Global horizontal irradiance (GHI) forecasts by numerical weather prediction (NWP) often contain model-led bias. There is thus a strong consensus on using post-processing techniques, such as model output statistics (MOS), to correct such errors. As opposed to the conventional parametric methods, this article considers a nonparametric approach for post-processing, namely, kernel conditional density estimation (KCDE). Essentially, KCDE constructs a relationship between the bias error (difference between the NWP-based GHI forecast and measurement) and NWP output variables, such as clear-sky index, zenith angle, air temperature, humidity, or surface pressure. Hence, when a new set of explanatory variables becomes available, the conditional expectation of the bias error can be estimated. Since the ground-based GHI measurements are not available everywhere, the possibility of using satellite-derived GHI data to correct NWP forecasts is also explored. In the case study, two years of GHI forecasts made using the North American Mesoscale forecast system are corrected using both ground-measured and satellite-derived GHI references. As compared to Lorenz's fourth-degree polynomial MOS, additional 10%–16% (using ground-measured GHI) and 5%–13% (using satellite-derived GHI) reductions in the forecast error are observed at 7 test stations across the continental United States.

1.
D.
Yang
,
J.
Kleissl
,
C. A.
Gueymard
,
H. T.
Pedro
, and
C. F.
Coimbra
, “
History and trends in solar irradiance and PV power forecasting: A preliminary assessment and review using text mining
,”
Sol. Energy
168
,
60
101
(
2018
).
2.
P. A.
Jiminez
,
J. P.
Hacker
,
J.
Dudhia
,
S. E.
Haupt
,
J. A.
Ruiz-Arias
,
C. A.
Gueymard
,
G.
Thompson
,
T.
Eidhammer
, and
A.
Deng
, “
WRF-Solar: Description and clear-sky assessment of an augmented NWP model for solar power prediction
,”
Bull. Am. Meteorol. Soc.
97
,
1249
1264
(
2016
).
3.
S.
Alessandrini
,
L. D.
Monache
,
S.
Sperati
, and
G.
Cervone
, “
An analog ensemble for short-term probabilistic solar power forecast
,”
Appl. Energy
157
,
95
110
(
2015
).
4.
G.
Cervone
,
L.
Clemente-Harding
,
S.
Alessandrini
, and
L. D.
Monache
, “
Short-term photovoltaic power forecasting using artificial neural networks and analog ensemble
,”
Renewable Energy
108
,
274
286
(
2017
).
5.
E.
Lorenz
,
J.
Hurka
,
D.
Heinemann
, and
H. G.
Beyer
, “
Irradiance forecasting for the power prediction of grid-connected photovoltaic systems
,”
IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
2
,
2
10
(
2009
).
6.
S.
Pelland
,
G.
Galanis
, and
G.
Kallos
, “
Solar and photovoltaic forecasting through post-processing of the global environmental multiscale numerical weather prediction model
,”
Prog. Photovoltaics: Res. Appl.
21
,
284
296
(
2013
).
7.
D.
Yang
, “
On post-processing day-ahead NWP forecasts using Kalman filtering
,”
Sol. Energy
182
,
179
181
(
2019
).
8.
M.
Pierro
,
F.
Bucci
,
C.
Cornaro
,
E.
Maggioni
,
A.
Perotto
,
M.
Pravettoni
, and
F.
Spada
, “
Model output statistics cascade to improve day ahead solar irradiance forecast
,”
Sol. Energy
117
,
99
113
(
2015
).
9.
F. J.
Lima
,
F. R.
Martins
,
E. B.
Pereira
,
E.
Lorenz
, and
D.
Heinemann
, “
Forecast for surface solar irradiance at the Brazilian Northeastern region using NWP model and artificial neural networks
,”
Renewable Energy
87
,
807
818
(
2016
).
10.
S.
Sperati
,
S.
Alessandrini
, and
L. D.
Monache
, “
An application of the ECMWF ensemble prediction system for short-term solar power forecasting
,”
Sol. Energy
133
,
437
450
(
2016
).
11.
H.
Verbois
,
A.
Rusydi
, and
A.
Thiery
, “
Probabilistic forecasting of day-ahead solar irradiance using quantile gradient boosting
,”
Sol. Energy
173
,
313
327
(
2018
).
12.

This number is based on the number of models implemented in the caret package in R.

13.
G. M.
Yagli
,
D.
Yang
, and
D.
Srinivasan
, “
Automatic hourly solar forecasting using machine learning models
,”
Renewable Sustainable Energy Rev.
105
,
487
498
(
2019
).
14.
R. J.
Hyndman
,
D. M.
Bashtannyk
, and
G. K.
Grunwald
, “
Estimating and visualizing conditional densities
,”
J. Comput. Graphical Stat.
5
,
315
336
(
1996
).
15.
D.
Ruppert
,
S. J.
Sheather
, and
M. P.
Wand
, “
An effective bandwidth selector for local least squares regression
,”
J. Am. Stat. Assoc.
90
,
1257
1270
(
1995
).
16.
L.
Wasserman
,
All of Nonparametric Statistics
(
Springer Science & Business Media
,
New York
,
2006
).
17.
A. M.
Habte
,
M.
Sengupta
,
A. J.
Lopez
,
Y.
Xie
, and
G. J.
Maclaurin
, “
Assessment of the national solar radiation database (nsrdb 1998-2016)
,”
Technical Report No. NREL/CP-5D00-71607
, National Renewable Energy Laboratory, Golden, CO,
2018
.
18.
R.
Perez
,
J.
Schlemmer
,
K.
Hemker
,
S.
Kivalov
,
A.
Kankiewicz
, and
J.
Dise
, “
Solar energy forecast validation for extended areas and economic impact of forecast accuracy
,” in
2016 IEEE 43rd Photovoltaic Specialists Conference (PVSC)
,
Portland, OR
,
5–10 June
(
2016
), pp.
1119
1124
.
19.
D.
Yang
, “
A correct validation of the National Solar Radiation Data Base (NSRDB)
,”
Renewable Sustainable Energy Rev.
97
,
152
155
(
2018
).
20.
D.
Yang
, “
Kriging for NSRDB PSM version 3 satellite-derived solar irradiance
,”
Sol. Energy
171
,
876
883
(
2018
).
21.
D.
Yang
,
C. A.
Gueymard
, and
J.
Kleissl
, “
Editorial: Submission of Data Article is now open
,”
Sol. Energy
171
,
A1
A2
(
2018
).
22.
D.
Yang
, “
SolarData: An R package for easy access of publicly available solar datasets
,”
Sol. Energy
171
,
A3
A12
(
2018
).
23.
See https://nsrdb.nrel.gov/api-instructions for instructions on downloading data via API.
24.
M.
Rosenblatt
, “
Conditional probability density and regression estimators
,” in
Multivariate Analysis II
, edited by
P. R.
Krishnaiah
(
Academic
,
New York
,
1969
), Vol.
2
, pp.
25
31
.
25.
G.
Lee
,
Y.
Ding
,
M. G.
Genton
, and
L.
Xie
, “
Power curve estimation with multivariate environmental factors for inland and offshore wind farms
,”
J. Am. Stat. Assoc.
110
,
56
67
(
2015
).
26.
C. C.
Taylor
, “
Automatic bandwidth selection for circular density estimation
,”
Comput. Stat. Data Anal.
52
,
3493
3500
(
2008
).
27.

In Eq. (20)), the weight terms can be calculated using Eq. (10), i.e., a total of q-2 times.

28.

The 3rd to qth variables include but are not limited to temperature, humidity, and pressure, which are likely to have different impacts across locations.

29.
J. F.
Mejia
,
M.
Giordano
, and
E.
Wilcox
, “
Conditional summertime day-ahead solar irradiance forecast
,”
Sol. Energy
163
,
610
622
(
2018
).
30.

The next set covers April–June data, then July–September, and finally October–December (see computer code for details). In this way, a total of 8 sets of MOS coefficients are fitted.

31.

If the metrics are computed based on Ghnsrdb, the errors are apparent errors.

32.

The error reduction is computed as, e.g., [nRMSE(LORENZG)−nRMSE(KCDEG)]/nRMSE(LORENZG).

33.
F. X.
Diebold
and
R. S.
Mariano
, “
Comparing predictive accuracy
,”
J. Business Econ. Stat.
13
,
253
263
(
1995
).
34.
D.
Yang
, “
Solar radiation on inclined surfaces: Corrections and benchmarks
,”
Sol. Energy
136
,
288
302
(
2016
).

Supplementary Material

You do not currently have access to this content.