Simple Statistical Methods for Comparative Evaluation of Air Quality Models

Hanna, Steven R.; Heinold, David W.

doi:10.1007/978-1-4757-9125-9_29

Cited by 14 publications

(15 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These metrics exhibit a greater penalty for substantial errors compared to minor errors, thus rendering them particularly sensitive to outliers. The normalized indicator can also be defined using the corrected indicator suggested by Hanna and Heinold [21]…”

Section: Evaluation Metricsmentioning

confidence: 99%

Machine Learning Based Photovoltaic Power and Energy Prediction in time-frequency domain

aouni,

Naimi,

Ayat

2024

Preprint

View full text Add to dashboard Cite

Efficient energy management in smart-grid systems relies heavily on accurate photovoltaic (PV) power production forecasting. In this study, we explore the benefits of employing frequency domain methods for PV panel power production forecasting. moreover two methods are proposed in this article, Discrete Wavelet Transform with Long Short-Term Memory (DWT-LSTM) for short-term forecasting and LSTM combined with Short-Time Fourier Transform and an Artificial Neural Network (LSTM-STFT-ANN) for long-term forecasting. to assess the robustness of the proposed methods, extensive testing and comparison was conducted with the state of the art methods LSTM combined with a conventional neural network (CNN-LSTM) on an unknown dataset spanning two years, enabling prediction performance evaluation under varying weather conditions, including foggy and cloudy scenarios. Additionally, we evaluated the total energy produced across different time horizons. The performance assessment results for 1-hour horizon power forecasting revealed that the DWT-LSTM method achieved an average normalized root mean squared error (HH) of approximately 0.16, outperforming the classical CNN-LSTM method, which yielded an HH of more than 0.22. For longer-horizon forecasts, the LSTM-STFT-ANN method exhibited higher efficiency. For instance, for 2-day horizon, the HH indicator was recorded as 0.35 for LSTM-STFT-ANN, compared to around 0.54 for the CNN-LSTM method.

show abstract

Section: Evaluation Metricsmentioning

confidence: 99%

Machine Learning Based Photovoltaic Power and Energy Prediction in time-frequency domain

aouni,

Naimi,

Ayat

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…statistical tests, t and F. Fox (1981), Hanna (1984), and Dickerson and Ermak (1988) have suggested that these statistics can be used to evaluate dispersion models. There have been many published air-quality studies that employ these statistics, such as those reported in the Hilton Head (Buckner, 1980) and Kiawah (Weber and Garrett, 1984) workshops.…”

Section: Comparison Against Observational Datamentioning

confidence: 99%

Evaluation protocol for the WIND system atmospheric models

Fast¹

1991

View full text Add to dashboard Cite

i DISTRfE}LJTION OF THIS DQCUMENT IS UNI.JMFIT_D ! DISCLAIMER • , MASTEB | This report waspreparedas an accountof work_po.nsored by an agencyof the UnitedStates ' i Government.Neither the United StatesGovernmentnorany agencythereof,nor any of their employees, makesany warranty,expresso.rimplied,or assumes any legalliabilityor resporLfi.. bility forthe accuracy,co,mp.leteness, or usefulnessof any information,apparatus,product,o,r process disclosed,or representsthat its use would not infringe privatelyownedrights.Refer. enceh,ereinto any specificccommercial product,process, o: servi.ce by _radename,trademark, i manufacturer, o.rotherwi=se does no.tneoessafilyconstitute or implyits end.otsement, recom-Li mendatio, n, o.rfavoringby the United States Governmentor any agencythereof.T'he views | arid opinions of authors exp:rcsscxJ i_r_,indu ,tui ..... _. UnitedS,tate_ Oc_vernment o,rany agency ther_f.

show abstract

“…The effective stack height due to momentum rise at a downwind distance of two building heights is computed (as in the ISC model) from the formula H=H a (1) where H e is effective stack height, H s is the stack height and AH m is the momentum rise. Letting L B equal the minimum of HB and i?w> where HB is the building height and Hw the projected building width, then the following formula for a z is used:…”

Section: Dispersion Coefficientsmentioning

confidence: 99%

“…To answer this question, two simplified model performance measures are used: the normalized mean square error, and the mean of the highest ten concentrations. 1 Confidence limits on the measures themselves and the differences between the measures for two different models are calculated using the bootstrap procedure. 9 It is suggested that this simple procedure is more useful than the long list of statistical measures recommended by the EPA.…”

Section: Model Performance Measuresmentioning

confidence: 99%