Research Article
Unified Quantile Regression Deep Neural Network with Time-Cognition for Probabilistic Residential Load Forecasting
Figure 9
Heat map of 80% interval AACE for four models. QGBRT, QCNN, QLSTM, and our proposed model in X-axis correspond to AACE for different models, while I_QGBRT, I_QCNN, and I_QLSTM represent the AACE improvement ratio of our proposed model over other three models. Y-axis expresses different 20 residences. AACEs of the same residence are listed in the same row, where darker green shades imply a better performance.