Data-driven optimal sensor placement for high-dimensional system using annealing machine

We propose a novel method for solving optimal sensor placement problem high-dimensional system using an annealing machine. The points are calculated as maximum clique of the graph, edge weight which is determined by proper orthogonal decomposition mode obtained from data based on fact that usually has low-dimensional representation. Since equivalent to independent set complement solved Fujitsu Digital Annealer. In contrast existing greedy methods, select point at each step and never reconsider selected previously, proposed superior because it able find points. As demonstration high dimensional system, pressure distribution measured pressure-sensitive paint method, optical flow diagnose reconstructed root mean square errors (RMSEs) between pressures transducers random selection compared. similar RMSE achieved approximately 1/5 number method. This great importance approach new engineering application