Antcliffe, David John (2017) Novel Techniques for Gas Demand Modeling and Forecasting. MPhil thesis, University of Sheffield.
Abstract
The ability to provide accurate forecasts of future gas demand has a major impact on several business processes for Gas Regions in the UK and elsewhere in the world. Long term forecasts provide the guidance for major structural needs, while short term forecasts guide the operations management on requirements of supply purchase, supply storage and delivery. Accurate forecasts guarantee optimum and safe gas supply at the lowest cost. Currently there is no single technique that produces the perfect forecast, this research will attempt to improve on current methods by applying Non-Linear techniques. The technique to be tested is defined as ”Non-Linear Autoregressive Moving Average with eXogeneous Inputs, polynomials, and a Forward Regression with Orthogonal Least Squares estimation procedure”. The goal of the research is is to produce a Mean Average Percentage Error of between 4-6% or better, which was proposed by DNV GL (supplier of software for the Gas Industry), as a valid level of error to make any new methodology of value.
Metadata
Supervisors: | Harrison, Robert |
---|---|
Keywords: | ARIMAX NARIMAX NARMAX ARMAX "Short Term Gas Demand Modeling and Forecasting" |
Awarding institution: | University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Automatic Control and Systems Engineering (Sheffield) |
Depositing User: | Mr David John Antcliffe |
Date Deposited: | 04 Sep 2017 07:55 |
Last Modified: | 04 Sep 2018 00:18 |
Download
MyMaster 20170809-to-publish
Filename: MyMaster 20170809-to-publish.pdf
Licence:
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License
Export
Statistics
You do not need to contact us to get a copy of this thesis. Please use the 'Download' link(s) above to get a copy.
You can contact us about this thesis. If you need to make a general enquiry, please see the Contact us page.