Dynamic strain signal monitoring and calibration with neural network based on hierarchical orthogonal artificial bee colony

Abstract

The dynamic strain monitoring system is an important means to reflect the deformation and evolution of the structure in the field of traffic and transportation visually. The exact measurement of dynamic strain measurement system is inseparable from the traceability system of high-precision strain gauge. In this paper, the resistance strain monitoring system installed on the web surface of the main girder box of Jiujiang bridge is calibrated online by using the parallel method at a similar location. By means of dynamic strain measurement system calibration sample signal feature extraction, put forward online calibration method based on neural network algorithm for dynamic strain signaler, move the noise of passive nonlinear signal under the circumstance of incentives, set up and designed a calibration method of HOABC-NN for dynamic signal analysis of on-line monitoring system. After verification, it is found that the algorithm is superior to the traditional neural network training method, the calibration precision is improved obviously, and can support the dynamic strain of bridge monitoring system online calibration