Predictive Maintenance System for Smart Home Appliances Using IoT Sensor Data and Deep Learning

The need for intelligent maintenance technologies that can guarantee appliance stability and lower accidental damages has increased due to the rising use of Internet of Things (IoT) technologies in smart homes. In this research, a deep learning and Internet of Things sensor-based predictive maintenance system for smart home appliances is presented. An IoT framework continually collects multi-modal sensor data from home appliances, such as temperature, vibration, voltage, and current. For fault analysis, data collected is preprocessed and converted into useful characteristics. To efficiently capture both time and space trends in sensor data, a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN– LSTM) model is suggested. The suggested model's performance is compared to both separate deep learning techniques and conventional machine learning. The CNN–LSTM model dramatically reduces prediction error while achieving improved accuracy, precision, recall, and F1-score, according to experimental data. The suggested method lowers maintenance costs, improves appliance performance, and facilitates informed choices in smart home environments.