Forecasting higher order statistical moments and probability densities for power system loads by the method of time series
Demand forecasting of electric power system loads has been conventionally done by prediction of the expected value of the demand. This mean value being subject to uncertainty does not fully characterize the quantity demanded since a stochastic process is not fully characterized by its expected value. In this thesis, not only the mean value but also the probability density function of power system loads are obtained. In achieving this goal the method of time series is used to predict the mean value and consequently the higher statistical moments. The latter is then used in a Gram-Charlier series to obtain the probability density function of the electrical power system loads. From this density function a wide variety of quantities may be calculated, namely, the mean value, the probability of the load exceeding some threshold, conditional probabilities, conditional expectation, and a figure of confidence of the forecasted mean. An example is given in which data from an actual electrical load are used.
- Research Organization:
- Purdue Univ., Lafayette, IN (USA). School of Electrical Engineering
- DOE Contract Number:
- AS02-77ET29102
- OSTI ID:
- 5189581
- Report Number(s):
- PCTR-92-80; TR-EE-80-18
- Country of Publication:
- United States
- Language:
- English
Similar Records
Monograph on the use of the multivariate Gram Charlier series Type A
Research on stochastic power-flow study methods. Final report
Related Subjects
24 POWER TRANSMISSION AND DISTRIBUTION
29 ENERGY PLANNING, POLICY, AND ECONOMY
292000* -- Energy Planning & Policy-- Supply
Demand & Forecasting
DATA
EXPERIMENTAL DATA
FORECASTING
INFORMATION
MATHEMATICAL MODELS
NUMERICAL DATA
POWER DEMAND
POWER SYSTEMS
PROBABILITY
STATISTICAL MODELS
THEORETICAL DATA