An IoT-Based Prediction Technique for Efficient Energy Consumption in Buildings

Today, there is a crucial need for precise monitoring and prediction of energy consumption at the building level using latest technologies including Internet Things (IoT) data analytics to determine enhance usage. Data-driven models could be used prediction. However, due high non-linearity between inputs outputs models, these improvement in terms accuracy robustness. Therefore, this work aims predict usage optimum outline building-extensive distribution strategies based on lightweight IoT framework. To calculate accurate consumption, an enhanced hybrid model was developed Auto-Regressive Integrated Moving Average (ARIMA) Imperialist Competitive Algorithm (ICA). The parameters ARIMA were optimized by adapting ICA technique that improved fitting while preventing over-fitting acquired data. Then, Exponentially Weighted (EWMA) applied monitor predicted values. proposed AIK-EWMA assessed actual power validated mathematical tests. As compared previous works, findings revealed accurately green automation applications.