Application of multiple linear regression to analysis of data from factory energy surveys

Publisher Website
Google Scholar
Abstract
Summary: When energy surveys are carried out in factories, the usual method is to install expensive metering equipment and to measure energy flows to various parts of the plant. The costs involved in conducting surveys by this method are often prohibitively high. This paper describes an alternative survey method which is simple and inexpensive to use. An estimate of energy use for each product is obtained by the use of multiple linear regression to relate total energy consumption data from existing meters to production of the various products. The accuracy of the estimates depends on the form of the data set collected, but in most factories accurate values can be obtained for the major products. In all seventy‐three factories where the technique was applied, it led to a useful analysis of total energy consumption. A summary of data collected in the New Zealand food manufacturing industry by application of the method is presented.

This publication has 1 reference indexed in Scilit: