AI-Based Power Degradation Assessment in PV Cells

Monitoring and remotely controlling off-grid solar power systems using machine learning to predict and compare operating parameters versus actual measurements to detect abnormal operating conditions. A first ML model predicts solar panel power, intensity, load profile, and battery voltage based on sensor readings. A second ML model compares predicted vs actual values to detect anomalies. If abnormalities are found, the second ML model sends alerts and provides maintenance recommendations like UAV inspection, cleaning, or load reduction.

Source

A method for monitoring solar panel performance that separates short-term and long-term variations. The method uses a trained model to simulate average panel performance based on environmental parameters, and compares this simulated performance to actual panel output. By analyzing the difference between simulated and actual performance over time, the method can detect long-term degradation such as soiling or UV damage, and identify maintenance or weather-related events. The method can be applied to photovoltaic panels, thermal solar panels, and hybrid solar panels.

Source

System and method for determining degradation loss in solar panels of photovoltaic power plants using machine learning models. The system obtains location, weather, and configuration data for the panels, as well as real-time operating parameters from sensors. This data is input into a pre-trained machine learning model to determine degradation loss for each panel, with alerts triggered when loss exceeds a threshold.

Source

A process for maximum power point tracking (MPPT) of photovoltaic modules that monitors degradation of the modules. The process involves periodic I/V operating point measurements of the module, positioning of the operating point at the maximum power point (MPP), and monitoring of the degradation of the module. The monitoring includes measuring electrical parameters associated with the current-voltage characteristics of the module, comparing the measured parameters with modeled data, and deducing degradation paths and causes from the comparison.

Source

Real-time monitoring of photovoltaic (PV) systems through simultaneous measurement of individual panels and strings under varying environmental conditions, enabling continuous performance assessment without inverter shutdown.

Source

A monitoring system for photovoltaic systems that identifies the faulty component causing reduced performance without requiring field intervention. The system comprises electronic control means that analyze data from the photovoltaic panel, inverter, and optimizer to pinpoint the source of the degradation. This enables rapid diagnosis and repair, reducing downtime and maintenance costs.

Source

Method for determining degradation rate of photovoltaic installations without weather data, using production data from clear-sky days to estimate degradation rate through monthly aggregation and comparison of production decreases across years.

Source

Method for diagnosing internal loss mechanisms in solar cells by analyzing the type of simulated current density-voltage (JV) curve, enabling identification of specific loss mechanisms such as recombination, series resistance, and shunt resistance.

Source

Photovoltaic power generation is influenced not only by variable environmental factors, such as solar radiation, temperature, and humidity, but also by the condition of equipment, including solar modules and inverters. In order to preserve energy production, it is essential to maintain and operate the equipment in optimal condition, which makes it crucial to determine the condition of the equipment in advance. This paper proposes a method of determining a degradation of efficiency by focusing on photovoltaic equipment, especially inverters, using LSTM (Long Short-Term Memory) for maintenance. The deterioration in the efficiency of the inverter is set based on the power generation predicted through the LSTM model. To this end, a correlation analysis and a linear analysis were performed between the power generation data collected at the power plant to learn the power generation prediction model and the data collected by the environmental sensor. With this analysis, a model was trained using solar radiation data and power data that are highly correlated with power generation. The result... Read More

Source

The transition to renewable energy requires innovations in new renewable energy sources, such as wind turbines and photovoltaic (PV) systems. Challenges arise in ensuring efficient and reliable performance in their operation and maintenance. Predictive maintenance using machine learning (PdM-ML) is relevant for addressing these challenges by enhancing failure predictions and reducing downtime. This study examines the effectiveness of PdM-ML in wind turbine and PV systems by analyzing operational data, performing data preprocessing, and developing machine learning models for each system. The results indicate that the model for wind turbines can predict failures in critical components such as gearboxes and blades with high accuracy. In contrast, the model for PV systems is effective in predicting efficiency declines in inverters and solar panels. Regarding operational complexity, each model has advantages and disadvantages of its own, but when compared to conventional maintenance techniques, both provide lower costs with greater operational efficiency. In conclusion, machine learning-b... Read More

Source

Recent advancements in residential solar electricity have revolutionized sustainable development. This paper introduces a methodology leveraging machine learning to forecast solar panels power output based on weather and air pollution parameters, along with an automated model for fault detection. Innovations in high-efficiency solar panels and advanced energy storage systems ensure reliable electricity supply. Smart inverters and grid-tied systems enhance energy management. Government incentives and decreasing installation costs have increased solar power accessibility. The proposed methodology, utilizing machine learning techniques, achieved an R-squared value of 0.95 and a Mean Squared Error of 0.02 in forecasting solar panel power output, demonstrating high accuracy in predicting energy production under varying environmental conditions. By improving operational efficiency and anticipating power output, this approach not only reduces carbon footprints but also promotes energy independence, contributing to the global transition towards sustainability.

Source

System and method for detecting and reversing potential-induced degradation (PID) in solar modules based on their positioning in a string. The system monitors negative-to-ground voltage and fuse conditions in real-time to identify PID-prone modules, which are then switched from the negative side to the positive side to reverse the degradation. The method enables early detection and prevention of PID, which can cause up to 80% power loss in solar modules.

Source

A method for predicting performance of solar cell structures using machine learning algorithms. The method involves training a machine learning model on a dataset of known solar cell structures and their corresponding performance characteristics. The trained model is then used to predict the performance of new solar cell structures based on their structural parameters. The method enables rapid and accurate prediction of solar cell performance, enabling optimization of solar cell design and fabrication processes.

Source

Photovoltaic (PV) systems are integral to renewable energy, demanding accurate performance modeling for optimal functionality. This paper presents a pragmatic, data-driven approach employing Polynomial Regression (PR) for solar panel modeling to boost accuracy and adaptability to environmental variables. By emphasizing the advantages of data-driven models in attaining predictive precision, this study aims to reconcile theoretical concepts with practical solar panel performance. PR is employed as a black-box machine learning (ML) algorithm to transcend the limitations of traditional modeling methods, revealing ML's ability to grasp intricate relationships and ensure precise predictions. Utilizing real and simulated datasets encompassing solar irradiance, ambient temperature, and applied load through hardware in the loop (HIL), the model is trained to forecast the electrical outputs of solar panels in two approaches to estimate output voltage and current as well as key points on the panels' I-V curve. Assessment of model accuracy using Root Mean Square Error (RMSE) showcases the PR mod... Read More

Source

A method for predicting the performance and reliability of photovoltaic systems over time, taking into account factors such as installation quality, component specifications, environmental conditions, and maintenance practices. The method determines system parameters representing expected output, availability, and failure probability by analyzing parameters associated with the system, inverter, and components, and modeling their degradation over time. The system parameters are calculated based on the expected efficiency of the photovoltaic system, which is determined by analyzing the expected time course of output power and the degree of aging.

Source

The commercial development of perovskite solar cells (PSCs) has been significantly delayed by the constraint of performing time-consuming degradation studies under real outdoor conditions. These are necessary steps to determine the device lifetime, an area where PSCs traditionally suffer. In this work, we demonstrate that the outdoor degradation behavior of PSCs can be predicted by employing accelerated indoor stability analyses. The prediction was possible using a swift and accurate pipeline of machine learning algorithms and mathematical decompositions. By training the algorithms with different indoor stability data sets, we can determine the most relevant stress factors, thereby shedding light on the outdoor degradation pathways. Our methodology is not specific to PSCs and can be extended to other PV technologies where degradation and its mechanisms are crucial elements of their widespread adoption.

Source

A system for monitoring solar power plant performance that quantifies losses in energy generation using data from multiple inverters. The system collects inverter and weather station data, analyzes it using machine learning models to segregate and quantify losses, and generates maintenance tickets for underperforming inverters. The system compares inverter attributes to determine actual power losses, enabling real-time analysis and diagnosis of performance issues.

Source

This paper presents a case study for analysis and prediction of photovoltaic generation in the metropolitan region of Fortaleza-CE, applying machine learning techniques in the paradigms of time series prediction and system identification. Computational simulations were performed using linear (least squares) and non-linear (artificial neural networks and kernel methods) estimators that are part of the state-of-the-art in machine learning. In time series scenario, using only power measurements, the best results were obtained with the MLP network, with a prediction horizon of seven days ahead. In system identification, using power and solar radiation measurements, the least squares estimator achieved the best performance among all tested estimators, even in the free simulation scenario, i.e., infinite steps ahead.

Source

Photovoltaic (PV) solar energy has become a reference in electrical generation.The plants currently installed, and those planned have a huge capacity and occupy large areas.The increase in size of the plants presents new challenges in operation and maintenance areas, such as the optimization of the number of sensors installed, large data management and the reduction of the timework in maintenance.The aim of this paper is to show a methodology, to diagnose failures, based on the measured data in the plant.The methodology used is supervised regression machine learning and comparison algorithms.This methodology allows the study of the sensors, the inverters, the joint boxes and the power reduction caused by soiling.The result would allow the detection of around 1-5% of production loss in the plant.The algorithms have been tested with real data of PV plants, and have detected common failures such as production drops in strings and losses due to soiling.

Source

The integration of renewable energy sources and the advancement of smart grid technologies have revolutionized the power distribution landscape. As the smart grid evolves, the monitoring and control of power converters play a crucial role in ensuring the stability and efficiency of the overall system. This research paper introduced a converter monitoring system in photovoltaic systems, the main concern is to protect the electrical system from disastrous failures that occur when the system is in operating condition. The reliability of the converters is significantly influenced by the degradation of their passive components, which can be characterized in various ways. For instance, the aging of inductors and capacitors can be char-acterized by a decrease in their inductance and capacitance values. Identifying which component is undergoing degradation and assessing whether it is in a critical condition or not, is crucial for implementing cost-effective maintenance strategies. This paper explores a set of classification algorithms, leveraging machine learning, trained on data collected f... Read More

Source

As the energy demand continues to rise, renewable energy sources such as photovoltaic (PV) systems are becoming increasingly popular. PV systems convert solar radiation into electricity, making them an attractive option for reducing reliance on traditional electricity sources and decreasing carbon emissions. To optimize the usage of PV systems, intelligent forecasting algorithms are essential. They enable better decisionmaking regarding cost and energy efficiency, reliability, power optimization, and economic smart grid operations. Machine learning algorithms have proven to be effective in estimating the power of PV systems, improving accuracy by allowing models to understand complex relationships between parameters and evaluate the output power performance of photovoltaic cells. This work presents a study on the use of machine learning algorithms Catboost, LightGBM, XGboost and Random Forest to improve prediction. The study results indicate that using machine learning algorithms LightGBM can improve the accuracy of PV power prediction, which can have significant implications for opt... Read More

Source

This article introduces a data-driven approach to assessing failure mechanisms and reliability degradation in outdoor photovoltaic (PV) string inverters. The manufacturer's stated PV inverter lifetime can vary due to the impact of operating site conditions. To address limitations in degradation estimation through accelerated testing, condition monitoring, or degradation modeling, we propose a machine learning (ML) oriented approach. Utilizing data from a 1.4 MW PV power plant operational since 2016, with 46 string PV inverters tied to the grid, we employ the unsupervised one-class support vector machine ML technique to analyze inverter and sensor data, capable of classifying humidity cycling and temperature fluctuations as dominant failure mechanisms. Utilizing the anomaly alert relationship and alert details specific to the inverter, the level of PV inverter output is considered as its availability or available reliability. Subsequently, a continuous Markov model is applied to six-month alert data, revealing an average stated reliability of 20% after 20 years of continuous operation... Read More

Source

The solar power plant is a rapidly growing renewable energy source that has a potential role in reducing climate change and replacing fossil fuels. Estimation of the power generated by a solar power plant is required to determine the energy supply. Unfortunately, the solar power generated is highly uncertain due to highly dependence to nature, such as solar radiation and weather. This makes the estimation of solar power generation to be very difficult. This study presents a development of machine learning to model a solar power plant for estimating the generated power. The machine learning is developed by implementing the k-NN algorithm. A data set of power generated in a solar power plant is applied to build the machine learning. The development resulted in a machine learning that models the solar power plant. Simulation test result show the machine learning was able to estimate the solar power generated with an accuracy of 69.6%. The developed model is very useful to estimate potential of solar power resource in a region. The developed model is very useful in feasibility studies to... Read More

Source

A method for determining the operating state of a photovoltaic array based on real-time output current values of photovoltaic strings. The method involves acquiring present output current values, determining a target abnormality score predicting model, and comparing the predicted abnormality score with a threshold to determine the operating state of the array. The method improves accuracy by filtering out virtually high current values and using a machine learning-based abnormality score predicting model.

Source

Solar photovoltaic (PV) power prediction plays a pivotal role in optimizing energy management within the re-newable energy industry. In this investigation, we explore the utilization of artificial neural networks (ANNs) to model solar PV ageing and, at the same time, forecast power generation. Diverse factors impacting power output are examined, and multiple ANNs are explored for prediction purposes. Real-world PV power data is collected and subjected to preprocessing to facilitate the training and testing of ANNs, including recurrent neural networks, autoencoders, and convolutional neural networks. The findings demonstrate the accurate short-term forecasting capa-bilities of ANNs, with particular emphasis on Long Short-term Memory (LSTM) networks. Additionally, the study delves into the effects of panel ageing on PV power by leveraging machine learning models and data analysis, leading to the identification of effective performance degradation prediction. The dataset is further segmented into subsets representing sunny and cloudy conditions, and employing separate models for each su... Read More

Source

A method to accurately determine the operating state of a solar power generation system with multiple panels. The method involves grouping the panels based on the variation in their output readings, calculating a reference value for each group, and comparing it to the target group's output to determine its state. By grouping panels with similar output trends, it provides more accurate analysis compared to comparing all panels together.

Source

Solar photovoltaic (PV) power forecasting is a crucial aspect of efficient energy management in the renewable energy sector. This study examines the use of artificial neural networks (ANNs) to forecast solar PV power output. It considers various factors influencing power output and investigates different ANNs for prediction. Real-world PV power data is collected and preprocessed for training and testing ANNs such as recurrent neural networks, autoencoders, and convolutional neural networks. The results show that ANNs, particularly Long Short-term memory (LSTM), accurately forecast PV power output in the short term. The study also analyzes the impact of panel ageing on PV power using machine learning models, revealing effective prediction of performance degradation. Clustering the dataset into sunny and cloudy subsets, and using separate models for each subset improves prediction accuracy. The study presents a comprehensive analysis of ANNs for PV power forecasting and the influence of panel ageing, highlighting the potential of machine learning for precise and reliable predictions.

Source

Among renewable energies, photovoltaic power generation, which can be introduced relatively easily in buildings and houses, is being used, and its further introduction is desired. Therefore, there is a need for technology to accurately predict the amount of electricity generated at potential sites for photovoltaic power generation facilities. In this study, we tried various machine learning methods for predicting the amount of electricity generated by photovoltaic power generation without using the information of the solar radiation meters, and examined the effect of the training period of machine learning on the accuracy of the prediction.

Source

This paper presents a data analysis of energy production by photovoltaic systems at Prince of Songkla University, Surat Thani Campus, with a focus on the installation of a solar cell system on the campus roof. The energy generated by this system is transmitted to the main electrical control cabinet of the building. However, in case of a malfunction in the solar cell set, inspecting and identifying the cause of the abnormality can be challenging and time-consuming. Therefore, the use of current measuring devices is necessary to assess the operation of each inverter set in the photovoltaic system, facilitating the identification and prediction of abnormalities within the solar cell set. To achieve this, we conducted a linear regression analysis employing a machine learning model. Our findings indicate that the Mean Absolute Error (MAE) model is the most suitable machine learning model for determining the efficiency of the photovoltaic system in terms of current sensing, with analysis results of Phase A = 0.8414, Phase B = 1.0128, and Phase C = 0.8129. Additionally, for voltage sensing,... Read More

Source

Determining photovoltaic array operating state through real-time data analysis, eliminating the need for infrared imaging. The method compares current and irradiance values from the array's output strings against standard parameters, using filtered historical data to enhance accuracy. The analysis eliminates false alarms by removing high current values and high discrete rates, ensuring reliable detection of operational states.

Source

A method for monitoring solar panel performance that separates short-term and long-term variations. The method uses a trained model to simulate average panel performance based on environmental parameters, and compares actual performance to the simulated average to detect long-term degradation. The model is trained on historical data and can account for factors like temperature and irradiation, enabling detection of degradation events like soiling or UV damage.

Source

Photovoltaic (PV) systems have become ubiquitous worldwide, harnessing solar energy to generate power on a global scale. Solar output power is varying and largely dependent on environmental factors. These factors consist of humidity, wind speed, irradiance, and the PV module surface temperature, etc. Since photovoltaic generation is not constant in nature, it is critical to prepare for contingencies related to its generation, as the prediction of solar power plant output is important for the electric grid to be designed to accommodate any fluctuations. This paper delves into an examination of how diverse environmental factors impact the output of a solar PV system. Additionally, it explores the application of various Machine Learning (ML) algorithms, including Linear regression, Polynomial regression, Ridge regression, and Lasso regression techniques. Recurrent Neural Network are also implemented on the dataset discussed in [10] to forecast the solar power output. The effectiveness of these techniques is evaluated using nRMSE values. Out of these techniques, polynomial regression out... Read More

Source

A photovoltaic power generation monitoring system that determines the need for maintenance by analyzing power generation data from solar panels using machine learning. The system detects power reduction events, classifies them based on factors such as power change rate, duration, and spatial range, and generates event information indicating whether maintenance is required.

Source

A computer-implemented method for optimizing photovoltaic system performance through real-time monitoring and predictive maintenance. The method processes performance data from heterogeneous sources, compares it to reference performance, detects degradation, and identifies root causes. Based on a system data model, the method generates recommendations to address identified issues, which are output to operators and controllers. The system continuously monitors data sources and assesses recommendation reliability.

Source

Understanding and quantifying the performance of PV plants is crucial for predicting energy production, optimizing maintenance activities to cost-effectively increase the energy production, and improving future designs and construction. PV plant performance data and weather inputs along with models to generate expected power are the basis of performance analysis. This work evaluates the accuracy of expected power models across sources of weather input data and model types. Variations include free and commercial satellite-based and ground-based weather data. For PV plant models machine learning models are compared with physics-based and simple regression models. Additionally, the sensitivity of the resulting performance loss rate (PLR) calculation to different weather inputs is studied.

Source

A photovoltaic power generation system abnormality determination device that improves determination accuracy by calculating a performance index based on power generation data from a target unit and a reference year, rather than relying solely on current power generation levels. The device detects abnormalities in the target unit based on the calculated performance index, enabling accurate determination even when units are replaced or degraded over time.

Source

Photovoltaics supply a growing share of power to the electric grid worldwide. To mitigate resource intermittency issues, these systems are increasingly being paired with electrochemical energy storage devices, e.g., Li-ion batteries, for which ensuring long and safe operation is critical. However, in this operation framework, secondary Li-ion batteries undergo sporadic usage, which prevents the application of standard diagnostic methods. Here, we propose a diagnostic methodology that uses machine learning algorithms trained directly on data obtained from photovoltaic charging of Li-ion batteries. The training is carried out on synthetic voltage data at various degradation conditions calculated from clear sky model irradiance data. The method is validated using synthetic voltage responses calculated from plane of array irradiance observations for a photovoltaic system located in Maui, HI, USA. We report an average root mean square error of 2.75% obtained for more than 10,000 different degradation paths with 25% or less degradation on the Li-ion cells.

Source

A method and apparatus for determining the state of a photovoltaic power station, enabling unified maintenance of multiple devices. The method includes analyzing the operation health of each device, classifying the overall station health based on device conditions, and displaying the station health status through a user interface. The apparatus includes a data processing module, a health analysis module, and a displaying module, which work together to provide a comprehensive view of the station's operational status.

Source

With the continuous development of photovoltaic power generation technology, the problems of intermittence and randomness of photovoltaic power generation become prominent. Therefore, the connection of the photovoltaic system to the grid will impact the stability of the power system and power dispatching. If the photovoltaic power generation can be accurately predicted, it will improve the coordination of power generation of the photovoltaic system and the stability of the power grid after the system grid connection. In a photovoltaic system, there are many factors affecting photovoltaic power, and there are different algorithms for power prediction. In this paper, long short-term memory (LSTM) is used to predict the power generation of the photovoltaic power system. LSTM can learn the correlation features of the time series data without the problems of data gradient disappearance of the traditional recurrent neural network algorithm. The prediction results are then directly applied to the existing integrated photovoltaic power storage system. Through the experiments, it is verified ... Read More

Source

A fault detection system for solar arrays uses machine learning and neural networks to identify and classify faults in photovoltaic (PV) systems. The system analyzes real-time measurements from individual PV modules, including voltage, current, temperature, and irradiance, to detect faults such as ground faults, series and parallel arc faults, open circuits, short circuits, soiling, and partial shading. The system employs semi-supervised learning and neural networks to classify faults, including those that are difficult to detect using traditional methods, such as soiling and short circuits. The system enables real-time fault detection and identification, reducing mean time to repair (MTTR) and improving the efficiency and reliability of utility-scale PV arrays.

Source

This paper proposes a new prognostics analysis approach for power electronics by combining physics-based and data-driven techniques. Starting with Weibull degradation model, machine learning (ML) techniques are applied to degradation data progressively for continuous reliability monitoring and predictive maintenance decision-making. No prior knowledge of components or mission profiles is required for model training and prediction. Extracted features from analysis can be used to cluster the iteration-based predictions effectively. Another advantage is abrupt change of operation condition can be captured through machine learning for potential lifetime improvement through predictive maintenance. Proposed method can be generalized to other hardware components beyond power electronics.

Source

System and method for monitoring solar panel performance in a solar power plant to identify defects and quantify energy losses. The system uses aerial vehicles to capture visual and thermographic images of the solar panels, which are then stitched together to create orthomosaic images. Computer vision models and object detection techniques are used to identify defects such as hotspots, bypass diode activation, and module shorts. The system also captures real-time electrical parameters from inverters and string monitoring boxes to analyze underperforming components and calculate energy losses.

Source

Solar photo voltaic (PV) energy system backbone of the renewable energy system. Energy system is depended on weather conditions such as temperature and radiation intensity. The role of machine learning (ML) for solar energy generation and radiation forecasting. This paper presents ML algorithm or methods review for prediction of solar energy generation and radiation. This paper also presents the state of art on different ML methods and parameters for forecasting solar energy production and radiation.

Source

Method for detecting potential-induced degradation (PID) of photovoltaic (PV) modules in PV installations. The method employs a series of discrete-point measurements at specific voltage points, with a focus on the parallel resistance (Rpar) between the PV module and the inverter. The measurements are performed during operation, using a controlled voltage approach that maintains consistent ambient conditions. By analyzing the voltage-current characteristic at these points, the method identifies deviations that correspond to potential degradation, allowing early detection and potential repair.

Source

The power production of photovoltaic plants can be affected throughout its operational lifetime by multiple losses and degradation mechanisms. Although long-term degradation has been widely studied, most methodologies assume a specific degradation behaviour and require detailed metadata. This paper presents a methodology for the calculation of long-term degradation of a photovoltaic plant based on neural networks. The goal of the neural network is to model the photovoltaic plant's power production as a function of environmental conditions and time elapsed since the plant started operating. A big advantage of this method with respect to others is that it is completely data-driven, requires no additional information, and makes no assumptions related to degradation behaviour. Results show that the model can derive a long-term degradation trend without overfitting to shorter-term effects or abrupt changes in year-to-year operation.

Source

Degradation is a technical and market hurdle in the development of novel photovoltaics and other energy devices. Understanding and addressing degradation requires complex, time-consuming measurements on multiple samples. To address this challenge, we present \textit{DeepDeg}, a machine learning model that combines deep learning, explainable machine learning, and physical modeling to: 1) forecast hundreds of hours of degradation, and 2) explain degradation in novel photovoltaics. Using a large and diverse dataset of over 785 stability tests of organic solar cells, totaling 230,000 measurement hours, DeepDeg is able to accurately predict degradation dynamics and explain the physiochemical factors driving them using few initial hours of degradation. We use cross-validation and a held-out dataset of over 9,000 hours of degradation of PCE10:OIDTBR to evaluate our model. We demonstrate that by using DeepDeg, degradation characterization and screening can be accelerated by 5-20x.

Source

Photovoltaic (PV) energy development has increased in the last years mainly based on large utility-scale plants. These plants are characterised by a huge number of panels connected to high-power inverters occupying a large land area. An accurate estimation of the power production of the PV plants is needed for failure detection, identifying production deviations, and the integration of the plants into the power grid. Various studies have used Machine Learning estimation techniques developed on very small PV plants. This paper deals with large utility-scale plants and uses all the available information to represent the non-uniform radiation over the whole studied solar field. Variables measured in up to four meteorological stations and distributed across the plant are used. Three PV plants with 1, 2 and 4 meteorological stations have been used to develop Machine Learning models. The hyperparameters were systematically optimised, demonstrating the improvements by comparing with a simple model based on Multiple Linear Regression. The best results were obtained with the Random Forest tec... Read More

Source

Degradation rate in Solar photovoltaic systems is truly an important factor affecting economic feasibility. A linear drop in performance is typically assumed during the operational life-time of the PV system. However, the operational ground performance data reveal that the degradation rate for PVmodule based systems is often non-linear similar to the present investigation. This, if neglected, can cause financial uncertainty. So, this paper presents an experimental field estimation coupled with an analytical model-based approach for prediction of degradation of Solar PV systems. A section of 380 kWp PV system is considered for the present work. Furthermore, comparison of regressors encompassing non-linearities like random forest and decision tree-based are attempted for model formulation. The work is also supported with a comprehensive review on modelling approaches for assessment of degradation-based failure modes in Solar PV system. The experimental average value of degradation rate of the 380 kWp PV system is 7.78% or 0.0778. The reported analytical models possess a close match wit... Read More

Source

Solar power-based photovoltaic energy conversion could be considered one of the best sustainable sources of electric power generation. Thus, the prediction of the output power of the photovoltaic panel becomes necessary for its efficient utilization. The main aim of this paper is to predict the output power of solar photovoltaic panels using different machine learning algorithms based on the various input parameters such as ambient temperature, solar radiation, panel surface temperature, relative humidity and time of the day. Three different machine learning algorithms namely, multiple regression, support vector machine regression and gaussian regression were considered, for the prediction of output power, and compared on the basis of results obtained by different machine learning algorithms. The outcomes of this study showed that the multiple linear regression algorithm provides better performance with the result of mean absolute error, mean squared error, coefficient of determination and accuracy of 0.04505, 0.00431, 0.9981 and 0.99997 respectively, whereas the support vector machi... Read More

Source

A photovoltaic module fault detection method and device that uses IV curve analysis to identify faults in photovoltaic modules. The method involves collecting IV curves from multiple modules, normalizing and fitting the curves to obtain a reference curve, and then comparing the characteristic information of the reference curve with the characteristic information of the test curve to detect faults such as open circuits, diode shorts, component misconfigurations, potential-induced degradation, dust accumulation, component aging, shadows, and hotspots.

Source