Parallel Bionic Algorithm Used to Optimize Sensor Placement for Self-diagnostic Smart Structures

Smart structures are a multi-sensor architecture, and the problem of multi-sensor optimal placement involves the cost-effective factor. It has been of great practical value and is worth to researching further to optimize sensors´ locations and number for smart structures. For piezoelectric smart composite laminated plates, least square wavelet support vector machine (LS-WSVM) with wavelet kernel function is proposed to establish the performance function of damage detection, and then an improved genetic algorithm (GA) is applied to optimize the performance function. To enhance the algorithm speed, LS-WSVM adopted as parallel mode are combined with GA, that is, a parallel genetic algorithm integrated neural network is constructed to realize the optimal sensor placement corresponding to its primal sensor placement based on damage detection. Simulation results show that the optimal placements of different number of sensors are consistent with engineering judgments, and the optimal sensors´ number corresponding to its primal sensorspsila number can be determined through the method of parallel genetic algorithm integrated neural network, and the method possesses the advantages of faster speed, etc..