You need JavaScript to view this

Estimation of clearness index using neural network with meteorological forecast; Kisho yoho wo nyuryoku toshita neural network ni yoru seiten shisu no yosoku

Abstract

Discussions were given on estimation of clearness index in order to operate stably a solar energy utilizing system. All-sky insolation amount varies not only by change in the climate, but also seasonal change in the sun`s altitude. Therefore, a clearness index (ratio of all-sky insolation to out-of-atmosphere insolation) was used. The larger the value, the higher the solar ray permeability. The all-sky insolation amount is a measured value, while the out-of-atmosphere insolation amount is a calculated value. Although the clearness index may be roughly estimated by weather forecast, the clearness index varies largely even on the same weather forecast, especially for cloudy days, if a weather forecast actually having error is used. Therefore, discussions were given on estimation of the clearness index by using a neural network which uses meteorological information such as air temperatures and precipitation probabilities as inputs. Using multiple number of meteorological forecast information simultaneously has reduced the average square error to 49% of that using only the weather forecast. The estimation accuracy depends on the accuracy of meteorological forecast, but using multiple number of forecast information can improve the accuracy. 6 refs., 7 figs., 1 tab.
Authors:
Nishimura, S; Kenmoku, Y; Sakakibara, T; [1]  Nakagawa, S; [2]  Kawamoto, T [3] 
  1. Toyohashi University of Technology, Aichi (Japan)
  2. Maizuru National College of Technology, Kyoto (Japan)
  3. Shizuoka University, Shizuoka (Japan)
Publication Date:
Nov 25, 1997
Product Type:
Conference
Report Number:
ETDE/JP-98753622; CONF-9711143-
Reference Number:
SCA: 580000; 140100; PA: JP-98:0G1092; EDB-98:076895; SN: 98001948782
Resource Relation:
Conference: 1997 JSES/JWEA joint conference, Taiyo/furyoku energy koen, Aichi (Japan), 28-29 Nov 1997; Other Information: PBD: 25 Nov 1997; Related Information: Is Part Of Proceedings of JSES/JWEA Joint Conference (1997); PB: 454 p.; Taiyo/Furyoku energy koen ronbunshu (1997)
Subject:
58 GEOSCIENCES; 14 SOLAR ENERGY; METEOROLOGY; FORECASTING; NEURAL NETWORKS; AMBIENT TEMPERATURE; PROBABILITY; RAIN; WEATHER; CLOUD COVER; SOLAR ENERGY; INSOLATION; SUN CHARTS; SEASONAL VARIATIONS; LIGHT TRANSMISSION; ERRORS; CLOUDS; LEAST SQUARE FIT
OSTI ID:
625356
Research Organizations:
Japan Solar Energy Society, Tokyo (Japan)
Country of Origin:
Japan
Language:
Japanese
Other Identifying Numbers:
Other: ON: DE98753622; TRN: JN98G1092
Availability:
Available from Japan Solar Energy Society, 44-14, Yoyogi 2-chome, Shibuya-ku, Tokyo,(Japan); OSTI as DE98753622
Submitting Site:
NEDO
Size:
pp. 305-308
Announcement Date:

Citation Formats

Nishimura, S, Kenmoku, Y, Sakakibara, T, Nakagawa, S, and Kawamoto, T. Estimation of clearness index using neural network with meteorological forecast; Kisho yoho wo nyuryoku toshita neural network ni yoru seiten shisu no yosoku. Japan: N. p., 1997. Web.
Nishimura, S, Kenmoku, Y, Sakakibara, T, Nakagawa, S, & Kawamoto, T. Estimation of clearness index using neural network with meteorological forecast; Kisho yoho wo nyuryoku toshita neural network ni yoru seiten shisu no yosoku. Japan.
Nishimura, S, Kenmoku, Y, Sakakibara, T, Nakagawa, S, and Kawamoto, T. 1997. "Estimation of clearness index using neural network with meteorological forecast; Kisho yoho wo nyuryoku toshita neural network ni yoru seiten shisu no yosoku." Japan.
@misc{etde_625356,
title = {Estimation of clearness index using neural network with meteorological forecast; Kisho yoho wo nyuryoku toshita neural network ni yoru seiten shisu no yosoku}
author = {Nishimura, S, Kenmoku, Y, Sakakibara, T, Nakagawa, S, and Kawamoto, T}
abstractNote = {Discussions were given on estimation of clearness index in order to operate stably a solar energy utilizing system. All-sky insolation amount varies not only by change in the climate, but also seasonal change in the sun`s altitude. Therefore, a clearness index (ratio of all-sky insolation to out-of-atmosphere insolation) was used. The larger the value, the higher the solar ray permeability. The all-sky insolation amount is a measured value, while the out-of-atmosphere insolation amount is a calculated value. Although the clearness index may be roughly estimated by weather forecast, the clearness index varies largely even on the same weather forecast, especially for cloudy days, if a weather forecast actually having error is used. Therefore, discussions were given on estimation of the clearness index by using a neural network which uses meteorological information such as air temperatures and precipitation probabilities as inputs. Using multiple number of meteorological forecast information simultaneously has reduced the average square error to 49% of that using only the weather forecast. The estimation accuracy depends on the accuracy of meteorological forecast, but using multiple number of forecast information can improve the accuracy. 6 refs., 7 figs., 1 tab.}
place = {Japan}
year = {1997}
month = {Nov}
}