ISO 50001 Data Driven Methods for Energy Efficiency Analysis of Thermal Power Plants

This paper proposes an energy management system based on Artificial Neural Network (ANN) to be integrated with the standard ISO 50001 and aims describe definition enhancement of baselines by means artificial intelligence techniques applied tested real electrical absorption data auxiliary units different thermal power plants in Italy. Power plant optimized operations are important both for cost performance reasons related effects environment next future transition scenario. The improvement consists determining more accurate consumption monitoring models that able track inefficiencies drifts through analytics Intelligence. Starting from analysis vectors at production site level, we performed a multi-scale define macro areas level finally find most relevant within plants. A comparison ANNs several was model complex architecture optimize savings respect pre-set thresholds according procedure. determined available plants’ Distributed Control System (DCS), can identify derived unit’s proper operation. Based reported numerical simulations, improved have been reached up 5% threshold sub-units, thus representing overall saving terms alert control efficiency: potential about 140 MWh throughout considered three-year dataset obtained taking into account cooling tower sub-unit, considerable economic benefit. results highlight neural technique efficiency defining represents valuable tool large asset face observed last years