A Probabilistic Fault Detection Approach for the Management of Energy and Environment in Buildings Using BiGRU-SA Approach

Over the next two decades, energy use in buildings is expected to increase by more than 40 percent. Already high electricity demand is anticipated to increase. Lessening the burden of the increased demand requires energy efficiency measures for buildings. Energy use in business buildings has increased due to space heating and other equipment, whereas residential building energy use has increased due to home appliances, water, and space heating. New and existing buildings, commercial and residential alike, can all benefit from the adoption of energy management systems to cut costs and carbon emissions. To achieve maximum energy efficiency, several management strategies are required for various building types Energy and environment building is used to analyze the data taken in order to locate the source of the problem. The proposed technique has three stages: preprocessing, feature extraction, and model training, all of which occur after a data has been received. Among the preprocessing techniques used are PCA features. the feature will be employed once the preprocessing is complete. Feature extraction makes use of the mean, standard deviation, and maximum deviation. The package of tools used to measure performance is known as GRU, BiGRU, and BiGRU-SA. The proposed BiGRU-SA strategy performs quite well when compared to the other two established methods.