Abstract
Mine water inrush can cause property losses and casualties, but current theoretical and technological approaches cannot accurately predict such events. Through the networked deployment of water level sensors along a mine roadway, a mine water inrush monitoring network was developed, and a multi-constraint and multi-objective optimal deployment method was established. By setting practical constraints of the mining area, water inrush risk level, and installation at specified locations, and considering two objective functions of minimum total cost and minimum average monitoring time, a mathematical model was established. The non-dominated sorting genetic algorithm II (NSGA-II) was designed to solve the model. The method temporally and spatially optimized the network, which was then verified in the Beiyangzhuang coal mine in north China. The average response time of the monitoring network was 916 s using only 28 water level sensors. The higher the water inrush risk level, the shorter the monitoring network response time. Under the 2, 3, and 4 risk levels, the network's response time to simulated water inrush accidents was less than 3000, 2100, and 900 s, respectively. The multi-constraint and multi-objective optimization layout method further enhanced the effectiveness of the network, providing a novel system for the early warning of mine water inrush.