Cloud-Based Data Storage and Big Data Analytics for Fault Prediction in Intelligent High-Voltage Power Supplies

The convergence of cloud computing, IoT, and predictive analytics has reshaped the operation of intelligent high-voltage power supplies. By integrating sensors and communication interfaces, these systems continuously collect data on voltage, current, temperature, insulation resistance, and operational cycles.
Edge processing units perform local pre-filtering and compression before transmitting the data to cloud platforms for long-term storage and analysis. Using machine learning algorithms such as random forests, LSTM neural networks, and anomaly detection models, the system identifies degradation trends and predicts component failures before they occur.
Health indices derived from feature parameters enable predictive maintenance strategies, replacing reactive service models with proactive system management. Remote dashboards visualize power supply health, performance metrics, and fault probabilities in real time.
This data-driven framework allows centralized control, remote parameter adjustment, and intelligent fault prevention, ensuring higher reliability, lower maintenance cost, and safer long-term operation of modern intelligent high-voltage power systems.