1,721,007 research outputs found
Non-Iterative methods for the extraction of the single-diode model parameters of photovoltaic modules: a review and comparative assessment
Full text linkThe extraction of the photovoltaic (PV) model parameters remains to this day a long-standing and popular research topic. Numerous methods are available in the literature, widely differing in accuracy, complexity, applicability, and their very nature. This paper focuses on the class of non-iterative parameter extraction methods and is limited to the single-diode PV model. These approaches consist of a few straightforward calculation steps that do not involve iterations; they are generally simple and easy to implement but exhibit moderate accuracy. Seventeen such methods are reviewed, implemented, and evaluated on a dataset of more than one million measured I-V curves of six different PV technologies provided by the National Renewable Energy Laboratories (NREL). A comprehensive comparative assessment takes place to evaluate these alternatives in terms of accuracy, robustness, calculation cost, and applicability to different PV technologies. For the first time, the irregularities found in the extracted parameters (negative or complex values) and the execution failures of these methods are recorded and are used as an assessment criterion. This comprehensive and up-to-date literature review will serve as a useful tool for researchers and engineers in selecting the appropriate parameter extraction method for their application
An efficient MPPT algorithm for partially shaded PV strings
No full textUnder partial shading conditions, several power peaks (maximum power points - MPPs) are presented on the P-V curve of a photovoltaic system, hindering the effectiveness of typical maximum power point tracking (MPPT) algorithms, due to possible convergence to a local suboptimal MPP. In this paper, a global MPPT (GMPPT) method for PV strings is proposed, which exploits the theoretical MPP characterization to detect the shading conditions and estimate all MPPs on the P-V curve. The calculations performed do not involve unnecessary operating point variations and output power fluctuations. The proposed method is designed for PV strings illuminated at two irradiance levels and only needs the standard voltage and current sensors of the DC/DC converter
A method for the analytical extraction of the single-diode PV model parameters
No full textDetermination of PV model parameters usually requires time consuming iterative procedures, prone to initialization and convergence difficulties. In this paper, a set of analytical expressions is introduced to determine the five parameters of the single-diode model for crystalline PV modules at any operating conditions, in a simple and straightforward manner. The derivation of these equations is based on a newly found relation between the diode ideality factor and the open circuit voltage, which is explicitly formulated using the temperature coefficients. The proposed extraction method is robust, cost-efficient, and easy-to-implement, as it relies only on datasheet information, while it is based on a solid theoretical background. Its accuracy and computational efficiency is verified and compared to other methods available in the literature through both simulation and outdoor measurements.</p
Modeling and analysis of zig-zag boost converter for battery charging applications
No full textIn a DC micro-grid (MG), the energy storage system (ESS) plays a vital role in doing the generation-load power management, DC bus voltage control and/or the power smoothening to provide a quality power to the load. In this paper, a new Bi-directional Boost converter topology denoted as Zig-Zag Boost converter (ZZB) is proposed as a charge controller for energy storage system for DC MG. The superiority of the ZZB over the conventional boost converter is that it can provide a comparatively higher gain with a better efficiency for the same duty cycle. Although this topology uses an extra switch and an additional inductor, the voltage and current stress of the switching and magnetic components is reduced, as well as the size of the inductors, resulting in lower overall cost and higher efficiency. Small deviations in the inductors and switches characteristics are perfectly acceptable, as the mismatch current is self-regulated to zero. This paper provides the complete model and an efficiency assessment of ZZB over other common topologies in the literature, while the results are validated via simulations in MATLAB/SIMULINK.</p
Load profiling using grey relational analysis for power system state estimation
No full textPower system state estimation (PSSE) is a critical tool for power system operation. Load/generation profiles are essential for performing accurate PSSE, and enable PSSE algorithms to handle errors in real and pseudo measurements. The classic approach of modelling the measurement error in PSSE is applying the mixture reduction algorithm to Gaussian mixture models (GMMs) fitted to existent load/generation profiles. However, this approach has inherent limitations in the mixing process. We propose a novel algorithm based on grey relational analysis (GRA) to derive a smaller load/generation profile from the original extensive profiles based on the forecast results of the next day. Our algorithm addresses the issues of mixture reduction and is applied before mixture reduction to improve PSSE accuracy. A case study is presented to evaluate the performance of the proposed algorithm in improving the accuracy of PSSE
Simple PV Performance Equations Theoretically Well Founded on the Single-Diode Model
Full text linkThere are several photovoltaic (PV) performance models in the literature, but most of them either employ complex and tedious calculations or require additional measurements apart from datasheet information. In this paper, a new set of performance equations to evaluate the short-circuit current, open-circuit voltage, and maximum power point at any operating conditions is introduced. The proposed expressions are simple functions of the irradiance and temperature, while they are generally applicable to any crystalline PV module and require only datasheet information as input data. This is achieved by introducing new formulas to determine the irradiance and temperature coefficients that are not provided in the datasheet, thus avoiding empirical constants or additional measurements. The novelty of the performance equations is their solid theoretical background, as they are in excellent agreement with the single-diode PV model, combined with simple and easy application. The proposed PV model is validated and compared with other methods found in the literature through simulations in MATLAB and outdoor measurements on commercial PV modules
An algorithm to detect partial shading conditions in a PV system
No full textMultiple local maxima (MPPs) are presented on the P-V curve of a PV system under partial shading conditions. In general, standard maximum power point tracking (MPPT) algorithms have trouble locating the global maximum, leading often to suboptimal operation at a local MPP, and thus to decreased efficiency. In commercial inverters, this situation is mitigated by performing periodically scans of the characteristic curve to relocate the global MPP. However, this procedure entails fluctuation of the power output, as well as, short-term power losses. To limit these implications, a simple algorithm is introduced in this paper, which mathematically determines whether a PV system is partially shaded or not, thus avoiding unnecessary curve scans at uniform illumination. The proposed method needs only datasheet information and a temperature sensor, while it is applicable to any PV topology, multiple irradiance levels and is readily implemented as an enhancement to any existing MPPT algorithm.</p
Data-driven modelling of the irradiance and temperature effect on the photovoltaic model parameters
No full textThe irradiance and temperature impact on the photovoltaic(PV) model has been extensively modelled in the past using analytical physics-based expressions. However,such methods tend to under perform under low irradiance conditions and certain PV technologies. This paper explores the potential of four machine learning (ML) alternatives on this task, i.e. Neural Networks, Support Vector Machine, K Nearest Neighbours, and Extreme Gradient Boosting. This analysis is performed on datasets of mono- and poly- crystalline PV modules provided by NREL. The findings demonstrate that ML methods generally outperform the widely adopted analytical approach, particularly in the parasitic resistances representation. Extreme Gradient Boosting is found to be the front-runner, while the results also indicate potential for transferability of models trained on one PV module to another of the same PV technology
Network-agnostic adaptive PQ adjustment control for grid voltage regulation in PV systems
No full textThe service of grid voltage regulation is required nowadays from inverter-based resources (IBRs) particularly at the lower voltage level. In the transmission network, this is easily managed by leveraging solely the reactive power (Q) capability of the IBR, but in distribution networks that are mix of L and r the voltage magnitude is coupled with both active (P) and reactive power injection. There are methods in the literature designed for these cases that utilize both P and Q in voltage regulation, but they usually require network and load data, which may not be readily available. Furthermore, they often disregard an apparent nonmonotonic relation between the inverter terminal voltage and the P/Q ratio risking instability. To fill this gap, this article introduces a network-Agnostic P-Q adjustment technique for photovoltaic (PV) systems or other IBRs. The proposed technique tracks in a step-like manner the reference voltage set point if it is feasible, or the maximum grid voltage otherwise. This allows identification of the critical P/Q ratio without any prior information at the cost of limited voltage ripple due to a variable step-size strategy implemented. The superior performance of the proposed scheme is validated through simulations in MATLAB-Simulink in a reduced UKGDS 95-bus system and through lab experiments on a scaled down laboratory grade prototype.</p
Statistical Analysis of Solar Irradiance Variability
No full textSolar photovoltaic (PV) generation forecasting is an important tool to power system operators, but struggles under conditions of intermittent solar irradiance. Although studying and forecasting irradiance itself has been the subject of muchresearch, little progress has been made on the variability (or fluctuation) of irradiance and its statistical properties, despite it being an important parameter in generation forecasting, state estimation and other power system applications. This paper takes a close look into the statistical nature of irradiance variability and shows that it can be sufficiently modeled by a Gaussian Mixture Model (GMM) of six components. Furthermore, an investigation on the required time resolution demonstrates that sub-minute resolution is necessary to accurately capture irradiance variability.The analysis is performed on a one-second resolution irradiancedataset provided by NREL
- …
