Real‐Time Data‐Driven Fault Diagnosis of Photovoltaic Arrays Using an Edge‐Server Machine‐Learning Framework

Fast and reliable fault detection is critical in photovoltaic (PV) systems to improve reliability and energy yield and reduce maintenance costs, ensuring safe and efficient operation under varying operating conditions. Although recent data-driven PV fault detection techniques (FDTs) in literature have demonstrated high diagnostic accuracies, they often suffer from practical limitations, offline operation, lack of fault localization and/or inability to concurrently identify faults. To address these challenges, a unified framework is proposed that simultaneously integrates real-time operation, fault classification and localization, and concurrent-fault identification in a single compact diagnostic approach. This is realized by developing a parallel feedforward neural network (FFNN) architecture whose performance is enhanced by a residual model-based structure, resulting in a more interpretable, scalable, reliable and accurate scheme. In addition, Grey Wolf Optimizer–Support Vector Machine (GWO–SVM) feature selection is incorporated to select the most influential diagnostic features, thus reducing data redundancy, enhancing diagnostic accuracy, and shortening training time. The proposed approach was tested to diagnose five types of PV faults (open circuit, short circuit, partial shading, degradation, and simultaneous faults), as well as classify their intensity and location. Simulation results show that the proposed FFNNs consistently outperform conventional Support Vector Machines (SVMs) in classification accuracy, with accuracies exceeding 98% and 99% for fault classification and localization, respectively, and above 95% for noisy data. Moreover, GWO-SVM proved to offer more stable feature subsets compared to Particle Swarm Optimization–SVM (PSO–SVM) in the considered feature selection structure. Simulation results validated the effectiveness of the proposed unified multiclassification fault diagnosis approach, even under system uncertainties, making it suited for real-world PV systems.

Photovoltaic (PV) systems have been gaining great attention during the last decade. One of the main faults in PV arrays is partial shading, affecting its electrical parameters. Failure to detect this fault may result in optimal generated power reduction and hot spotting leading to damage to cell encapsulant and second breakdown. This paper develops a partial shading detection algorithm that only requires the available measurements of array voltage and current. By quantifying the wave-shape of the super-imposed component of PV array power by the skewness function, it can discriminate a partial shading condition from the short-circuit and high-resistance faults. The developed scheme is implemented in a central intelligent electronic device and does not require a communication link and training data set. The merits of the proposed partial shading detection algorithm are demonstrated through several fault scenarios using a 5×5 grid-connected PV generation system.

In photovoltaic (PV) system, the output power is depending on the smooth operation of PV array. PV array may suffer from different fault condition due to open circuit, short circuit and partial shading condition. Fault is a abnormal condition that can cause hindrance to the smooth operation of PV system. Fault identification is major challenge to the design engineers to avoid power loss as well as damage to the PV system. Open circuit, short circuit and partial shading are most frequent faults that occur in PV array system. Many Researchers proposed fault diagnosis methods to recognize faults in PV array, majority failed in accurate and fast detection of faults. In this paper a novel method consists of Stockwell transform and machine learning based algorithm is used to detect and classify faults in PV array. The Necessary simulations and classification tasks are carried out in MATLAB Software.

Fault diagnosis in photovoltaic (PV) arrays is essential in enhancing power output as well as the useful life span of a PV system. Severe faults such as Partial Shading (PS) and high impedance faults, low location mismatch, and the presence of Maximum Power Point Tracking (MPPT) make fault detection challenging in harsh environmental conditions. In this regard, there have been several attempts made by various researchers to identify PV array faults. However, most of the previous work has focused on fault detection and classification in only a few faulty scenarios. This paper presents a novel approach that utilizes deep two-dimensional (2-D) Convolutional Neural Networks (CNN) to extract features from 2-D scalograms generated from PV system data in order to effectively detect and classify PV system faults. An in-depth quantitative evaluation of the proposed approach is presented and compared with previous classification methods for PV array faults – both classical machine learning based and deep learning based. Unlike contemporary work, five different faulty cases (including faults in PS – on which no work has been done before in the machine learning domain) have been considered in our study, along with the incorporation of MPPT. We generate a consistent dataset over which to compare ours and previous approaches, to make for the first (to the best of our knowledge) comprehensive and meaningful comparative evaluation of fault diagnosis. It is observed that the proposed method involving fine-tuned pre-trained CNN outperforms existing techniques, achieving a high fault detection accuracy of 73.53%. Our study also highlights the importance of representative and discriminative features to classify faults (as opposed to the use of raw data), especially in the noisy scenario, where our method achieves the best performance of 70.45%. We believe that our work will serve to guide future research in PV system fault diagnosis.

The active Maximum Power Point Tracking (MPPT), presence of blocking diodes, low insolation and the similarity in I-V characteristics among certain Line-Line (LL) faults and partial shading conditions imply the detection of fault as a major task. The undetected faults in Photovoltaic (PV) array lead to power loss, reduced reliability, fast degradation and may even cause fire risks. This paper proposes an effective Fault Detection Algorithm (FDA) using Maximal Overlap Discrete Wavelet Transform (MODWT) that would identify and discriminate the faults from partial shading and uniform irradiation changing conditions. The FDA uses input data as PV array voltage, current, irradiation and temperature. Various LL faults under discrete environmental conditions are verified using MATLAB/Simulink and is laboratory tests on a 4×4 PV array using LABVIEW software.

The manifested merits associated with solar energy including high sustainability, zero greenhouse gas emission and economic operation have encouraged wide penetration of photovoltaic (PV) systems in the microgrid, during the last few decades. However, the intermittency caused due to the fluctuating nature of solar irradiance demands an efficient maximum power point tracking (MPPT) algorithm for PV-integrated microgrids. The scenarios related to partial shading and faults in PV array impact the voltage–current behaviour resulting in the failure of conventional MPPT techniques in accurately estimating the operating point. The incorrect estimation by MPPT techniques quite often affects the operation of overcurrent protection modules. In this regard, this chapter presents an accurate sine cosine optimization (SCA)-based MPPT algorithm which will search the global operating point irrespective of the condition (i.e. in the event of partial shading or array faults), while avoiding undesired activation of the protection system. Besides this, a reliable protection scheme is proposed to detect and classify the faults in the distribution line under dual operating modes of microgrid (i.e. grid-connected and islanding). The instantaneous voltage–current signals recorded at the relaying bus are preprocessed through discrete wavelet transform (DWT) to obtain the discriminatory attributes, which are further utilized by the hybrid framework of artificial neural network (ANN) and SCA to perform the intended protection tasks under both modes of microgrid operation. The performance of the proposed MPPT technique and protection scheme has been analysed against a wide range of operating scenarios with real-time validation on OPAL-RT digital simulation platform.

Fault diagnosis model for photovoltaic array using a dual-channels convolutional neural network with a feature selection structure

The output power of a photovoltaic (PV) system is dependent on the perfect operation of the PV array. PV arrays may experience various fault conditions as a result of open circuit, short circuit, and partial shading. A fault is an aberrant situation that can interfere with the effective execution of a PV system. To reduce power loss and harm to the PV system, design engineers must identify faults. The most common problems in PV array systems are open circuit, partial shading, hotspots, cracks and short circuit faults. Several scientists suggested fault diagnosis methods to recognize faults in PV arrays, however the majority failed to detect problems accurately and quickly. In this paper, a two-stage algorithm based on wavelet transform and ensemble KNN method is used to detect and classify faults in PV array. A 6X6 series parallel PV array is used to test the proposed methodology. MATLAB Simulink is used to model the 6X6 PV array. Wavelet Analysis tool bar is used for signal analysis.

Line-to-Line (L-L) faults in Photovoltaic (PV) arrays prevent the PV system from producing maximum power, and if not cleared, may result in serious energy and revenue losses and cause fire hazards. Maximum Power Point Tracking (MPPT), a technique employed to maximize the power output of a PV array at different irradiance level, may potentially mask certain faults and make them undetectable by protection devices, especially when these faults occur under low solar irradiance condition or with high fault impedance. This paper proposes an innovative algorithm for detecting L-L faults in PV arrays based on support vector machine (SVM).

Detecting fault in a solar photovoltaic (PV) array in the presence of protection diodes (bypass and reverse blocking) is complex. This is because variations in the electrical parameters (due to a fault) are often not distinguishable from one type of fault to another or from fault to partial shading. Depending on the fault condition, an open fault may appear like a short-circuit fault and a severe partial shading occasionally may be detected as a short-circuit fault. Identifying a short-circuit fault and separating it from an open fault or partial shading are important as the former demands immediate attention to avoid damage to the PV modules and fire hazard. It is observed that the available techniques, based on transient changes in array parameters and model comparison, cannot identify faults under all practical scenarios. In this article, a fault detection and classification technique using array voltage, current, irradiance, and temperature measurements is proposed. The technique computes Thevenin equivalent resistance of the PV array for accurate fault identification. The simulation results demonstrate that the proposed technique can detect and classify faults correctly and differentiate them from partial shading. The comparative results show its superior performance compared with available techniques. The proposed technique is also validated by the results of hardware experiments.

In this paper, we address the problem of fault classification in PhotoVoltaic (PV) arrays using a semi-supervised graph signal processing approach. Traditional fault detection and classification methods require large amounts of labeled data for training. In utility scale solar arrays, obtaining labeled data for different fault classes is resource intensive. We propose a graph based classification technique that relies on a limited amount of labeled data. We compare our results with the well known supervised machine learning classifiers such as the K-nearest neighbour classifier, random forest classifier, support vector machines, and artificial neural networks. We also show that the graph-based classifiers require lower training computational cost compared to the standard supervised machine learning algorithms. The proposed method also achieves good classification performance with unseen data. We validate our method on a real-time dataset and show significant improvements over existing approaches.

This paper propounds a novel technique for detection of the faults in photovoltaic array by using Artificial Neural Network. By using a simulation model, the power variation under different faulty conditions such as open circuit fault, short circuit fault, and bridging fault are measured under normal and partial shading conditions. The simulated attributes are given to the Artificial Neural Network to predict the type of fault occurred in or between photovoltaic modules. Finally, three different training algorithms of Artificial Neural Network are compared for fault detection with help of mean square error as the performance parameter.

The widespread adoption of green energy resources worldwide, such as photovoltaic (PV) systems to generate green and renewable power, has prompted safety and reliability concerns. One of these concerns is fault diagnostics, which is needed to manage the reliability and output of PV systems. Severe PV faults make detecting faults challenging because of drastic weather circumstances. This research article presents a novel deep stack-based ensemble learning (DSEL) approach for diagnosing PV array faults. The DSEL approach compromises three deep-learning models, namely, deep neural network, long short-term memory, and Bi-directional long short-term memory, as base learners for diagnosing PV faults. To better analyze PV arrays, we use multinomial logistic regression as a meta-learner to combine the predictions of base learners. This study considers open circuits, short circuits, partial shading, bridge, degradation faults, and incorporation of the MPPT algorithm. The DSEL algorithm offers reliable, precise, and accurate PV-fault diagnostics for noiseless and noisy data. The proposed DSEL approach is quantitatively examined and compared to eight prior machine-learning and deep-learning-based PV-fault classification methodologies by using a simulated dataset. The findings show that the proposed approach outperforms other techniques, achieving 98.62% accuracy for fault detection with noiseless data and 94.87% accuracy with noisy data. The study revealed that the DSEL algorithm retains a strong generalization potential for detecting PV faults while enhancing prediction accuracy. Hence, the proposed DSEL algorithm detects and categorizes PV array faults more efficiently, reliably, and accurately.

Photovoltaic (PV) array fault diagnosis is important because it helps reduce energy and revenue losses to PV system operators. It also reduces fire hazards and electric shocks caused by PV array faults. As a result, many machine-learning-based fault diagnosis techniques have been proposed in recent times. Although the fault diagnosis accuracies associated with these techniques have been impressive, most machine learning algorithms rely on manual feature extraction, which is time consuming, expensive, and diagnostic expertise exacting. To address the problem of manual feature extraction, this paper proposes a new PV array fault diagnosis technique capable of automatically extracting features from raw data for PV array fault classification. The proposed technique utilizes long short-term memory networks, which is a deep learning algorithm, for feature extraction. The extracted features feed into a softmax regression classifier for fault diagnosis. The proposed technique exhibits high fault diagnosis accuracies on both noisy and noiseless data. In addition, the results of the proposed technique compare favorably with those of other techniques. It can, therefore, be inferred from the results that the proposed fault diagnosis technique offers an effective approach to automatically extract useful features from raw data and thus remove the need for the manual feature extraction.

Improved fault detection and classification in PV arrays using stockwell transform and data mining techniques