Georgi Hristov Yordanov

Overirradiance (Cloud Enhancement) Events at High Latitudes

Contrary to intuition, solar irradiance peaks at partially cloudy conditions. Clouds can boost sunlight by over 1.5 times, even at high latitudes. Depending on cloud velocity, the bursts last from seconds to minutes. Measuring irradiance on a tilted surface with 10-ms resolution allows for a detailed study of such events in Southern Norway, almost at sea level. All monthly maxima from April through September 2011 exceeded 1300 W/m2. The slow sensor registered an annual maximum of 1413 W/m2, while the fast sensors range was found insufficient. A burst reaching 1528 W/m2 was registered in June 2012. Near the Equator, bursts exceeding 1800 W/m2 have been observed. These numbers are striking since the extraterrestrial solar irradiance peaks in January at about 1400 W/m2. The phenomenon is attributed mainly to forward scattering of light in optically thin clouds (adjacent to the sun), which is much stronger for angles within 5° around the solar disk.

Development of a test station for accurate in situ I-V curve measurements of photovoltaic modules in Southern Norway

The development of an outdoor test station for accurate in situ I-V curve measurements of photovoltaic (PV) modules is described. The modules are installed in an open-rack configuration at the University of Agder in Southern Norway. Seven new and three aged PV modules of different type and make are being tested, including mono-and multicrystalline silicon from differing manufacturing routes, triple-junction amorphous silicon, and CIS. Data acquisition is controlled with a multichannel electronic load system and LabVIEW software, recording high-resolution I-V curves at one-minute intervals. Between I-V curve sweeps, each module is operated at the maximum power point. Characteristic electrical parameters are extracted and stored together with values of module temperatures and in-plane solar irradiance, the latter recorded at sub-second resolution. The paper describes the experimental set-up in more detail, discusses some local environmental effects, and presents a sample of test results.

100-millisecond Resolution for Accurate Overirradiance Measurements

Cloud enhancement of sunlight results in peak irradiance well exceeding extraterrestrial levels, even at high latitudes. Values above 1.8 kW/m2 are possible in the equatorial regions. Recently, we detected bursts over 1.5 kW/m2 in Northern Europe at latitude close to 60°N. Overirradiance events (intensities much higher than 1 sun) can last tens of minutes, as well as less than 1 s. They may have caused series arcing in photovoltaic modules, leading to fires and loss of property. The accurate measurement of short bursts requires sensors with response times on the order of milliseconds. The long response times of thermopile pyranometers smooth out important details of very short-lived peaks and may cause underestimations of more than 30%. However, sampling with very high resolution can result in very large data files. Thus, the question of an optimal sampling rate arises. This paper will show a method to answer this question. We propose two simple equations that relate the optimal sampling and averaging times to the second temporal derivative of irradiance at the narrowest peak. We studied records with 10-ms resolution from Southern Norway and arrived at an upper limit of 132 ms for accuracy level of 10 W/m2.

Contributions to the knowledge base on PV performance: Evaluation of the operation of PV systems using different technologies installed in southern Norway

To assist in establishing an accepted knowledge base on PV-modules and systems performance using a representative range of technologies, devices have to be installed at diverse locations, covering a broad range of environmental conditions. For the example of a high latitude location, modules and systems are installed and under investigation in southern Norway (Kristiansand region) by the University of Agder in cooperation with industrial partners. This paper presents first results of the analysis of module performance. The operational behavior of the modules is used to derive a modeling scheme applicable for performance prediction. This use is demonstrated by giving the expected annual performance of different module technologies for a set of sites in southern Norway.

Physically-consistent parameterization in the modeling of solar photovoltaic devices

This research tests the standard one-diode model of a crystalline-Si photovoltaic cell, focusing on the physical accuracy. In particular, the (apparent) shunt resistance and the diode ideality factor are studied. Current-voltage characteristics of illuminated crystalline-Si photovoltaic modules are analyzed, and some limits of applicability of the standard model are given. Typical values of the ideality factor for crystalline-Si devices are derived from own experimental data as well as from recently published literature. It is shown that the contribution of the apparent shunt resistance is only significant for cell voltages below about 0.45 V, and depends on irradiance. This result is consistent with earlier research. Some reference books on Photovoltaics give a wrong shape of the electrical characteristic based on a non-physical interpretation of the shunt resistance. This paper may be particularly useful for power electronics engineers designing inverters and maximum-power-point tracking algorithms.

PV modules with variable ideality factors

The classical single-exponential I-V curve model for crystalline-Si PV devices assumes a constant ideality factor. The model is widely applied in e.g. modeling the power output of PV systems at different irradiances and temperatures, and also in calculating the Equivalent Cell Temperature (ECT) from the open-circuit voltage. The present paper presents results from several crystalline-Si (c-Si) module types suggesting that the ideality factor may change with irradiance. The possible implications of such variability are discussed, in particular the need for a new ECT calculation procedure as well as for revision of some PV performance models.

Modeling and parameter identification of crystalline silicon photovoltaic devices

This paper tests the standard single-exponential model of the electrical characteristics of crystalline-Si photovoltaic devices, focusing on the (apparent) shunt current. Measured characteristics of illuminated polycrystalline-Si photovoltaic modules are modeled, and the apparent shunt current is analyzed. It is shown that an Ohmic-like behavior only takes place at voltages well below the maximum-power point. At higher voltages, the apparent shunt current quickly drops to negligible values. Modeling a crystalline-Si PV device with a fixed shunt resistance may therefore lead to underestimation of the maximum power exceeding 10% at certain irradiance levels.

Extracting parameters from semi-log plots of polycrystalline silicon PV modules outdoor I–V data: Double-exponential model revisited

This paper presents a method for extracting physically meaningful parameters from measured I–V curves of PV modules. The 7-parameter double-exponential model is applied in the modeling. The method is based on linear fitting of semi-logarithmic plots. The paper demonstrates a new technique to estimate the series resistance of a module with high accuracy from such plots. As a result, also the reverse saturation current and the quality factor of the diffusion diode can be determined. The method is applied to outdoor I–V data from a test station with three similar, but not identical, polycrystalline-Si modules. The values of the series resistances found with this method deviate somewhat from the values found by indoor measurements by an independent laboratory. The quality factors of the diffusion diodes were in this case found to be somewhat larger than 1, indicating a good, but not perfect quality of the material.

Ideality factor behavior between the maximum power point and open circuit

The local ideality factor analysis of dark and light I-V curves has been used in the past to study various performance degradation effects in solar cells. Trapping, edge recombination and injection-level-dependent recombination are expressed as “lumps, humps and bumps” in the plots of the local ideality factor over cell voltage (m-V plots). Earlier applications of this differential technique did not correct the plots for the series resistance effect. Thus, the bumps at the higher voltages introduced by some mechanisms were more difficult to quantify. A possible solution is to analyze ISC-VOC curves, but their measurement is not always possible. We present a formula for calculation of the RS-corrected local ideality factor. The m-I plot can be more informative than the m-V plot which varies strongly with the cell temperature. We present results from several crystalline-silicon (c-Si) PV technologies showing the m-I plot behavior between the MPP and open circuit. For some of the tested modules, the ideality factor is much higher at the MPP.

Optimal temporal resolution for detailed studies of cloud-enhanced sunlight (Overirradiance)

Enhancement of sunlight by clouds can lead to irradiance peaks much exceeding the extraterrestrial levels - close to 2 suns near the Equator and at least 1.5 suns at latitudes of about 60°. Some extreme overirradiance events can last many minutes, but durations in the order of 1 second are also possible. The present paper reports the shortest bursts recorded in Southern Norway in the years 2012 and 2013. Our records of 10-millisecond resolution from 2012 show that the optimal instantaneous irradiance sampling interval is less than 0.15 s at the present test site, while the optimal averaging time is less than 0.13 s. We propose simple equations for deriving these times in an arbitrary geographical location from data recorded with very fine temporal resolution. The small values in the present coastal site are attributed to rather high cloud speeds.