Research Of Truss Structural Health Monitoring Based On Electromechanical Impedance Method

As one of common civil structures, truss structure is widely used in highwaytraffic construction and roof structure of buildings for its advantages of easytransportation, convenient assembly and uniform loading. However, truss structure istemporary structures, and has the characters of multiple building styles, moreuncertain factors, bearing a large load, computational theory is incomplete and so on.So that nearly a quarter of construction accidents in China was related to the failure oftruss structure, and was caused a great number of economic losses. Because of thesereasons, how to monitor the security of truss structures timely and accurately is notonly related to the safety of workers, but also construction process.Electromechanical impedance (EMI) method is a new kind of method used tomonitor the health of structure. The basic idea of EMI method is to determine thehealth state of structure by measuring impedance spectra of PZT sensors. The sensorsare bonded on different positions of the structure to monitor the health state of thisstructure. EMI technique also has the advantages of nonparametric model analysis andhigh local sensitivity. However, the impedance measuring instruments in the markettoday are bulky, expensive and not cconvenient to use in project sites, therefore asmall portable wireless impedance measurement system is developed. Finally, neuralnetwork technique is adopted to process measured data, and the health state ofstructure is recognized and analyzed intelligently.Main contents and results of this dissertation are as follows:(1) Domestic and international research profile about electromechanicalimpedance method applied in structural health monitoring and impedancemeasurement is analyzed, and the basic theory on electromechanical impedancemethod applied in structural health monitoring and neural network based damageintelligent identification method is researched in detail.(2) The hardware circuit of impedance measurement system is developed basedon high precision impedance measurement chip AD5933, the circuit including theAD5933peripheral circuit, microprocessor chip and wireless communication module.Upper computer is developed based on MATLAB and visual C++. Then developedsystem is used to monitor the health of pipeline, and effectiveness of this system canbe verified. (3) Then, EMI method is applied to monitor the health state of truss structure.First, the frequency band of spectrum, which is the most sensitive to the variation ofthe truss structure, is determined experimentally. Then, impedance spectra data aremeasured and analyzed when the truss structure is under different health state.Furthermore, these processes include pre-processing, such as mathematical statisticsmethod, data compression and et al and neural network based damage identification.And finally, this developed system was used to monitor the health state of trussstructure. Experimental results show that, EMI method has far reaching applicationpotential to monitor the health state of complex structure. By using neural networktechnique to process measured data, damage can be confirmed not only the existencebut also its severity and location. Finally, developed system was used to carry out thestudy of structure health monitoring, and proved this system is practical.