Multi-objective parrot optimizer based on fine stagnation detection and elite guidance for solving helicopter rotor blade sensor layout problem

In helicopter rotor-blade structural health monitoring (SHM) systems, sensor-placement optimization (SPO) needs to reconcile two conflicting goals: accurate modal-response acquisition and lightweight sensor deployment. This paper proposes a multi-objective parrot optimizer with elite guidance and fine-grained stagnation detection (MOPO_ES) to solve the multi-objective SPO problem for rotor blades. A bi-objective optimization model is established, where one of seven critical SHM performance metrics and the total number of sensors are taken as optimization objectives. The elite-guided greedy selection (EGGS) strategy draws high-quality elite individuals from the external non-dominated-solution archive. It performs dimension-wise greedy iterations to refine solutions locally and greatly boost convergence efficiency. The presented fine-grained stagnation-detection mechanism employs a sliding window and improved threshold-based sub-objective tracking. It identifies stagnation for each objective independently and dynamically tunes the Gaussian-noise amplitude, global resampling probability and exploration-enhancement factor to balance global exploration and local exploitation precisely. Comparative simulations are conducted against mainstream multi-objective optimizers, including NSGA-III, NSGA-II, MOPSO, MOEA/D and SPEA2, for performance verification. The experimental results show that the proposed algorithm yields well-distributed Pareto-optimal solutions with competitive convergence capability.