the high-scale penetration of photovoltaic systems, reliable and efficient grid-connected photovoltaic (PV) systems utilizing the advances of power electronics and control system technology are desirable. Thus, single-stage grid-connected photovoltaic systems, have gained attention, especially in low voltage applications. However, PV systems exhibit nonlinear behavior that could negatively affect the performance of the system if they are not adequately compensated for. In this dissertation, using the general structure for the synchronous dq0 frame, a single-stage three-phase grid-connected photovoltaic system, and a single-stage single-phase grid-connected PV system, both with a nonlinear control strategy, are developed to track the maximum power and to control the active and reactive power, without the necessity of an additional power converter. A novel trajectory of the reference current is obtained online taking into account the dynamics of the DC link capacitor and the switching function of the inverter. Unlike to the three-phase system, the single-phase system includes a novel method to mitigate the double line-frequency current ripple of the PV array, which is the major drawback of the single-phase PV inverter. Moreover, based on the preceded work, the nonlinear controller is combined with adaptive control to estimate the unknown disturbances that physically could appear in the circuit and affect the performance of the system. The stability of the systems and boundedness of signals are demonstrated by Lyapunov stability analysis. Simulation results show the effectiveness and robustness of the proposed controllers to track the maximum power and to control the active and reactive power.

Performance results and analysis of 3 kW grid-connected PV systems

Due to the Randomness and fluctuation of the photovoltaic (PV) system, Maximum Power Point Tracking (MPPT) algorithm is employed to ensure the maximum power extraction from PV arrays, when it is connected to the grid. The previous studies have indicated that the MPPT algorithm is one of the main causes of interharmonics emission in PV systems. The MPPT algorithm parameters such as perturbation step size inevitably have a strong impact on the interharmonics in the current injected to the grid. In general, a smaller MPPT perturbation step size will reduce the interharmonics emission level, but it will result in poor tracking performance of the MPPT algorithm. Thus, there is a trade-off between the MPPT efficiency and the interharmonics emission in PV system. In this paper, an interharmonics suppression scheme with variable step size MPPT algorithm was proposed. In the suppression scheme, two voltage values near by the MPPT voltage and two tracking step size are preset. When the dc-link voltage of PV is between the two preset voltage values, it means that the PV power near its MPP and the smaller perturbation step size implemented in the MPPT algorithm. Otherwise, the bigger perturbation step size is employed. It can greatly reduce interharmonics emission level and maintain the MPPT efficiency. Simulations have been carried out to verify the effectiveness of the suppression scheme.

Nowadays, grid-connected photovoltaic (PV) power system is quite popular in many countries. For grid-connected PV power system, to achieve maximum power and good power quality of the system are considered as big challenges. In order to achieve this, artificial intelligent (AI) controller-based maximum power point tracking (MPPT) algorithm has been investigated for PV system as well as a new cascade multilevel inverter (MLI) is proposed for grid integration of PV system. The proposed cascade MLI has been designed with a smaller number of power electronic switches and it can be operated at asynchronous voltage sources, which is most adaptable for PV system. This proposed inverter can reduce the total harmonic distortion (THD) at output side by means of increasing the output voltage level with this power quality of the system has been improved. The microgrid integration of proposed inverter has been controlled by using AI-based voltage source controller and this proposed system is designed and simulated in MATLAB environment at various weather conditions and loading conditions. Finally, prototype model has been developed in laboratory and evaluating its performance. The simulation results and prototype results were verified with IEEE 1547 standard that proves the proposed system effectiveness.

Bypass diode faults are among the most hard-to-detect but impactful anomalies in photovoltaic (PV) systems, especially under partial shading conditions, where their electrical signatures often resemble those caused by non-critical irradiance variations. This study presents a systematic simulation-based investigation into how different bypass diode fault types—short-circuited, open-circuited, and healthy—affect the electrical behavior of PV strings under diverse irradiance profiles. A high-resolution MATLAB/Simulink model is developed to simulate 27 unique diode fault configurations across multiple shading scenarios, enabling the extraction of key features from resulting I–V curves. These features include global and local maximum power point parameters, open-circuit voltage, and short-circuit current. To address the challenge of feature redundancy and classification ambiguity, a preprocessing step is applied to remove near-duplicate instances and improve model generalization. An artificial neural network (ANN) model is then trained to classify the number of faulty bypass diodes based on these features. Comparative evaluations are conducted with support vector machines and random forests. The results indicate that the ANN achieves the highest test accuracy (93.57%) and average AUC (0.9925), outperforming other classifiers in both robustness and discriminative power. These findings highlight the importance of feature-informed, data-driven approaches for fault detection in PV systems and demonstrate the feasibility of diode fault classification without precise fault localization.

As the penetration level of grid-connected Photovoltaic (PV) systems increases, the power quality is one of the major concerns for system operators and the demands are becoming even stricter. The impact of interharmonics on the grid has been acknowledged in recent research when considering a large-scale adoption of PV inverters. However, the origins of interharmonics remain unclear. Thus, this paper performs tests on a commercial PV inverter to explore interharmonic generation and more important investigates the mechanism of interharmonic emission. The investigation reveals that the perturbation of the Maximum Power Point Tracking (MPPT) algorithm is one of the sources that induce interharmonics in the grid current, especially at low-power operating conditions. Accordingly, three mitigation solutions are discussed to address this issue, and simulations have been carried out to verify the effectiveness of the solutions. Simulation results indicate that the constant-voltage MPPT method is the most suitable solution to the mitigation of interharmonics introduced by the MPPT operation, as it avoids the perturbation in the PV voltage during operation.

The lack of control in voltage overshoot, transient response, and steady state error are major issues that are frequently encountered in a grid-connected photovoltaic (PV) system, resulting in poor power quality performance and damages to the overall power system. This paper presents the performance of a control strategy for an inverter in a three-phase grid-connected PV system. The system consists of a PV panel, a boost converter, a DC link, an inverter, and a resistor-inductor (RL) filter and is connected to the utility grid through a voltage source inverter. The main objective of the proposed strategy is to improve the power quality performance of the three-phase grid-connected inverter system by optimising the proportional-integral (PI) controller. Such a strategy aims to reduce the DC link input voltage fluctuation, decrease the harmonics, and stabilise the output current, voltage, frequency, and power flow. The particle swarm optimisation (PSO) technique was implemented to tune the PI controller parameters by minimising the error of the voltage regulator and current controller schemes in the inverter system. The system model and control strategies were implemented using MATLAB/Simulink environment (Version 2020A) Simscape-Power system toolbox. Results show that the proposed strategy outperformed other reported research works with total harmonic distortion (THD) at a grid voltage and current of 0.29% and 2.72%, respectively, and a transient response time of 0.1853s. Compared to conventional systems, the PI controller with PSO-based optimization provides less voltage overshoot by 11.1% while reducing the time to reach equilibrium state by 32.6%. The consideration of additional input parameters and the optimization of input parameters were identified to be the two main factors that contribute to the significant improvements in power quality control. Therefore, the proposed strategy effectively enhances the power quality of the utility grid, and such an enhancement contributes to the efficient and smooth integration of the PV system.

Assesment of grid-connected photovoltaic systems in the Kuwaiti climate