An Adaptive Cyberinfrastructure for Threat Management in Urban Water Distribution Systems

Abstract

Abstract. Threat management in drinking water distribution systems involves real-time characterization of any contaminant source and plume, design of control strategies, and design of incremental data sampling schedules. This requires dynamic integration of time-varying measurements along with analytical modules that include simulation models, adaptive sampling procedures, and optimization methods. These modules are compute-intensive, requiring multi-level parallel processing via computer clusters. Since real-time responses are critical, the computational needs must also be adaptively matched with available resources. This requires a software system to facilitate this integration via a highperformance computing architecture such that the measurement system, the analytical modules and the computing resources can mutually adapt and steer each other. This paper describes the development of such an adaptive cyberinfrastructure system facilitated by a dynamic workflow design.

A dynamic simulation–optimization model for adaptive management of urban water distribution system contamination threats

“…The adaptive simulation-optimization scheme proposed here should be ideally coupled with a dynamic source identification model [20] that continuously updates the perceived contamination scenario using streaming data. A general framework for dynamic integration of these analytical modules for development of an adaptive cyberinfrastructure for threat management is described in [31]. 6 In the event that a contaminant is introduced to a WDS, water utility operators may take different preventive and protective actions to protect public health.…”

Using Niched Co-Evolution Strategies to Address Non-Uniqueness in Characterizing Sources of Contamination in a Water Distribution System

“…The parameters shown in Table 1 and the computation time of approximately 83 days on an average desktop computer reflect the need to have a large number of individuals generated to search Mesopolis. To reduce the computation time, NCES was executed on a computer cluster containing eleven nodes with two 2.2 GHz processors, four GB RAM, and 80 GB HD per node (Mahinthakumar et al 2006). Using parallelized versions of the Java framework and EPANET, the computation time for one optimization trial using the settings in Table 1 was reduced to approximately seven hours.…”

Introduction to the ICCS2006 Workshop on Dynamic Data Driven Applications Systems

“…In projects [9][10][11][12], the applications are in the areas of environmental and natural resource management. In [9] the project employs previous work performed by the investigators in the Instrumented Oil-Field DDDAS project that has enabled a new generation of data-driven, interactive and dynamically adaptive strategies for subsurface characterization and oil reservoir management.…”

“…Work conducted in [10] develops capabilities to convert a traditional data collection sensor and ocean model into a DDDAS enabled system for identifying contaminants, dynamically coupling different models, simulations, and sensing strategies, in a symbiotic manner. The paper in [11] is centered on methods for detection, management and mitigation of threat in drinking water distribution systems; it involves real-time characterization of any contaminant source and plume, design of control strategies, and design of incremental data sampling schedules. This requires dynamic integration of time-varying measurements along with analytical modules hat include simulation models, adaptive sampling procedures, and optimization methods.…”