Holmgren, William F.; Andrews, Robert W.; Lorenzo, Antonio T.; Stein, Joshua S.
We describe improvements to the open source PVLIB-Python modeling package. PVLIB-Python provides most of the functionality of its parent PVLIB-MATLAB package and now follows standard Python design patterns and conventions, has improved unit test coverage, and is installable. PVLIBPython is hosted on GitHub.com and co-developed by GitHub contributors. We also describe a roadmap for the future of the PVLIB-Python package.
Sandia National Laboratories (Sandia) manages four of the five PV Regional Test Centers (RTCs). This report reviews accomplishments made by the four Sandia-managed RTCs during FY2015 (October 1, 2014 to September 30, 2015) as well as some programmatic improvements that apply to all five sites. The report is structured by Site first then by Partner within each site followed by the Current and Potential Partner summary table, the New Business Process, and finally the Plan for FY16 and beyond. Since no official SOPO was ever agreed to for FY15, this report does not include reporting on specific milestones and go/no-go decisions.
PV performance models are used to quantify the value of PV plants in a given location. They combine the performance characteristics of the system, the measured or predicted irradiance and weather at a site, and the system configuration and design into a prediction of the amount of energy that will be produced by a PV system. These predictions must be as accurate as possible in order for finance charges to be minimized. Higher accuracy equals lower project risk. The Increasing Prediction Accuracy project at Sandia focuses on quantifying and reducing uncertainties in PV system performance models.
The Characterizing Emerging Technologies project focuses on developing, improving and validating characterization methods for PV modules, inverters and embedded power electronics. Characterization methods and associated analysis techniques are at the heart of technology assessments and accurate component and system modeling. Outputs of the project include measurement and analysis procedures that industry can use to accurately model performance of PV system components, in order to better distinguish and understand the performance differences between competing products (module and inverters) and new component designs and technologies (e.g., new PV cell designs, inverter topologies, etc.).
IEEE Standard 1547-2003 conformance of several interconnected microinverters was performed by Sandia National Laboratories (SNL) to determine if there were emergent adverse behaviors of co-located aggregated distributed energy resources. Experiments demonstrated the certification tests could be expanded for multi-manufacturer microinverter interoperability. Evaluations determined the microinverters' response to abnormal conditions in voltage and frequency, interruption in grid service, and cumulative power quality. No issues were identified to be caused by the interconnection of multiple devices.
The proper modeling of Photovoltaic(PV) systems is critical for their financing, design, and operation. PV LIB provides a flexible toolbox to perform advanced data analysis and research into the performance modeling and operations of PV assets, and this paper presents the extension of the PV LIB toolbox into the python programming language. PV LIB provides a common repository for the release of published modeling algorithms, and thus can also help to improve the quality and frequency of model validation and inter comparison studies. Overall, the goal of PV LIB is to accelerate the pace of innovation in the PV sector.
The Sandia Inverter Performance Test Protocol defined two possible weighted-average efficiency values for use in comparing inverter performance, of which one definition was selected by the California Energy Commission for use in their Buydown incentive program leading to widespread use in the photovoltaic inverter market. This paper discusses the derivation of the efficiency weights originally proposed, and investigates the potential for defining new weights in light of increased array-to-inverter (DC-to-AC) system rating ratios in modern PV systems.
We present a method for measuring the series resistance of the PV module, string, or array that does not require measuring a full IV curve or meteorological data. Our method relies only on measurements of open circuit voltage and maximum power voltage and current, which can be readily obtained using standard PV monitoring equipment; measured short circuit current is not required. We validate the technique by adding fixed resistors to a PV circuit and demonstrating that the method can predict the added resistance. Relative prediction accuracy appears highest for smaller changes in resistance, with a systematic underestimation at larger resistances. Series resistance is shown to vary with irradiance levels with random errors below 1.5% standard deviation.
The performance of photovoltaic systems must be monitored accurately to ensure profitable long-term operation. The most important signals to be measured—irradiance and temperature, as well as power, current and voltage on both DC and AC sides of the system—contain rapid fluctuations that are not observable by typical monitoring systems. Nevertheless these fluctuations can affect the accuracy of the data that are stored. This report closely examines the main signals in one operating PV system, which were recorded at 2000 samples per second. It analyzes the characteristics and causes of the rapid fluctuations that are found, such as line-frequency harmonics, perturbations from anti-islanding detection, MPPT searching action and others. The operation of PV monitoring systems is then simulated using a wide range of sampling intervals, archive intervals and filtering options to assess how these factors influence data accuracy. Finally several potential sources of error are discussed with real-world examples.
It is important to be able to accurately simulate the variability of solar PV power plants for grid integration studies. We aim to inform integration studies of the ease of implementation and application-specific accuracy of current PV power plant output simulation methods. This report reviews methods for producing simulated high-resolution (sub-hour or even sub-minute) PV power plant output profiles for variability studies and describes their implementation. Two steps are involved in the simulations: estimation of average irradiance over the footprint of a PV plant and conversion of average irradiance to plant power output. Six models are described for simulating plant-average irradiance based on inputs of ground-measured irradiance, satellite-derived irradiance, or proxy plant measurements. The steps for converting plant-average irradiance to plant power output are detailed to understand the contributions to plant variability. A forthcoming report will quantify the accuracy of each method using application-specific validation metrics.
Imagery from GOES satellites is analyzed to determine how solar variability is related to the NOAA classification of cloud type. Without using a model to convert satellite imagery to average insolation on the ground, this paper investigates using cloud categories to directly model the expected statistical variability of ground irradiance. Hourly cloud classified satellite images are compared to multiple years of ground measured irradiance at two locations to determine if measured irradiance, ramp rates, and variability index are correlated with cloud category. Novel results are presented for ramp rates grouped by the cloud category during the time period. This correlation between satellite cloud classification and solar variability could be used to model the solar variability for a given location and time and could be used to determine the variability of a location based on the prevalence of each cloud category.
Module temperature is modeled using a transient heat-flow model. Module temperature predicted in this fashion is important in the calculation of cell temperature, a vital input in performance modeling. Parameters important to the model are tested for sensitivity, and optimized to a single day of measured module temperature using simultaneous non-linear least squares regression. These optimized parameters are then tested for accuracy using a year's worth of data for one location. The results obtained from this analysis are compared with modeled data from a different site, as well as to results obtained using a steadystate model. We find that the transient model best captures the variability in module temperature, and that the transient model works best when calibrated for a specific location.
Module temperature is modeled using a transient heat-flow model. Module temperature predicted in this fashion is important in the calculation of cell temperature, a vital input in performance modeling. Parameters important to the model are tested for sensitivity, and optimized to a single day of measured module temperature using simultaneous non-linear least squares regression. These optimized parameters are then tested for accuracy using a year's worth of data for one location. The results obtained from this analysis are compared with modeled data from a different site, as well as to results obtained using a steadystate model. We find that the transient model best captures the variability in module temperature, and that the transient model works best when calibrated for a specific location.
Clear sky models estimate the terrestrial solar radiation under a cloudless sky as a function of the solar elevation angle, site altitude, aerosol concentration, water vapor, and various atmospheric conditions. This report provides an overview of a number of global horizontal irradiance (GHI) clear sky models from very simple to complex. Validation of clear-sky models requires comparison of model results to measured irradiance during clear-sky periods. To facilitate validation, we present a new algorithm for automatically identifying clear-sky periods in a time series of GHI measurements. We evaluate the performance of selected clear-sky models using measured data from 30 different sites, totaling about 300 site-years of data. We analyze the variation of these errors across time and location. In terms of error averaged over all locations and times, we found that complex models that correctly account for all the atmospheric parameters are slightly more accurate than other models, but, primarily at low elevations, comparable accuracy can be obtained from some simpler models. However, simpler models often exhibit errors that vary with time of day and season, whereas the errors for complex models vary less over time.
Sandia National Laboratories (Sandia) and SunPower Corporation (SunPower) have completed design and deployment of an autonomous irradiance monitoring system based on wireless mesh communications and a battery operated data acquisition system. The Lanai High-Density Irradiance Sensor Network is comprised of 24 LI-COR{reg_sign} irradiance sensors (silicon pyranometers) polled by 19 RF Radios. The system was implemented with commercially available hardware and custom developed LabVIEW applications. The network of solar irradiance sensors was installed in January 2010 around the periphery and within the 1.2 MW ac La Ola PV plant on the island of Lanai, Hawaii. Data acquired at 1 second intervals is transmitted over wireless links to be time-stamped and recorded on SunPower data servers at the site for later analysis. The intent is to study power and solar resource data sets to correlate the movement of cloud shadows across the PV array and its effect on power output of the PV plant. The irradiance data sets recorded will be used to study the shape, size and velocity of cloud shadows. This data, along with time-correlated PV array output data, will support the development and validation of a PV performance model that can predict the short-term output characteristics (ramp rates) of PV systems of different sizes and designs. This analysis could also be used by the La Ola system operator to predict power ramp events and support the function of the future battery system. This experience could be used to validate short-term output forecasting methodologies.
This report describes in-depth analysis of photovoltaic (PV) output variability in a high-penetration residential PV installation in the Pal Town neighborhood of Ota City, Japan. Pal Town is a unique test bed of high-penetration PV deployment. A total of 553 homes (approximately 80% of the neighborhood) have grid-connected PV totaling over 2 MW, and all are on a common distribution line. Power output at each house and irradiance at several locations were measured once per second in 2006 and 2007. Analysis of the Ota City data allowed for detailed characterization of distributed PV output variability and a better understanding of how variability scales spatially and temporally. For a highly variable test day, extreme power ramp rates (defined as the 99th percentile) were found to initially decrease with an increase in the number of houses at all timescales, but the reduction became negligible after a certain number of houses. Wavelet analysis resolved the variability reduction due to geographic diversity at various timescales, and the effect of geographic smoothing was found to be much more significant at shorter timescales.
We present an approach to simulate time-synchronized, one-minute power output from large photovoltaic (PV) generation plants in locations where only hourly irradiance estimates are available from satellite sources. The approach uses one-minute irradiance measurements from ground sensors in a climatically and geographically similar area. Irradiance is translated to power using the Sandia Array Performance Model. Power output is generated for 2007 in southern Nevada are being used for a Solar PV Grid Integration Study to estimate the integration costs associated with various utility-scale PV generation levels. Plant designs considered include both fixed-tilt thin-film, and single-axis-tracked polycrystalline Si systems ranging in size from 5 to 300 MW{sub AC}. Simulated power output profiles at one-minute intervals were generated for five scenarios defined by total PV capacity (149.5 MW, 222 WM, 292 MW, 492 MW, and 892 MW) each comprising as many as 10 geographically separated PV plants.
During the development of a solar photovoltaic (PV) energy project, predicting expected energy production from a system is a key part of understanding system value. System energy production is a function of the system design and location, the mounting configuration, the power conversion system, and the module technology, as well as the solar resource. Even if all other variables are held constant, annual energy yield (kWh/kWp) will vary among module technologies because of differences in response to low-light levels and temperature. A number of PV system performance models have been developed and are in use, but little has been published on validation of these models or the accuracy and uncertainty of their output. With support from the U.S. Department of Energy's Solar Energy Technologies Program, Sandia National Laboratories organized a PV Performance Modeling Workshop in Albuquerque, New Mexico, September 22-23, 2010. The workshop was intended to address the current state of PV system models, develop a path forward for establishing best practices on PV system performance modeling, and set the stage for standardization of testing and validation procedures for models and input parameters. This report summarizes discussions and presentations from the workshop, as well as examines opportunities for collaborative efforts to develop objective comparisons between models and across sites and applications.
Design and operation of the electric power grid (EPG) relies heavily on computational models. High-fidelity, full-order models are used to study transient phenomena on only a small part of the network. Reduced-order dynamic and power flow models are used when analysis involving thousands of nodes are required due to the computational demands when simulating large numbers of nodes. The level of complexity of the future EPG will dramatically increase due to large-scale deployment of variable renewable generation, active load and distributed generation resources, adaptive protection and control systems, and price-responsive demand. High-fidelity modeling of this future grid will require significant advances in coupled, multi-scale tools and their use on high performance computing (HPC) platforms. This LDRD report demonstrates SNL's capability to apply HPC resources to these 3 tasks: (1) High-fidelity, large-scale modeling of power system dynamics; (2) Statistical assessment of grid security via Monte-Carlo simulations of cyber attacks; and (3) Development of models to predict variability of solar resources at locations where little or no ground-based measurements are available.
Photovoltaic systems are often priced in $/W{sub p}, where Wp refers to the DC power rating of the modules at Standard Test Conditions (1000 W/m{sup 2}, 25 C cell temperature) and $ refers to the installed cost of the system. However, the true value of the system is in the energy it will produce in kWhs, not the power rating. System energy production is a function of the system design and location, the mounting configuration, the power conversion system, and the module technology, as well as the solar resource. Even if all other variables are held constant, the annual energy yield (kWh/kW{sup p}) will vary among module technologies because of differences in response to low-light levels and temperature. Understanding energy yield is a key part of understanding system value. System performance models are used during project development to estimate the expected output of PV systems for a given design and location. Performance modeling is normally done by the system designer/system integrator. Often, an independent engineer will also model system output during a due diligence review of a project. A variety of system performance models are available. The most commonly used modeling tool for project development and due diligence in the United States is probably PVsyst, while those seeking a quick answer to expected energy production may use PVWatts. In this paper, we examine the variation in predicted energy output among modeling tools and users and compare that to measured output.