1 February 2011 in Journal of Physics: Conference Series
10 October 2024 in IEEE Transactions on Instrumentation and Measurement
User-side carbon emission accounting is a means to sort out the amount and source of user carbon emissions. Real-time carbon estimation methods based on non-intrusive load monitoring (NILM) tend to miss the fact that there is not a negative correlation between equipment identification accuracy and carbon emission estimation error. Higher accuracy does not mean lower error. A two-stage training NILM network, LRMSE-ResNeSt, is proposed to reduce the error of real-time carbon emission monitoring results. It emphasizes the low Root-Mean-Square Error (RMSE) of the carbon emissions monitoring. The feature extraction network is first constructed using convolution and split attention mechanism, and then, the model is trained using a two-stage training approach. In the first stage, backpropagation is used to update the network parameters for accurate device identification. In the second stage, the parameters of the fully connected layer are tuned using Particle Swarm Optimization to make the classifier more focused on the identification accuracy of devices with high carbon emissions. Finally, the proposed LRMSE-ResNeSt is validated using the IAID industrial dataset. The experimental results show that the LRMSE-ResNeSt successfully reduces the RMSE of real-time carbon estimation by an average of 14.94%, which addresses the problem that the NILM method only focuses on the accuracy of the equipment identification but ignores the error of the carbon emission estimation results.
8 July 2025 in Journal of Cleaner Production
29 July 2025 in IEEE Sensors Journal
Carbon monitoring is critical in reducing carbon emissions, especially as the increasing electricity consumption in industrial processes offers a unique opportunity to monitor emissions from the electrical perspective, enabling carbon emission monitoring through electrical load monitoring. However, current methods have proven ineffective in providing high-precision monitoring due to the neglect of the individual equipment’s and their different states differing contributions to carbon emissions. Therefore, we propose a carbon emission monitoring method based on multi-state load identification in Advanced Metering Infrastructure, which incorporates both the gram component and the residual component. The gram component converts load series into images to acquire more features, while the residual component identifies the multi-state devices and monitors carbon emission, accurately. Additionally, a novel loss based on focal loss and Huber loss is designed to minimize carbon emission monitoring errors.This study presents a model that excels in both load identification and carbon emission monitoring. Experiments are conducted using industrial users from the IAID dataset, with performance evaluated using metrics such as ACC, F1macro, F1micro and carbon monitoring error (AEEcar). The proposed method outperforms existing approaches, demonstrating superior accuracy in emission monitoring, with the AEEcar below 5%.