基于Q学习超启发算法的FBG混叠光谱解调方法OA

To address the reflection spectrum aliasing problem in fiber Bragg grating(FBG)multiplexing systems,a Q-learning hyper-heuristic differential evolution(QHDE)algorithm is proposed in this paper.Based on the differential evolution(DE)framework,the algorithm constructs an operator library containing four heterogeneous mutation strategies,and introduces the Q-learning mechanism to achieve the adaptive selection of mutation strategies.By evaluating the performance of the strategy in real time,computational resource scheduling is dynamically optimized.The demodulation results show that the QHDE algo-rithm achieves picometer-level accuracy for wavelength demodulation of FBG aliasing spectrum in the test range,and its de-modulation performance is significantly better than the DE algorithm with single mutation strategy,exhibiting excellent robust-ness and demodulation capability.