Electric demand prediction using artificial neural network technology
- Science Applications International Corp., San Diego, CA (United States)
As a means of promoting demand-side management (DSM) technologies, electric utilities have developed increasingly complex electric rate structures. Electric rates are typically based on both demand and energy use and, in some instances, can change on an hourly basis. The ability of a building's owner or operator to react to the variability of these rates would be greatly enhanced if a building's electric demand and energy use could be accurately predicted on a daily basis. This is especially true for buildings that are equipped with thermal energy storage (TES) systems for building cooling. TES systems are designed to shift the electric demand associated with building cooling to night-time hours when electric rates are usually lowest. TES systems are typically designed to provide the maximum benefit under design day weather and building usage conditions. As a result, TES systems are often under-utilized (with an associated reduction in savings) during time periods when less than design day conditions exist. To optimize TES system equipment operation, it is first necessary to predict building electric and cooling demand under non-design day conditions. A personal computer-based software package that operates in conjunction with a building's energy management and control system (EMCS) to automatically optimize TES system operation is currently installed and operating in an office building in the northeastern United States. This software package uses artificial neural network (ANN) technology to model several parameters related to building energy use and TES system operation. The purpose of this article is to report on the initial performance of the artificial neural network in its prediction of building electric load.
- OSTI ID:
- 5967119
- Journal Information:
- ASHRAE Journal (American Society of Heating, Refrigerating and Air-Conditioning Engineers); (United States), Vol. 35:3; ISSN 0001-2491
- Country of Publication:
- United States
- Language:
- English
Similar Records
Hourly thermal load prediction for the next 24 hours by ARIMA, EWMA, LR and an artificial neural network
A supervisory controller for optimization of building central cooling systems
Related Subjects
OFFICE BUILDINGS
POWER DEMAND
FORECASTING
ARTIFICIAL INTELLIGENCE
ENERGY EFFICIENCY
NEURAL NETWORKS
OFF-PEAK ENERGY STORAGE
PERSONAL COMPUTERS
THERMAL ENERGY STORAGE EQUIPMENT
BUILDINGS
COMPUTERS
DEMAND
DIGITAL COMPUTERS
EFFICIENCY
ENERGY STORAGE
EQUIPMENT
MICROCOMPUTERS
STORAGE
320105* - Energy Conservation
Consumption
& Utilization- Building Services- (1987-)
320603 - Energy Conservation
Consumption
& Utilization- Municipalities & Community Systems- Public Utilities- (1980-)