Unified Quantile Regression Deep Neural Network with Time-Cognition for Probabilistic Residential Load Forecasting

<div>Forecasting results and ground truth of residence 2 for one week. <i>X</i>-axis expresses time with a half-hour unit, and <i>Y</i>-axis represents electricity load. Eight intervals between 0.1 and 0.9 are shown in different colors, where the solid black lines represent the ground truth. (a) QGBRT, (b) QCNN, (c) QLSTM, and (d) our proposed model.</div>

Complexity

fig12

Figure 12

Figure 12: Unified Quantile Regression Deep Neural Network with Time-Cognition for Probabilistic Residential Load Forecasting 