ABSTRACT
The spread of the global COVID-19 epidemic has resulted in significant shifts in electricity consumption compared to regular days. It is unknown if standard single-task, single-indicator load forecasting algorithms can accurately reflect COVID-19 load patterns. Power practitioners urgently want a simple, efficient, and accurate solution for anticipating reliable load. In this paper, we first propose a unique collaborative TCN-LSTM-MTL short-term load forecasting model based on mobility data, temporal convolutional networks, and multi-task learning. The addition of the parameter sharing layers and the structure with residual convolution improves the data input diversity of the forecasting model and enables the model to obtain a wider time series receptive field. Then, to demonstrate the usefulness of the mobility optimized TCN-LSTM-MTL, tests were conducted in three levels and twelve base regions using 19 different benchmark models. It is capable of controlling predicting mistakes to within 1 % in the majority of tasks. Finally, to rigorously explain the model, the Shapley additive explanations (SHAP) visual model interpretation technology based on game theory is introduced. It examines the TCN-LSTM-MTL model's internal mechanism at various time periods and establishes the validity of the mobility indicators as well as the asynchronous relationship between indicator significance and real contribution.
ABSTRACT
The pandemic of coronavirus disease 2019 (COVID-19) has posed serious challenges. It is vitally important to further clarify the epidemiological characteristics of the COVID-19 outbreak for future study and prevention and control measures. Epidemiological characteristics and spatial-temporal analysis were performed based on COVID-19 cases from 21 January 2020 to 1 March 2020 in Shandong Province, and close contacts were traced to construct transmission chains. A total of 758 laboratory-confirmed cases were reported in Shandong. The sex ratio was 1.27: 1 (M: F) and the median age was 42 (interquartile range: 32-55). The high-risk clusters were identified in the central, eastern and southern regions of Shandong from 25 January 2020 to 10 February 2020. We rebuilt 54 transmission chains involving 209 cases, of which 52.2% were family clusters, and three widespread infection chains were elaborated, occurring in Jining, Zaozhuang and Liaocheng, respectively. The geographical and temporal disparity may alert public health agencies to implement specific measures in regions with different risk, and should attach importance on how to avoid household and community transmission.