Real-time location algorithms of drinking water pollution sources based on domain knowledge

Area of Science:

• Environmental Science
• Water Resource Management
• Sensor Networks

Background:

• Real-time pollution source localization is crucial for water quality management.
• Water supply networks present challenges due to vast scale and limited sensor placement.
• Multiple solutions arise from sparse sensor data, hindering accurate source identification.

Purpose of the Study:

• To develop an effective algorithm for the real-time location of pollution sources in water supply networks.
• To address the challenge of multiple solutions caused by limited sensor placement.
• To improve the accuracy and reliability of pollution source identification.

Main Methods:

• A dynamic optimization model based on time-varying pollution concentration data from water quality sensors.
• A multi-strategy dynamic multi-mode optimization algorithm incorporating domain knowledge.
• A merging mechanism to maintain population diversity and prevent premature convergence.

Main Results:

• The proposed algorithm effectively solves the real-time pollution source location problem across various pipe networks.
• Simulation results demonstrate the algorithm's capability in different pollution scenarios.
• Increased monitoring data enhances the uniqueness and accuracy of the identified pollution source location.

Conclusions:

• The developed algorithm provides an effective solution for real-time pollution source localization in water systems.
• The integration of domain knowledge and advanced optimization techniques improves localization accuracy.
• This approach offers a robust method for managing and mitigating water pollution incidents.