Energy monitoring of process systems: Time-series segmentation-based targeting models

摘要: Energy monitoring systems calculate actual energy use, estimate needs at normal operation, track metrics, and highlight issues related to efficiency of process plants. Analysis key indicators (KEIs) allows the comparison different operating regimes. Based on extracted knowledge realistic targets KEIs can be determined. The performance data-driven targeting models depends how effective regimes are characterized. Till now this modeling task is performed manually based heuristic subjective evaluation operation. A goal-oriented time-series segmentation technique has been developed automate selection proper data used for identification models. With proposed novel algorithm targeting-models regions automatically concept resulted system demonstrated Heavy Naphtha Hydrotreater CCR Reforming Units MOL Hungarian Oil Gas Company.