Genetic algorithm-determined artificial neural network architecture for predicting power usage effectiveness (PUE) in a data center

The accurate estimation of a data center's power use effectiveness (PUE) is critical for refinery operations. The predictions of two machine learning models are compared in this research: genetic algorithms combined with artificial neural networks are both artificial neural networks. Using a new method for genetically improving artificial neural networks (ANN), PUE has been predicted (GA). The number of neurons in the hidden layer is determined by the genetic algorithm. The artificial neural network model has 18 variables as inputs. The best structure and training parameters for an ANN have been shown to be determined by the genetic algorithm. Furthermore, an artificial neural network model powered by a genetic algorithm was assessed, and the findings suggested that the PUE may be predicted with some accuracy. This method can help to increase forecast accuracy.