An effort to optimize similar days parameters for ANN based electricity price forecasting

Paras Mandal, Anurag K. Srivastava, Michael Negnevitsky, Jung Wook Park

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents a sensitivity analysis of similar days (SD) parameters to increase the accuracy of artificial neural network (ANN) and SD based short-term price forecasting. Presented work is an extended version of previous works done by authors to integrate ANN and similar days method for predicting electricity price. Focus here is on sensitivity analysis of similar days parameters while keeping the parameters same for ANN to forecast hourly electricity prices in the PJM (regional transmission organization in north-east America) electricity market. Several cases are simulated by choosing: (a) two; (b) three; (c) four; and (d) five similar days parameters to calculate the norm. Additionally, sensitivity analysis has been carried out by changing time framework of similar days (d=15, 30, 45, 60) and number of selected similar price days (N=5, 10). From sensitivity analysis, it is identified that the optimized mean absolute percentage error (MAPE) is obtained using case-c with d=30 and N=10. MAPE of reasonably small value along with forecast mean square error (FMSE) and mean absolute error (MAE) of around 2$/MWh and 1$/MWh are obtained for the PJM data, which has correlation coefficient of determination (R2) of 0.7758 between load and electricity price. Numerical results show that forecasts generated by developed ANN model based on the optimized case are accurate and efficient.

Original languageEnglish
Title of host publication2008 IEEE Industry Applications Society Annual Meeting, IAS'08
DOIs
Publication statusPublished - 2008 Dec 30
Event2008 IEEE Industry Applications Society Annual Meeting, IAS'08 - Edmonton, AB, Canada
Duration: 2008 Oct 52008 Oct 9

Other

Other2008 IEEE Industry Applications Society Annual Meeting, IAS'08
CountryCanada
CityEdmonton, AB
Period08/10/508/10/9

Fingerprint

Sensitivity analysis
Electricity
Neural networks
Mean square error

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Cite this

Mandal, P., Srivastava, A. K., Negnevitsky, M., & Park, J. W. (2008). An effort to optimize similar days parameters for ANN based electricity price forecasting. In 2008 IEEE Industry Applications Society Annual Meeting, IAS'08 [4658929] https://doi.org/10.1109/08IAS.2008.141
Mandal, Paras ; Srivastava, Anurag K. ; Negnevitsky, Michael ; Park, Jung Wook. / An effort to optimize similar days parameters for ANN based electricity price forecasting. 2008 IEEE Industry Applications Society Annual Meeting, IAS'08. 2008.
@inproceedings{3c0764be2e7a4c4ab4cc605e1df0292a,
title = "An effort to optimize similar days parameters for ANN based electricity price forecasting",
abstract = "This paper presents a sensitivity analysis of similar days (SD) parameters to increase the accuracy of artificial neural network (ANN) and SD based short-term price forecasting. Presented work is an extended version of previous works done by authors to integrate ANN and similar days method for predicting electricity price. Focus here is on sensitivity analysis of similar days parameters while keeping the parameters same for ANN to forecast hourly electricity prices in the PJM (regional transmission organization in north-east America) electricity market. Several cases are simulated by choosing: (a) two; (b) three; (c) four; and (d) five similar days parameters to calculate the norm. Additionally, sensitivity analysis has been carried out by changing time framework of similar days (d=15, 30, 45, 60) and number of selected similar price days (N=5, 10). From sensitivity analysis, it is identified that the optimized mean absolute percentage error (MAPE) is obtained using case-c with d=30 and N=10. MAPE of reasonably small value along with forecast mean square error (FMSE) and mean absolute error (MAE) of around 2$/MWh and 1$/MWh are obtained for the PJM data, which has correlation coefficient of determination (R2) of 0.7758 between load and electricity price. Numerical results show that forecasts generated by developed ANN model based on the optimized case are accurate and efficient.",
author = "Paras Mandal and Srivastava, {Anurag K.} and Michael Negnevitsky and Park, {Jung Wook}",
year = "2008",
month = "12",
day = "30",
doi = "10.1109/08IAS.2008.141",
language = "English",
isbn = "9781424422791",
booktitle = "2008 IEEE Industry Applications Society Annual Meeting, IAS'08",

}

Mandal, P, Srivastava, AK, Negnevitsky, M & Park, JW 2008, An effort to optimize similar days parameters for ANN based electricity price forecasting. in 2008 IEEE Industry Applications Society Annual Meeting, IAS'08., 4658929, 2008 IEEE Industry Applications Society Annual Meeting, IAS'08, Edmonton, AB, Canada, 08/10/5. https://doi.org/10.1109/08IAS.2008.141

An effort to optimize similar days parameters for ANN based electricity price forecasting. / Mandal, Paras; Srivastava, Anurag K.; Negnevitsky, Michael; Park, Jung Wook.

2008 IEEE Industry Applications Society Annual Meeting, IAS'08. 2008. 4658929.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - An effort to optimize similar days parameters for ANN based electricity price forecasting

AU - Mandal, Paras

AU - Srivastava, Anurag K.

AU - Negnevitsky, Michael

AU - Park, Jung Wook

PY - 2008/12/30

Y1 - 2008/12/30

N2 - This paper presents a sensitivity analysis of similar days (SD) parameters to increase the accuracy of artificial neural network (ANN) and SD based short-term price forecasting. Presented work is an extended version of previous works done by authors to integrate ANN and similar days method for predicting electricity price. Focus here is on sensitivity analysis of similar days parameters while keeping the parameters same for ANN to forecast hourly electricity prices in the PJM (regional transmission organization in north-east America) electricity market. Several cases are simulated by choosing: (a) two; (b) three; (c) four; and (d) five similar days parameters to calculate the norm. Additionally, sensitivity analysis has been carried out by changing time framework of similar days (d=15, 30, 45, 60) and number of selected similar price days (N=5, 10). From sensitivity analysis, it is identified that the optimized mean absolute percentage error (MAPE) is obtained using case-c with d=30 and N=10. MAPE of reasonably small value along with forecast mean square error (FMSE) and mean absolute error (MAE) of around 2$/MWh and 1$/MWh are obtained for the PJM data, which has correlation coefficient of determination (R2) of 0.7758 between load and electricity price. Numerical results show that forecasts generated by developed ANN model based on the optimized case are accurate and efficient.

AB - This paper presents a sensitivity analysis of similar days (SD) parameters to increase the accuracy of artificial neural network (ANN) and SD based short-term price forecasting. Presented work is an extended version of previous works done by authors to integrate ANN and similar days method for predicting electricity price. Focus here is on sensitivity analysis of similar days parameters while keeping the parameters same for ANN to forecast hourly electricity prices in the PJM (regional transmission organization in north-east America) electricity market. Several cases are simulated by choosing: (a) two; (b) three; (c) four; and (d) five similar days parameters to calculate the norm. Additionally, sensitivity analysis has been carried out by changing time framework of similar days (d=15, 30, 45, 60) and number of selected similar price days (N=5, 10). From sensitivity analysis, it is identified that the optimized mean absolute percentage error (MAPE) is obtained using case-c with d=30 and N=10. MAPE of reasonably small value along with forecast mean square error (FMSE) and mean absolute error (MAE) of around 2$/MWh and 1$/MWh are obtained for the PJM data, which has correlation coefficient of determination (R2) of 0.7758 between load and electricity price. Numerical results show that forecasts generated by developed ANN model based on the optimized case are accurate and efficient.

UR - http://www.scopus.com/inward/record.url?scp=57949102976&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57949102976&partnerID=8YFLogxK

U2 - 10.1109/08IAS.2008.141

DO - 10.1109/08IAS.2008.141

M3 - Conference contribution

SN - 9781424422791

BT - 2008 IEEE Industry Applications Society Annual Meeting, IAS'08

ER -

Mandal P, Srivastava AK, Negnevitsky M, Park JW. An effort to optimize similar days parameters for ANN based electricity price forecasting. In 2008 IEEE Industry Applications Society Annual Meeting, IAS'08. 2008. 4658929 https://doi.org/10.1109/08IAS.2008.141