Diffusion-based Time Series Forecasting for Sewerage Systems
This addresses forecasting challenges for sewerage system management, particularly during severe weather, but appears incremental as it applies existing diffusion and conformal methods to this domain.
The paper tackles the problem of forecasting in sewerage systems by developing a diffusion-based model that processes multivariate time series data, achieving robust predictions during extreme weather events with statistically reliable prediction intervals through conformal inference.
We introduce a novel deep learning approach that harnesses the power of generative artificial intelligence to enhance the accuracy of contextual forecasting in sewerage systems. By developing a diffusion-based model that processes multivariate time series data, our system excels at capturing complex correlations across diverse environmental signals, enabling robust predictions even during extreme weather events. To strengthen the model's reliability, we further calibrate its predictions with a conformal inference technique, tailored for probabilistic time series data, ensuring that the resulting prediction intervals are statistically reliable and cover the true target values with a desired confidence level. Our empirical tests on real sewerage system data confirm the model's exceptional capability to deliver reliable contextual predictions, maintaining accuracy even under severe weather conditions.