Case Study: Meeting electricity needs with small area peak demand forecasting
Queensland’s rapidly expanding population places extra strain on electricity supply and contributes to peak demand. At Ergon Energy (Ergon), millions of dollars are invested each year in the infrastructure required to support demand during peak times and to avoid power outages. The cost of investing in this infrastructure is one of the major causes of increasing electricity prices.
Total summer demand in Ergon’s network area is forecast to increase by 7.5 percent over the period to 2020 and network capacity augmentation is required to avoid system constraints. Network capacity augmentation needs differ by region yet peak demand forecasting across Australia generally uses whole of state inputs.
Ergon has moved to more granular energy and peak demand forecasting to match the differing levels of activity across Queensland. Ergon has developed customer-driven Peak Demand Forecast models for each of its six network regions using previously unavailable small area data on economic output, air conditioner use, historical summer and winter peak demand, and Bureau of Meteorology data.
To add a further level of granularity, tailored reference sheets were generated for more than 350 Ergon substations which showed information on population, land use, historical demand, temperatures and rainfall. These reference sheets supported Ergon’s Subject Matter Experts in a Delphi process which is used to supplement the mathematical demand forecasts with local knowledge.
This information feeds into Ergon’s network capital planning process to allow more accurate appraisal of when capital equipment is likely to need replacement. The greater granularity of the forecasts allows Ergon to delay some capital expenditure in lower growth regions and bring forward expenditure in higher growth regions. This leads to a network that can maintain the level of supply interruption risk at current levels but at a lower cost of capital investment.
Ergon saves costs from deferring and optimising capital expenditure, which contribute to lower prices for Ergon’s customers.
Other Ergon models which rely on peak demand forecasts, for example modelling of customer demand response from price movements, benefit from improved peak demand modelling.
Aenean leo ligula, porttitor eu, consequat vitae, eleifend ac, enim. Aliquam lorem ante, dapibus in, viverra quis, feugiat a, tellus. Phasellus viverra nulla ut metus varius laoreet. Quisque rutrum. Aenean imperdiet. Etiam ultricies nisi vel augue. Curabitur ullamcorper ultricies nisi.
CEO’S NAME HERE
How can this apply to other businesses?
Businesses should be forward looking when designing their data architecture. Successful businesses understand the links between their data strategy and business strategy.
Successful data models capture data with a consistent structure and process. Failing to do this limits the predictive value in data.
Quality data can create greater value and insights than management anticipates. This is especially true for data-rich businesses such as marketing solutions, energy and telecommunications