Multi-sensor optimal deployment based efficient and synchronous data acquisition in large three-dimensional physical similarity simulation

Purpose This paper aims to propose a deployment optimization and efficient synchronous acquisition method for compressive stress sensors used by stress distribution law research based on the genetic algorithm and numerical simulations. The authors established a new method of collecting the mining compressive stress-strain distribution data to address the problem of the number of sensors and to optimize the sensor locations in physical similarity simulations to improve the efficiency and accuracy of data collection. Design/methodology/approach First, numerical simulations were used to obtain the compressive stress distribution curve under specific mining conditions. Second, by comparing the mean square error between a fitted curve and simulation data for different numbers of sensors, a genetic algorithm was used to optimize the three-dimensional (3D) spatial deployment of sensors. Third, the authors designed an efficient synchronous acquisition module to allow distributed sensors to achieve synchronous and efficient acquisition of hundreds of data points through a built-in on-board database and a synchronous sampling communication structure. Findings The sensor deployment scheme was established through the genetic algorithm, A synchronous and selective data acquisition method was established for reduced the amount of sensor data required under synchronous acquisition and improved the system acquisition efficiency. The authors obtained a 3D compressive stress distribution when the advancement was 200 m on a large-scale 3D physical similarity simulation platform. Originality/value The proposed method provides a new optimization method for sensor deployment in physical similarity simulations, which improves the efficiency and accuracy of system data acquisition, providing accurate acquisition data for experimental data analysis.