Proposed Metric for Evaluation of Solar Forecasting Models

2012 ◽  
Vol 135 (1) ◽  
Author(s):  
Ricardo Marquez ◽  
Carlos F. M. Coimbra
Keyword(s):  
Time Windows ◽  
Correlation Coefficients ◽  
Forecast Model ◽  
Solar Variability ◽  
Time Averaging ◽  
Solar Forecasting ◽  
Model Quality ◽  
Forecasting Models ◽  
Forecasting Error ◽  
Persistence Model

This work presents an alternative metric for evaluating the quality of solar forecasting models. Some conventional approaches use quantities such as the root-mean-square-error (RMSE) and/or correlation coefficients to evaluate model quality. The direct use of statistical quantities to assign forecasting quality can be misleading because these metrics do not convey a measure of the variability of the time-series for the solar irradiance data. In contrast, the quality metric proposed here, which is defined as the ratio of solar uncertainty to solar variability, compares the forecasting error with the solar variability directly. By making the forecasting error to variability comparisons for different time windows, we show that this ratio is essentially a statistical invariant for each forecast model employed, i.e., the ratio is preserved for widely different time horizons when the same time averaging periods are used, and therefore provides a robust way to compare solar forecasting skills. We employ the proposed metric to evaluate two new forecasting models proposed here, and compare their performances with a persistence model.

A Novel Metric for Evaluation of Solar Forecasting Models

2011 ◽  
Author(s):  
Ricardo Marquez ◽  
Carlos F. M. Coimbra
Keyword(s):  
Time Window ◽  
Time Windows ◽  
Correlation Coefficients ◽  
Solar Variability ◽  
Solar Forecasting ◽  
Model Quality ◽  
Quality Metric ◽  
Forecasting Models ◽  
Forecasting Error ◽  
Irradiance Data

This work presents an alternative, time-window invariant metric for evaluating the quality of solar forecasting models. Conventional approaches use statistical quantities such as the root-mean-square-error and/or the correlation coefficients to evaluate model quality. The straightforward use of statistical quantities to assign forecasting quality can be misleading because these metrics do not convey a measure of the variability of the time-series included in the solar irradiance data. In contrast, the quality metric proposed here, which is defined as the ratio of solar uncertainty to solar variability, compares forecasting error with solar variability directly. By making the forecasting error to variability comparisons for different time windows, we show that this ratio is essentially a statistical invariant for each forecasting model employed, i. e., the ratio is preserved for widely different time horizons.

Review of onsite temperature and solar forecasting models to enable better building design and operations

2021 ◽  
Vol 14 (4) ◽  
pp. 885-907
Author(s):  
Bing Dong ◽  
Reisa Widjaja ◽  
Wenbo Wu ◽  
Zhi Zhou
Keyword(s):  
Building Design ◽  
Solar Forecasting ◽  
Forecasting Models

New Key Factors Discovery to Enhance Dengue Fever Forecasting Model

2014 ◽  
Vol 931-932 ◽  
pp. 1457-1461 ◽  
Author(s):  
Phatsavee Ongruk ◽  
Padet Siriyasatien ◽  
Kraisak Kesorn
Keyword(s):  
Wind Speed ◽  
Dengue Fever ◽  
Forecast Model ◽  
Key Factors ◽  
Dengue Virus Infection ◽  
Conventional Model ◽  
Forecasting Error ◽  
Proposed Model ◽  
Basic Set ◽  
Almost All

There are several factors that can be used to predict a dengue fever outbreak. Almost all existing research approaches, however, usually exploit the use of a basic set of core attributes to forecast an outbreak, e.g. temperature, humidity, wind speed, and rainfall. In contrast, this research identifies new attributes to improve the prediction accuracy of the outbreak. The experimental results are analyzed using a correlation analysis and demonstrate that the density of dengue virus infection rate in female mosquitoes and seasons have strong correlation with a dengue fever outbreak. In addition, the research constructs a forecast model using Poisson regression analysis. The result shows the proposed model obtains significantly low forecasting error rate when compared it against the conventional model using only temperature, humidity, wind speed, and rainfall parameters.

STORM SURGE FORECAST MODEL USING GENETIC PROGRAMMING

2020 ◽  
pp. 75-89
Author(s):  
Quyen
Keyword(s):  
Genetic Programming ◽  
Storm Surge ◽  
Storm Surges ◽  
Forecast Model ◽  
Data Sets ◽  
Model Learning ◽  
Forecasting Models ◽  
Accurate Forecast ◽  
Evolving Models ◽  
Storm Surge Forecasting

Stormsurge is a typical genuine fiasco coming from the ocean. Therefore, an accurate forecast of surges is a vital assignment to dodge property misfortunes and decrease the chance of tropical storm surges. Genetic Programming (GP) is an evolution-based model learning technique that can simultaneously find the functional form and the numeric coefficients for the model. Moreover, GP has been widely applied to build models for predictive problems. However, GP has seldom been applied to the problem of storm surge forecasting. In this paper, a new method to use GP for evolving models for storm surge forecasting is proposed. Experimental results on data-sets collected from the Tottori coast of Japan show that GP can become more accurate storm surge forecasting models than other standard machine learning methods. Moreover, GP can automatically select relevant features when evolving storm surge forecasting models, and the models developed by GP are interpretable.

scholarly journals A Short-Term Solar Forecasting Platform Using a Physics-Based Smart Persistence Model and Data Imputation Method

2021 ◽  
Author(s):  
Srinath Yelchuri ◽  
A. Rangaraj ◽  
Yu Xie ◽  
Aron Habte ◽  
Mohit Joshi ◽  
...  
Keyword(s):  
Imputation Method ◽  
Data Imputation ◽  
Solar Forecasting ◽  
Short Term ◽  
Persistence Model

Magnitude Estimation and Onsite Earthquake Early Warning using Cumulative Absolute Velocity in Taiwan

2021 ◽  
Author(s):  
Hao-Yun Huang ◽  
Yih-Min Wu
Keyword(s):  
Early Warning ◽  
Time Windows ◽  
Correlation Coefficients ◽  
Strong Motion ◽  
Earthquake Early Warning ◽  
Absolute Velocity ◽  
Cumulative Absolute Velocity ◽  
Critical Issues ◽  
The Relationship ◽  
Magnitude Determination

<p>Real-time magnitude determination is one of the critical issues for earthquake early warning (EEW). Magnitude determination may have saturation situation using initial seismic signals after an earthquake occurrence. Previous studies utilized eventual cumulative absolute velocity (eCAV) to determine magnitude up to 9.0 without any saturation. However, to determine eCAV will be too late for EEW application. In order to shorten time to obtain eCAV, 4,754 strong motion records from 64 events with M<sub>L </sub>large than 5.5 in Taiwan are used to establish the relationship between eCAV and initial shaking parameters (initial CAV, initial cumulative absolute displacement, initial cumulative absolute integral displacement,  P<sub>d</sub> and  τ<sub>c</sub>) from 1 s to 20 s after P arrival. Our preliminary results show that eCAV can be estimated using initial shaking parameters. Logarithm linear correlation coefficients vary from 0.78 to 0.97 with standard deviations from 0.27 to 0.10 for time windows from 1 s to 20 s after P arrival. Eventually, we can timely estimate eCAV for magnitude determination as well as or on-site EEW purpose.</p>

Hourly Photovoltaics Power Output Prediction for Malaysia Using Support Vector Regression

2015 ◽  
Vol 785 ◽  
pp. 591-595 ◽  
Author(s):  
Kyairul Azmi Baharin ◽  
Hasimah Abd Rahman ◽  
Mohammad Yusri Hassan ◽  
Chin Kim Gan
Keyword(s):  
Support Vector Regression ◽  
Power Output ◽  
Forecast Model ◽  
Support Vector ◽  
Pv System ◽  
Energy Forecasting ◽  
Global Irradiance ◽  
Grid Operators ◽  
Pv Penetration ◽  
Persistence Model

Reliable solar energy forecasting enables grid operators to manage the grid better as PV penetration level increases. This research explores the use of support vector regression to forecast hourly power output from a grid-connected PV system in Malaysia. Data is obtained from a grid-connected PV system that is equipped with a weather monitoring station. Three parameters are used as input to the forecast model; global irradiance, tilted irradiance and ambient temperature. Results were compared against a persistence model. The SVR model manages to forecast hourly power production with satisfactory accuracy.

scholarly journals Ozone and temperature decadal responses to solar variability in the mesosphere and lower thermosphere, based on measurements from SABER on TIMED

2016 ◽  
Vol 34 (1) ◽  
pp. 29-40 ◽  
Author(s):  
F. T. Huang ◽  
H. G. Mayr ◽  
J. M. Russell III ◽  
M. G. Mlynczak
Keyword(s):  
Solar Cycle ◽  
Multiple Regression ◽  
Solar Maximum ◽  
Lower Thermosphere ◽  
Correlation Coefficients ◽  
Solar Variability ◽  
Broadband Emission ◽  
Low Latitudes ◽  
Mesosphere And Lower Thermosphere ◽  
Temperature Responses

Abstract. We have derived ozone and temperature responses to solar variability over a solar cycle, from June 2002 through June 2014, 50 to 100 km, 48° S to 48° N, based on data from the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument on the Thermosphere-Ionosphere-Mesosphere-Energetics and Dynamics (TIMED) satellite. Results with this extent of coverage in the mesosphere and lower thermosphere have not been available previously. A multiple regression is applied to obtain responses as a function of the solar 10.7 cm flux (solar flux units, sfu). Positive responses mean that they are larger at solar maximum than at solar minimum of the solar cycle. From  ∼  80 to 100 km, both ozone and temperature responses are positive for all latitudes and are larger than those at lower altitudes. From  ∼  80 to 100 km, ozone responses can exceed 10 % (100 sfu)−1, and temperature responses can approach 4 °K. From 50 to  ∼  80 km, the ozone responses at low latitudes ( ∼  ±35°) are mostly negative and can approach  ∼  negative 3 % (100 sfu)−1. However, they are mostly positive at midlatitudes in this region and can approach  ∼  2 % (100 sfu)−1. In contrast to ozone, from  ∼  50 to 80 km, the temperature responses at low latitudes remain positive, with values up to  ∼  2.5 K (100 sfu)−1, but are weakly negative at midlatitudes. Consequently, there is a systematic and robust relation between the phases of the ozone and temperature responses. They are positively correlated (in phase) from  ∼  80 to 100 km for all latitudes and negatively correlated (out of phase) from  ∼  50 to 80 km, also for all latitudes. The negative correlation from 50 to 80 km is maintained even though the ozone and temperature responses can change signs as a function of altitude and latitude, because the corresponding temperature responses change signs in step with ozone. This is consistent with the idea that dynamics have the larger influence between  ∼  80 and 100 km, while photochemistry is more in control from  ∼  50 to 75 km. The correlation coefficients between the solar 10.7 cm flux and the ozone and temperature themselves from 2012 to 2014 are positive (negative) in regions where the responses are positive (negative). This supports our results since the correlations are independent of the multiple regression used to derive the responses. We also compare with previous results.

Sign in / Sign up

Export Citation Format

Share Document