A. Lisounkin and G. Schreck (Germany)
water consumption forecasting, time series analysis, wavelet filter, pattern matching, facility control.
This paper focuses on the use of the data-driven approach for a phenomenological modeling of water consumer behavior in an urban area. Such a modeling is subordinated to the middle-term and long-term facility control tasks and the operator training. The key modeling steps are the wavelet based filtering, clustering, syntactic coding, and approximate code matching. Using the developed tech nique, a specific set of codes for an urban area – signatures – will be built. The signatures represent the object domain knowledge which can be interpreted as typical profiles for the scenario based control and the operator training. This procedure was applied for analysis of water consumer behavior in middle-sized German cities.
Important Links:
Go Back
