An electric energy metering method based on big data self-diagnosis

Abstract

The present application belongs to the field of data processing, and particularly relates to an electric energy metering method based on big data self-diagnosis. The present application discloses an electric energy metering method based on big data self-diagnosis, which comprises the following steps: acquiring multi-dimensional heterogeneous sensing data in real time, training a dynamic metering floating baseline by using a space-time graph convolution network, applying a variational mode decomposition and independent component analysis algorithm to decouple the metering data stream into load characteristics, environmental disturbance and hardware drift components, comparing the hardware drift component with the floating baseline in real time to calculate an error correction coefficient, converting the coefficient into a digital compensation factor to act on a data processing module, realizing online correction of the metering result and synchronously generating a health diagnosis report. By constructing a dynamic baseline model and a characteristic decoupling algorithm, the present application realizes real-time self-detection and digital closed-loop self-recovery of metering accuracy, eliminates external environmental interference, and improves the robustness, management level and operation and maintenance efficiency of the electric energy metering system.

Images