Answers are grounded in this paper and cite their sources.
Trending using curve fitting to past small area load growth is an economical but not highly accurate method of forecasting loads for distribution planning. A new trending method uses clustering of historical load at the small area level as the forecast algorithm. This new method has significant performance advantages over normal curve fit methods and can be implemented in 16 bit integer arithmetic on most computers. The method and test results are discussed in detail.
No takes yet. Share an insight, caveat, or question.
Willis et al. (1983) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: