Improved Kalman Filtering-Based Information Fusion for Crack Monitoring Using Piezoelectric-Fiber Hybrid Sensor Network

Abstract

Multifunctional sensor network has become a research focus in the field of structural health monitoring. To improve the reliability and stability of the diagnosis results, it is necessary to fuse heterogeneous signals under the interference of the external load and damage. In this paper, a piezoelectric-fiber hybrid sensor network is integrated to monitor the crack growth around the hole in the aviation aluminum plate. The effect of the load change on the signals of piezoelectric transducers (PZTs) and optical fiber sensors is analyzed. To improve the damage diagnosis result obtained by ultrasonic guided wave imaging diagnosis based on PZTs and strain damage identification based on distributed optical fiber sensor, a fusion strategy of heterogeneous signals based on a two-stage Kalman filtering algorithm is proposed. In the first stage, the features extracted from two types of sensors are fused at a specific time at the feature level, and then the location of the damage center is predicted. Then, the second fusion is to fuse the predicted damage location results at multiple specific times at the decision level. Crack growth monitoring experiments in hot spots of metallic material under bending moment loading is carried out to verify the feasibility of the proposed fusion method. The experimental results indicate that the fusion damage diagnosis results are more stable, moreover, the accuracy of damage location and quantification is improved than the single signal diagnosis results.