Fingerprint-Based Deduplication for Renewable Energy Data On-Site Analyzing

Electric power, as a primary source of energy, is crucial to ensuring the healthy development of socio-economic systems. With the rapid growth of the economy, the energy demand continues to rise, necessitating the reduction of environmental pollution. Consequently, renewable energy sources such as wind, photovoltaic, and solar-thermal have been widely adopted. The advancement of microgrid autonomy and smart IoT technologies has improved energy utilization efficiency, optimized the energy consumption structure, reduced transmission losses, and alleviated environmental impacts through the use of renewable sources. Currently, smart power IoT focuses on the real-time collection and distribution of data, facing significant challenges due to the imperfections in edge IoT technologies and reliance on centralized cloud server processing, which substantially increases the computational load on central servers. We propose an edge IoT architecture that enhances the consumption capacity of distributed renewable energy at the microgrid level. By studying fine-grained data fingerprinting and neural network-based renewable energy data edge-analysis technologies, we have enabled edge processing of data, reducing latency and privacy issues, and enhancing the system’s real-time processing capabilities and responsiveness.