A Novel Method for Source Tracking of Chemical Gas Leakage: Outlier Mutation Optimization Algorithm.

Area of Science:

• Environmental Science
• Chemical Engineering
• Computer Science

Background:

• Chemical industrial parks are critical infrastructure requiring robust emergency response to gas leakage accidents.
• Gas leakages pose significant risks including poisoning, fire, and explosion, necessitating effective source tracking.
• Existing source tracking technologies require enhancement for speed and accuracy in emergency situations.

Purpose of the Study:

• To propose a novel Outlier Mutation Optimization (OMO) algorithm for rapid and accurate chemical gas leakage source tracking.
• To evaluate the performance of the OMO algorithm against other optimization techniques.
• To demonstrate the flexibility and robustness of the OMO algorithm in realistic scenarios.

Main Methods:

• Development of the Outlier Mutation Optimization (OMO) algorithm, incorporating random walk exploration and increased individual mutation probability.
• Utilizing Swarm Intelligence (SI) principles within the OMO algorithm.
• Simulation of chemical gas leakage scenarios using the Gaussian plume model with random parameters for algorithm testing.

Main Results:

• The OMO algorithm demonstrated superior comprehensive performance with extremely high calculation accuracy across different sensor numbers (e.g., optimal value of 2.464e-15 with 16 sensors).
• Satisfactory calculation times were achieved, indicating efficiency for emergency response (e.g., 12.743 s/50 times with 16 sensors).
• The OMO algorithm outperformed direct search, evolutionary, and other swarm intelligence algorithms in leakage source tracing accuracy.

Conclusions:

• The OMO algorithm is a flexible and robust method for chemical gas leakage source tracking.
• OMO offers significant advantages in terms of accuracy and speed compared to existing algorithms.
• This novel algorithm enhances emergency response capabilities for chemical industrial parks.