|Title:||A comparative analysis of forecasting methodologies of electricity consumption and application in Hong Kong system|
|Subject:||Electric power consumption -- China -- Hong Kong -- Forecasting|
Hong Kong Polytechnic University -- Dissertations
Department of Electrical Engineering
|Pages:||100 leaves : ill. ; 30 cm|
|Abstract:||Power system planning and decision support is a complex process requiring reliable and accurate forecasting data for future power consumption. Power infrastructure is characterised by high fixed cost, long construction lead time, long-life, slow supply adjustment to actual consumption. All these special features of power infrastructure put an extreme pressure on getting quality forecasting data for future power consumption. This paper contains a brief history of Hong Kong electricity consumption over the past few decades, description of three major, forecasting models, analysis on the accuracy of forecasting models and finally the conclusion. Two major types of quantitative forecasting models used within power industry are time-series and multiple linear regression (MLR) model. On the other hand, artificial neural network (ANN) is another popular forecasting model to predict future power consumption recently. In order to test the applicability and accuracy of these forecasting models in relation to the forecast of power consumption, two approaches were adopted, namely, total market forecast and combined market forecast. Total market forecast refers to the prediction of total electricity consumption which consists of the consumption in domestic, commercial, industrial, street lighting and export to China markets. Combined market forecast, unlike the total market forecast, divides the total market into individual market. We predict the forecasts for each market and add them to obtain the forecast of total electricity consumption combined from those individual markets. The result revealed that time series model should mainly be used to establish past trends to forecast the future whereas multiple linear regression model is often more suitable in the sense that the accuracy of forecasted data is higher. We also found that artificial neural network model forecast is generally as good as those generated by the traditional multiple linear regression model.|
|Rights:||All rights reserved|
Files in This Item:
|b15320996.pdf||For All Users (off-campus access for PolyU Staff & Students only)||3 MB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: