Regularized DNN Based Adaptive Compensation Algorithm for Gateway Power Meter in Ultra-High Voltage Substations

The development of the power Internet of Things (IoT) has made the data of ultra-high voltage substation metering critical information sources for supporting power energy scheduling and market transactions. However, as key data acquisition devices, the operation performance of ultra-high voltage substation energy meters exhibits instability under different environmental conditions, and even small metering errors can lead to significant discrepancies in settlement. This issue directly impacts the overall safety and stability of the power system. Therefore, this paper considers the influence of environmental factors such as temperature, humidity, and air quality on substation energy metering and proposes an adaptive compensation algorithm for ultra-high voltage substation energy meter based on regularized deep neural networks (DNN) to enhance metering accuracy. Finally, the effectiveness of this algorithm is validated through simulation experiments.