Ronald Stewart

Modeling daylight availability and irradiance components from direct and global irradiance

This paper presents the latest versions of several models developed by the authors to predict short time-step solar energy and daylight availability quantities needed by energy system modelers or building designers. The modeled quantities are global, direct and diffuse daylight illuminance, diffuse irradiance and illuminance impinging on tilted surfaces of arbitrary orientation, sky zenith luminance and sky luminance angular distribution. All models are original except for the last one which is extrapolated from current standards. All models share a common operating structure and a common set of input data: Hourly (or higher frequency) direct (or diffuse) and global irradiance plus surface dew point temperature. Key experimental observations leading to model development are briefly reviewed. Comprehensive validation results are presented. Model accuracy, assessed in terms of root-mean-square and mean bias errors, is analyzed both as a function of insolation conditions and site climatic environment.

A NEW SIMPLIFIED VERSION OF THE PEREZ DIFFUSE IRRADIANCE MODEL FOR TILTED SURFACES

A new, more accurate and considerably simpler version of the Perez[1] diffuse irradiance model is presented. This model is one of those used currently to estimate short time step (hourly or less) irradiance on tilted planes based on global and direct (or diffuse) irradiance. It has been shown to perform more accurately than other models for a large number of locations worldwide. The key assumptions defining the model remain basically unchanged. These include (1) a description of the sky dome featuring a circumsolar zone and horizon zone superimposed over an isotropic background, and (2) a parameterization of insolation conditions (based on available inputs to the model), determining the value of the radiant power originating from these two zones. Operational modifications performed on the model are presented in a step by step approach. Each change is justified on the basis of increased ease of use and/or overall accuracy. Two years of hourly data on tilted planes from two climatically distinct sites in France are used to verify performance accuracy. The isotropic, Hay and Klucher models are used as reference. Major changes include (1) the simplification of the governing equation by use of reduced brightness coefficients; (2) the allowance for negative coefficients; (3) reduction of the horizon band to an arc-of-great-circle; (4) optimization of the circumsolar region width; and (5) optimization of insolation conditions parameterization.

An anisotropic hourly diffuse radiation model for sloping surfaces: description, performance validation, site dependency evaluation

Abstract A model is described to estimate hourly or higher frequency diffuse sky radiation impinging on plane surfaces of any orientation, once knowing this value on the horizontal. This model features a simple geometrical sky hemisphere description, allowing for the observed effects of forward-scattered and back-scattered radiation and a parameterization of insolation conditions based on available radiative quantities. Model performance is studied through (1) long term independent tests performed against hourly ground-shielded tilted irradiance data from Trappes, France; Carpentras, France and San Antonio, Texas; (2) long term dependent tests performed against hourly data from the same stations plus Albany, New York; and (3) real time tests based on one-minute data from Albany, New York. Performance is assessed through comparison with three reference models: the isotropic, the Hay, and Klucher anisotropic models. Substantial performance improvement over the three reference models is found for all stations and all surface orientations. Additional performance improvements from independent to dependent testing can be explained logically on the basis of climate, altitude and latitude differences between stations.

Using satellite-derived insolation data for the site/time specific simulation of solar energy systems

In this paper, the question of satellite-derived irradiance is addressed on two levels. First, the question of physical accuracy is addressed by comparing satellite-predicted hourly/daily global and direct irradiance with controlled ground measurements in climatically distinct environments. This accuracy is compared to the error made as a function of distance when extrapolating the needed data from the closest ground measurement site. Second, the question of end-use accuracy is addressed by comparing satellite-derived, photovoltaic-utility load-match benchmarks with actual benchmarks for three US electric utilities where ground measurements were available.

Design and development of a rotating shadowband radiometer solar radiation/daylight network

Abstract Solar radiation data are needed to properly design solar/daylighting systems. The data that exist are often short-term, nonlocal, or of uncertain accuracy. The modest amount of high-quality data that exist were obtained with costly equipment, attentive maintenance, and conscientious quality control. A solar radiation measurement plan is being considered for New York State that seeks to minimize costs by deploying a network of rotating shadowband radiometers. These instruments perform three basis solar measurements (direct beam, diffuse, and global horizontal) with a single detector. Much of the data retrieval and quality control is performed by computer to minimize maintenance costs and downtime. Descriptions of the instrument, network as currently conceived, and data processing plans are presented.

Assessing the load matching capability of photovoltaics for US utilities based upon satellite-derived insolation data

The load matching capability of photovoltaic (PV) power generation is estimated for 20 electric utilities in the continental US. Load matching is determined experimentally by analyzing the interaction between the load requirements of each utility and the output of locally-sited PV plants. PV output is simulated using site/time specific hourly insolation data inferred from geostationary satellite-based remote cloud cover measurement. As quantified with four independent benchmarks, PV load matching capability is found to be substantial for several utilities. A well defined relationship is observed between a utilitys summer-to-winter peak load ratio (SWP ratio) and the load matching capability for that utility. Many of the highest load matching occurrences are found in locations not traditionally targeted for solar energy development, namely: the central US and the Mid-Atlantic seaboard.<<ETX>>

Variations of the luminous efficacy of global and diffuse radiation and zenith luminance with weather conditions—description of a potential method to generate key daylight availability data from existing solar radiation data bases

Abstract The Atmospheric Sciences Research Center and the New York State Energy Research and Development Authority have undertaken a comprehensive daylight resource assessment program which includes the coincident measurement of 25 radiometric and photometric quantities[1]. This paper contains selected results of the analysis performed on collected data. These include the study of the variations of global and diffuse luminous efficacy, and of zenith luminance with insolation conditions. These conditions are described using a parameterization based on two widely available radiative quantities (global and direct irradiance). Based on three 45-day monitoring periods, it appears that many of the observed variations can be accounted for from the knowledge of these two quantities.

Experimental evaluation of a photovoltaic simulation program

A widely used photovoltaic (PV) simulation code, PVFORM, is evaluated in a grid-connected configuration against experimental data from a prototype demand-side management PV array. Taking advantage of the comprehensive array monitoring program, each of the key algorithms composing the simulation code is evaluated independently. PVFORM as a whole was not found to have any major flaws, but was found to overpredict actual power output due mostly to assuming ideal array sun-tracking performance and ideal maximum power point tracking.

MODELING IRRADIANCE ON TILTED PLANES: A SIMPLER VERSION OF THE PEREZ MODEL; US-WIDE CLIMATIC/ENVIRONMENTAL EVALUATION

Preliminary findings of research undertaken in collaboration with Sandia National Laboratories (Menicucci and Fernandez, 1986) for improving photovoltaic simulation are presented. The objective is to systematically evaluate and improve the diffuse model for tilted planes proposed by Perez and coworkers (1986), by (1), simplifying the original model and, (2) constructively investigating site dependency. A new simplified algorithm is provided along with recommendations suggested by the analysis of the first data sets.

Early experiences of the 15 kW NMPC demand-side management photovoltaic project

Abstract The Niagara Mohawk Power Corporation has begun operation of a photovoltaic (PV) system in upstate New York to study the summer peak load reduction capability of grid-connected PV systems serving commercial buildings. The roof-retrofitted system consists of a 151 m 2 polycrystalline silicon module area rated at 15.4 kW d.c., three one-axis trackers, and a high efficiency power conditioning unit. Preliminary results from the first two months of operation indicate PV system output is at a high fraction of capacity when the building experiences its electrical demand peaks. Ongoing studies are evaluating a cross-section of commercial customer load profiles in terms of the probability of peak demand reduction.