Artificial neural networks and physical modeling for determination of baseline consumption of CHP plants

Abstract

An effective modeling technique is proposed for determining baseline energy consumption in the industry. A CHP plant is considered in the study that was subjected to a retrofit, which consisted of the implementation of some energy-saving measures. This study aims to recreate the post-retrofit energy consumption and production of the system in case it would be operating in its past configuration (before retrofit) i.e., the current consumption and production in the event that no energy-saving measures had been implemented. Two different modeling methodologies are applied to the CHP plant: thermodynamic modeling and artificial neural networks (ANN). Satisfactory results are obtained with both modeling techniques. Acceptable accuracy levels of prediction are detected, confirming good capability of the models for predicting plant behavior and their suitability for baseline energy consumption determining purposes. High level of robustness is observed for ANN against uncertainty affecting measured values of variables used as input in the models. The study demonstrates ANN great potential for assessing baseline consumption in energy-intensive industry. Application of ANN technique would also help to overcome the limited availability of on-shelf thermodynamic software for modeling all specific typologies of existing industrial processes.