Location of natural gas leakage sources on offshore platform by a multi-robot system using particle swarm optimization algorithm

Abstract

The offshore platform serves as the infrastructure for offshore mining. In order to guarantee safe operation of the platform and the development of the platform in the direction of intelligence, a natural gas leakage intelligent detection system with fast response is necessary to prevent potential accident when leakage occurs. In this work, a method to locate leaked natural gas is proposed based on the particle swarm optimization algorithm for the multi-robots to achieve. First, the advection diffusion equation is calculated to simulate the leakage gas transmission using finite difference method in different scene layouts. At the same time，a multi-robot collaborative detection strategy for gas detection, gas tracking, and gas source localization is proposed using an improved particle swarm optimization algorithm. After that, the process of multi robots search for leak sources was simulated. In addition, the control algorithm is analyzed, and the results show that the strategy has a high detection rate. Further analysis shows that the number of robots has increased from 2 to 7, the time it takes for the multi-robot system to successfully locate the leak source decreased; when the number of robots is constant, the time it takes for the multi-robot to successfully locate two leak sources is longer than the time to locate a single leak. Finally, search strategies based on particle swarm optimization, ant colony algorithm and cuckoo search algorithm are used to simulate in a single leak source environment. When the number of robots is changed from 3 to 7, compared with the other two algorithms, the accuracy of the search strategy search based on particle swarm is increased by 2%–28%, and the search time is accelerated by 27s–92s.the particle swarm optimization is superior than two others for control of multi-robot system in leakage source location.