Optimizing\nPoint-in-Space Continuous Monitoring System\nSensor Placement on Oil and Gas Sites

We propose a generic, modular framework to optimize the\nplacement\nof point-in-space continuous monitoring system sensors on oil and\ngas sites aiming to maximize the methane emission detection efficiency.\nOur proposed framework substantially expands the problem scale compared\nto previous related studies and can be adapted for different objectives\nin sensor placement. This optimization framework is comprised of five\nsteps: (1) simulate emission scenarios using site-specific wind and\nemission information; (2) set possible sensor locations under consideration\nof the site layout and any site-specific constraints; (3) simulate\nmethane concentrations for each pair of emission scenario and possible\nsensor location; (4) determine emissions detection based on the site-specific\nsimulated concentrations; and (5) select the best subset of sensor\nlocations, under a given number of sensors to place, using genetic\nalgorithms combined with Pareto optimization. We demonstrate the practicality\nand effectiveness of our framework through its application to an oil\nand gas emission testing facility with a large search space of possible\nsensor locations; a setting which is computationally infeasible to\nsolve with commonly used mixed-integer linear programming. Additionally,\na case study illustrates the successful application of our algorithm\nto an operating oil and gas site, showcasing its real-world applicability\nand effectiveness.