Research on the Construction Method of the Service-Oriented Web-SWMM System

Dawei Xiao, Min Chen, Yuchen Lu, Songshan Yue, Tao HouISPRS Int. J. Geo Inf 2019 · DOI

TL;DR

This paper introduces Web-SWMM, a novel service-oriented system that makes the powerful SWMM hydrological model accessible via a web browser. By encapsulating SWMM as a web service, it overcomes the limitations of traditional desktop installations, enabling easier data preparation, online computation, and visualization for urban flood simulation.

What Problem Does This Paper Address?

Urban flooding is a growing concern globally due to rapid urbanization. While the US EPA's Storm Water Management Model (SWMM) is a leading tool for simulating these events, its desktop-based nature presents significant hurdles. Users must download and install the software, which can be complex and time-consuming. This local dependency also hinders collaboration, data sharing, and the integration of real-time data for dynamic simulations. The authors aim to bridge this gap by creating a web-based platform that democratizes access to SWMM, making it more user-friendly and facilitating its broader application in urban flood management.

Methods

The core of the Web-SWMM system lies in its service-oriented architecture, built upon the OpenGMS platform. The process involves:

  • Model Encapsulation: SWMM is wrapped using a standardized model description language (MDL) that defines its input, output, and execution behavior. This creates a "model service wrapper."
  • Model Deployment: The encapsulated SWMM model is deployed onto a server and published as a web service. This involves matching the model's software and hardware requirements with available service wrappers.
  • Data Preparation (UDX Model): A key innovation is the use of the Universal Data eXchange (UDX) model. UDX provides a flexible, structured way to represent SWMM's complex input and output data. This enables:
  • Online Data Configuration: A tool allows users to create or configure SWMM input data directly in UDX format via a web interface.
  • Data Editing: Users can visually edit geographic objects and their attributes within the UDX data.
  • Conversion: Algorithms convert UDX data to SWMM's native .inp format for computation and SWMM's .rpt/.out files back to UDX for visualization.
  • Web Service System: The system comprises a data module (for UDX configuration, editing, and visualization) and a model module (for computing node selection, model migration, and control).
  • User Interface: A web-based interface, built with Java, JavaScript, and HTML5, allows users to interact with the system, upload data, run simulations, and visualize results.

Datasets

The paper describes a prototype implementation and experimental study using data from the "Suzhou Meisong Garden Community." This dataset includes:

  • Basic geographic data of the community.
  • Rainwater well information (numbered 1-108).
  • Topography factors and building information.
  • The experimental area was divided into 84 subcatchments.

This data was processed and organized into the UDX format required by the Web-SWMM system. Rainfall data was simulated using the Chicago rain type generator for a 2-year return period and 2-hour duration.

Key Insights and Ideas

  • Democratizing SWMM: The most significant insight is that by abstracting SWMM's complexity behind a web service, it becomes accessible to a much wider audience, including the general public and researchers without specialized software engineering skills.
  • UDX as a Universal Translator: The UDX model proves crucial for bridging the gap between SWMM's specific data formats and the need for flexible, web-friendly data representation and exchange. It simplifies data preparation and visualization.
  • Service-Oriented Architecture for Hydrological Models: The paper demonstrates a robust framework for encapsulating and deploying complex hydrological models as web services, paving the way for a more integrated and collaborative approach to water resource management.
  • Online Data Management: The integrated tools for online data configuration, editing, and visualization streamline the entire simulation workflow, reducing the time and effort required for model setup.

Key Results

The paper successfully implements and tests the Web-SWMM system. The experimental study on the Suzhou Meisong Garden Community demonstrated that:

  • Users can configure SWMM input data online using UDX.
  • The system can process simulated heavy rainfall data.
  • The system can visualize output data, such as conduit maximum depth and node maximum depth.
  • The simulation results indicated potential overflow issues for specific rainwater wells and exceeding the maximum limit for an underground conduit under the tested heavy rain scenario.
  • The system successfully avoids the need for users to download or install SWMM locally.

Limitations and Future Work

The authors acknowledge several limitations and suggest future research directions:

  • Manual Programming: The current implementation relies on manual programming for SWMM encapsulation. Future work should focus on developing an automated tool to reduce labor costs and technical thresholds.
  • Data Preprocessing: The data preprocessing function could be enhanced with more sophisticated tools, such as a real-time rainfall data to UDX converter, to further simplify data preparation and enable real-time simulations.
  • Model Extension: The web service mode should be extended to include other hydrological models, creating a comprehensive hydrological model service theme to promote broader sharing and reuse.

Summary updated

BibTeX
@article{xiao2019research,
  author = {Dawei Xiao and Min Chen and Yuchen Lu and Songshan Yue and Tao Hou},
  title = {{Research on the Construction Method of the Service-Oriented Web-SWMM System}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2019},
  doi = {10.3390/ijgi8060268},
}
Figure