Disaggregate Model to Forecast Transformer Usage

Disaggregate Model to Forecast Transformer Usage

Matthew Roman (University of Missouri, USA) and Wooseung Jang (University of Missouri, USA)
Copyright: © 2014 |Pages: 14
DOI: 10.4018/978-1-4666-5202-6.ch070

Chapter Preview


Data Preparation

Two databases of usage information were provided by the company, such as an inventory database and a Distribution Operations Job Management (DOJM) database. All transformers issued from inventory over the past five years were in the inventory database. However, there was no way to separate the data based on the reasons the transformers were used. The DOJM database was much more informative because it provided project activity codes and showed installation records of transformers, which allowed the separation of transformer usage into the three subcategories, i.e., NC, SE, and GM. Unfortunately, the DOJM database showed lower usage levels than the inventory database in almost every month because of reporting errors and delays during the installation operation. Figure 1 compares the monthly usage levels reported in the inventory and DOJM databases over the provided five years (60 months) of data.

Key Terms in this Chapter

Median Absolute Percentage Error (MdAPE): The middle value of all the percentage errors for a data set when the absolute values of the errors are ordered by size.

Exponential Smoothing: A technique that can be applied to time series data, either to produce smoothed data for presentation, or to make forecasts.

Economic Indicator: A statistic about the economy. Economic indicators allow analysis of economic performance and predictions of future performance.

Transformer: A static electrical device that transfers energy by inductive coupling between its winding circuits. Transformers range in size from a thumbnail-sized coupling transformer hidden in microphones to units weighing hundreds of tons used in electrical substations at power generation stations and to interconnect the power grid.

Triple Exponential Smoothing: An exponential smoothing model taking into account seasonal changes as well as trends.

Leading Indicator: A measurable economic factor that changes before the economy starts to follow a particular pattern or trend. Leading indicators are used to predict changes in the economy, but are not always accurate.

Mean Absolute Percentage Error (mape): The mean or average of the absolute percentage errors of forecasts, also known as mean absolute percentage deviation (MAPD). It is a measure of accuracy of a method for constructing fitted time series values in statistics, specifically in trend estimation. This measure is easy to understand because it provides the error in terms of percentages.

Simple Average: The simplest way to smooth a time series to calculate a simple, or un-weighted, moving average.

Complete Chapter List

Search this Book: