Fig 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
![Fig. 2. General category of machine learning method [24].](publication/345840713/figure/fig2/AS:966192281972737@1607369599587/General-category-of-machine-learning-method-24_Q320.jpg)
Fault detection and repair of the components of photovoltaic (PV) systems are essential to avoid economic losses and facility accidents, thereby ensuring reliable and safe systems. This article presents a method to detect faults in a PV system based on power ratio (PR), voltage ratio (VR), and current ratio (IR). The lower control limit (LCL) and u...
Contexts in source publication
Context 1
... from one of the strings could be lower than that of the other strings. This situation can occur when PV modules connected in strings are not capable of generating electricity in the normal way because the PV module has been somehow polluted or damaged. When such mismatching has occurred, the I-V curve of the system has the shape of curve 1 in Fig. ...
Context 2
... mismatching of modules in strings or arrays usually occurs because of differences in generated electricity. Curve 2 of Fig. 1 is a case where the front surface of a PV system has been generally contaminated or plenty of time has passed since installation, resulting in decreasing ...
Context 3
... the PV cell at the open-circuit voltage because the overall current flows through the PV cell, resulting in the series resistance of 0. By contrast, around the open-circuit voltage, the series resistance greatly affects the I-V curve. The I-V curve of a PV module as the series resistance increases could be drawn in a similar form as curve 5 in Fig. ...
Context 4
... formed, the current generated from the PV module and opencircuit voltage decreases. A fault caused by the shunt resistance is even more drastic at low-irradiation conditions because the light-induced current is lesser in this situation. The I-V curve of the PV module as the shunt resistance decreases could be drawn in a similar form as curve 6 in Fig. 1. The parallel resistance can be approximated as the slope of the I-V curve at the short-circuit current ...
Context 5
... ¯ x i and ¯ y i are the average values of x i and y i , respectively, and n is the size of the samples of each variable. In Figs. 1, 3, and 5, the numbers in the upper half refer to the correlation coefficients between the variables, and the red lines in the lower half indicate the linearity between the variables [32]. The data used for the Pearson correlation analysis were collected from the test site, which is discussed in detail in Section ...
Context 6
... (24), the V R for various cases of defective modules was calculated and is illustrated in Fig. 10. A comparison was made for four configurations: 6 (series) by 6 (parallel), 9 by 4, 12 by 3, and 18 by ...
Context 7
... proportionally to the number of parallel connections. In a series connection of PV modules, the current is insensitive to the number of faulty modules in a string when the bypass diode is activated. However, when the entire string becomes faulty, the current of the PV array decreases, which is the definition of parallel fault. This is shown in Fig. 11, as the current of each configuration decreases in a step-like ...
Context 8
... Power Ratio (P R ): The power ratio measured at maximum power point is calculated by dividing the measured value of P mp by the multiplied value of the voltage and current, which were predicted using the model, as presented in Table I. P R is calculated using Fig. 12 illustrates the change in the calculated P R as the number of faulty modules changes in the PV system, which consists of the same number of PV modules but in a different configuration of series and parallel connections. In a parallel fault, all three configurations showed a similar reduction rate of PR. Notably, the reduction rates of ...
Context 9
... fault detection and diagnosis algorithm for PV systems was designed, as shown in Fig. 13. In the first stage of the algorithm, the measured values of the output data and environmental variables are fed to the PV system. Subsequently, P R , V R , and I R are calculated using the equations presented in Section II. These values of P R , V R , and I R are the standard values used to determine if faults have occurred in the PV ...
Context 10
... outdoor testing site was designed to verify the fault detection/diagnosis algorithm, as presented in Section III, and is shown in Fig. 14. The PV system was composed of 36 PV modules with the characteristics of 50 W (18 V and 2.77 A). A total of 12 modules are connected in series forming a string and three strings are connected in parallel to form an array. The specifications of the PV modules used in the test are listed in Table II. The geographical coordinates of the ...
Context 11
... types of fault conditions in the PV system were designed: series, parallel, and total. These are shown in Fig. 15. The series and parallel faults were defined in Section III-C. The total fault is a case where both voltage and current of PV array have dropped due to complex reasons, such as open-/short-circuited problem and shading. In this study, the total fault condition has been emulated by decreasing the current and voltage of the array. ...
Context 12
... change in P R value for each fault situation is shown in Fig. 16. In the case of a series fault, P R decreased in phase to 0.91, 0.85, 0.76, and 0.68 as the number of faulty modules increased. This result is outside the control range of LCL and UCL, which is defined as 0.93-1.02. In the case of a total fault, P R shows a similar tendency as a series fault by decreasing to 0.88, 0.80, 0.71, and 0.62 ...
Context 13
... similar tendency as a series fault by decreasing to 0.88, 0.80, 0.71, and 0.62 as the number of faulty modules increased. Finally, for the case of a parallel fault, the P R decreased sharply to 0.66 and 0.33 as connections between the array were short circuited. The result showed that, in general, P R decreases in phase for all three fault types. Fig. 17 shows changes in V R of the PV system under each fault condition. In the case of a series fault, V R dropped to 0.95, 0.87, 0.79, and 0.69 on average, which is below the control range of LCL and UCL, defined as 0.99-1.01. In the case of a total fault, the decrease in V R is similar to that of a series fault with the values of 0.91, ...
Context 14
... total fault conditions but not under a parallel fault condition. Therefore, V R can be used as a criterion to diagnose parallel faults and the other two fault types: If V R is within a designated range, and yet still there is a drop in P R , it means that there is a parallel fault in the system; otherwise, there is either series or total fault. Fig. 18 shows the change in I R under each fault condition. In a series connection, the series fault does not affect the current of the PV array; therefore, there are no changes in I R under series fault ...
Citations
... For example, incoming shade from trees that have grown taller or buildings that were constructed later can reduce the PV system's output. Other possible effects are pollution [11] by dust and leaves or degradation of the PV modules [12]. With information about irradiation, wind speed, ambient temperature or sun angle, it is not possible to represent such caused drops in power. ...
This paper presents a mahcine learning based solar power forecast methode that can take into account shading related fluctuations. The generated PV power is difficult to predict because there are various fluctuations. Such fluctuations can be weather related when a cloud passes over the array. But they can also occur due to shading caused by stationary obstacles, and this paper addresses this form of shading. In this work an approach is presented that improves the forecast under such fluctuations caused by shading. A correction of the prediction could successfully reduce error due to shading. The evaluation of the model is based on five sets of recorded shading data, where shading resulted from intentionally placed structures. The correction uses internal inverter data and irradiance values of the previous day to perform the correction and was able to reduce the RMSE of four 10 kWp systems with different orientation and tilt angle under shading and thus improve the prediction accuracy by up to 40%. The model can detect how intense the shading is and correct the forecast by itself.
... Pengembangan implementasi EBT dapat direalisasikan dengan berbagai macam cara. Salah satu contoh adalah membangun pembangkit listrik tenaga surya (PLTS) [5][6] [7][8] [9]. Perusahaan Listrik Negara (PLN) telah membuka peluang kepada para pelanggan untuk membangun PLTS secara mandiri di atas bangunan masing-masing pelanggan. ...
Global warming is one of the problems facing the world today. One solution to this problem is to utilize new and renewable energy-based power plants (EBT). The contribution of this research is to analyze the use of energy produced by PLTS FTI UII, as well as the effect of climate change on the production of PLTS FTI UII. The research steps are collecting data, calculating the value of PV mini-grid efficiency, and conducting climate change analysis on PV mini-grid production. The research data is the amount of energy production and energy consumption of PLTS FTI UII for the period March 2017 to July 2022. Efficiency calculations are carried out by calculating the ratio of production and total capacity of PLTS. Analysis of the effect of climate change on PV mini-grid energy production is to compare the GHI potential data with the total production of PV mini-grid each year. The results of the study show that the lowest level of production and efficiency of PLTS is in 2018 and the highest is in 2019. The factor that affects the low efficiency is the occurrence of many problems in 2018 so that it requires a lot of maintenance processes. This process will affect the production of PLTS so that the conversion process of solar energy decreases. Furthermore, the factor that causes the low efficiency is due to global warming which causes the climate to become uncertain and the production of carbon dioxide increases. Thus, energy production in PLTS decreases. From some of these analyzes, it can be concluded that the factors causing the decline in production and efficiency in the PLTS FTI UII system are the excessive and long maintenance process of the PLTS system, as well as climate change that occurs around the PLTS FTI UII system. The results of this study can be used for the evaluation process of the development of the PLTS FTI UII system in the future. Pemanasan global merupakan salah satu permasalahan yang dihadapi dunia saat ini. Salah satu solusi terhadap permasalahan tersebut adalah dengan memanfaatkan pembangkit listrik berbasis energi baru dan terbarukan (EBT). Kontribusi penelitian ini adalah menganalisis pemanfaatan energi yang dihasilkan oleh PLTS FTI UII, serta pengaruh perubahan iklim terhadap produksi PLTS FTI UII. Langkah penelitian yaitu pengumpulan data, perhitungan nilai efisiensi PLTS, serta melakukan analisis perubahan iklim terhadap produksi PLTS. Data penelitian adalah jumlah produksi energi serta konsumsi energi PLTS FTI UII dengan periode Maret 2017 sampai Juli 2022. Perhitungan efisiensi dilakukan dengan cara menghitung rasio produksi dan total kapasitas PLTS. Analisis pengaruh perubahan iklim terhadap produksi energi PLTS adalah membandingkan data potensi GHI dengan total produksi PLTS di setiap tahun. Hasil dari penelitian menunjukkan tingkat produksi dan efisiensi terendah PLTS adalah tahun 2018 dan tertinggi tahun 2019. Faktor yang mempengaruhi rendahnya efisiensi adalah terjadinya banyak masalah pada tahun 2018 sehingga mengharuskan banyak proses pemeliharaan. Proses tersebut akan mempengaruhi produksi PLTS sehingga proses konversi energi matahari menjadi menurun. Selanjutnya, faktor yang menyebabkan rendahnya efisiensi tersebut dikarenakan adanya pemanasan global yang mengakibatkan iklim menjadi tidak menentu dan meningkatnya produksi karbon dioksida. Sehingga, produksi energi pada PLTS menjadi menurun. Dari beberapa analisis tersebut, dapat disimpulkan bahwa faktor-faktor penyebab menurunnya produksi dan efisiensi pada sistem PLTS FTI UII adalah proses pemeliharaan sistem PLTS yang terlalu banyak dan lama, serta perubahan iklim yang terjadi di sekitar sistem PLTS FTI UII. Hasil dari penelitian ini dapat digunakan untuk proses evaluasi pembangunan sistem PLTS FTI UII di masa depan.
... Detecting faults of the PV components and fix it is necessary to avoid economic losses and big incidents that may established in this systems, thus ensuring secure and robust systems [5]. Moreover, more time and costs are suffered when malfunction is failed to be detected in a timely manner in the system. ...
... Moreover, more time and costs are suffered when malfunction is failed to be detected in a timely manner in the system. Therefore, to ensure a highquality system for prolonged, it is essential to recognize the times and locations of faults and failures immediately [5]. ...
Introduction. This research work focuses on the design and experimental validation of fault detection techniques in grid-connected solar photovoltaic system operating under Maximum Power Point Tracking mode and subjected to various operating conditions. Purpose. Six fault scenarios are considered in this study including partial shading, open circuit in the photovoltaic array, complete failure of one of the six IGBTs of the inverter and some parametric faults that may appear in controller of the boost converter. Methods. The fault detection technique developed in this work is based on artificial neural networks and uses discrete wavelet transform to extract the features for the identification of the underlying faults. By applying discrete wavelet transform, the time domain inverter output current is decomposed into different frequency bands, and then the root mean square values at each frequency band are used to train the neural network. Results. The proposed fault diagnosis method has been extensively tested on the above faults scenarios and proved to be very effective and extremely accurate under large variations in the irradiance and temperature. Practical significance. The results obtained in the binary numerical system allow it to be used as a machine code and the simulation results has been validated by MATLAB / Simulink software.
... Gyu Gwang Kim et al. [12] presented a method to detect faults in a photovoltaic system based on the power ratio (PR), the voltage ratio (VR), and the current ratio (IR). Each ratio's lower control limit (LCL) and upper control limit (UCL) were defined using data from a test site system under normal operating conditions. ...
One of the critical aspects in the mining sector is energy, being of great importance for the operation since if it were to stop, one of the consequences would be the loss of large amounts of money. The research objective is to predict the State of Charge of Batteries of equipment powered by photovoltaic solar panels in the mining sector based on automatic supervised learning techniques. A monitoring system records each energy variable programmed in the photovoltaic system, for which an analysis of the data extracted from the monitoring system was carried out. The data were evaluated using automatic supervised learning techniques using the RapidMiner tool, whose prediction average was 90.12%. The technique of automatic supervised learning of artificial neural networks was chosen to predict the state of charge of batteries for photovoltaic systems. A software tool was built with the neural network. The analysis and discussion of the results of the training of the model were carried out, the contribution of this research being to determine the prediction of the state of charge of batteries in photovoltaic systems in the mining sector using techniques of supervised machine learning which was the neural network. Finally, with the model correctly trained, validation was carried out that allowed comparing the predictive data with the data in real-time, obtaining a good relationship and satisfactory results.
... It was found that System 1 under investigation showed an early failure symptom where the cumulative percentage of AR < 0.9 ranges between 34% to 71%, meanwhile System 2 and 3 were identified as fault-free GCPV systems with cumulative percentage AR < 0.9 ranging from 5% until 19% (Shukor et al., 2021). Likewise, a study was also conducted on failure detection at the PV array level, which involved DC AR (Kim et al., 2021). The study proposed that DC AR must range between 0.93 until 1.02 for a normal operating condition. ...
... In addition, various type of failure was diagnosed from this study, such as series, parallel and total failure. These identified failures were the factors that led to the decrease in the electrical output of the system (Kim et al., 2021). ...
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.
... In (Zuniga-Reyes et al. 2021), K-nearest neighbors algorithm and Discrete Fourier transform are used in analyzing ripple voltage and current to detect and identify partial shading and short-circuit in a PV system. A multivariate analysis based on voltage, current and power ratio is proposed by Kim et al. (2021) to diagnose faults in a PV array. Three concurrent faults including partial shading, line-to-line fault and open circuit have been identified with Fisher discrimination dictionary learning for sparse representation (Xi et al. 2021). ...
Improving the efficiency of photovoltaic (PV) systems has gained priority in current research due to the large volumes of PV panels installed. Moreover, the remarkable efforts made to investigate different methods of diagnosing PV failures have multiplied, giving additional impetus to research on the efficiency of PV systems. However, most of these methods are limited in the number of faults that can be identified; some are expensive and complex, and others require huge amounts of data to train. In this paper, a simple and robust multivariate statistical analysis method is proposed for the diagnosis and identification of faults in a PV system. From the multitude of data that can be collected on a PV system in operation or available on most new inverters on the market, the method used here is based on kernel principal component analysis, which looks at and analyses the variance between these data. Combined with the Hotelling statistic (T2) and Squared Prediction Error (SPE) index associated with Rolle’s theorem, this analysis identifies six operating states of the PV system: normal operation, short-circuited panels, open-circuit panels, partially shaded panels, serial resistance degradation, and MPPT error. 200 samples of different faults at different irradiance have been generated between an irradiance from 50 to 500 w/m². The accuracy rate was 64% for the degradation of series resistance, 91.11% for partial shading, 100% for open circuit, 100% for short circuit, and 100% for MPPT error. The various results obtained first from a Matlab-Simulink model, and then from a real system of 18 modules, demonstrate the efficiency and performance of the proposed algorithms.
Owing to the consumers’ growing tendency to solar energy, monitoring and detecting possible faults in the photovoltaic (PV) systems are very important. Since common protection systems are not able to detect faults in photovoltaic systems, many approaches have been proposed to detect faults and increase system reliability. In this paper, various states of faults occurring in PV systems, e.g. partial shading fault (PSF), line-to-line fault (LLF) including intra-string (IS) and cross-string (CS), open circuit fault (OCF), and combinations of these faults such as hybrid faults are investigated. Moreover, the characteristic curve of PV systems for different faults are delineated, and mathematical equations for the extreme points of each curve are presented in order to study the effect of various faults on power-voltage (P-V) curve, comparing them with the healthy mode. The obtained results were used to present an algorithm based on definition of the fault indexes for identification, diagnosis and localization of the faults. The effect of maximum power point tracking (MPPT) was considered in the process of fault detection, and finally, different scenarios of faults were considered in simulation and experimental cases to validate the suggested algorithm.
Photovoltaic (PV) power generation is susceptible to various permanent electrical faults, including line-to-line (LL), line-to-ground (LG), and open-circuit (OC) faults. Additionally, non-catastrophic, temporary shading faults can also occur. These faults can lead to changes in the I-V characteristics and cause a shift in the maximum power point (MPP). Permanent faults such as LL/LG and OC can significantly alter the array's open circuit voltage and short circuit current. This research proposes a novel sensor-less approach for detecting, discriminating, and locating different faults in a PV array. The proposed approach utilizes the mandatory Maximum Power Point Tracking (MPPT) reference to assess the likelihood of fault occurrence. It then analyzes the altered I-V characteristics to discriminate and locate the fault location within the PV array. To facilitate fault detection and categorization, a new Fault Detector-Array Combiner Box (FD-ACB) with power electronic switches has been developed. The effectiveness of the proposed algorithm is evaluated using a test setup consisting of three 4×4 sub-arrays. Detailed case studies involving LL/LG, OC, and shade faults are presented. The results demonstrate that the sensor-less FD-ACB setup has the potential to find (i) undetected and undiagnosed LL/LG and OC faults, (ii) discriminate shade and permanent faults, and (iii) locate the faulty PV sub-array vulnerable to permanent faults.
