| As one of the most commonly used connection forms,bolt connection has the characteristics of direct force transmission and convenient assembly,etc.Since the bolt connection is economical and has a broad range of application fields,it is widely used in mechanical engineering,civil engineering,aerospace engineering,et al.However,due to the various adverse factors,including the structural vibration,temperature changes and so on,the bolt will inevitably lose in varying degrees.The looseness of bolt joints may seriously weaken the joint stiffness,reduce the reliability of connectors,and affect the safety of bolt joints and even the whole structures.In this paper,based on the smart material of piezoelectric ceramics,the key technical problems,including the ambient noise interferences,multi-bolt looseness monitoring and the initial slight bolt looseness monitoring are investigated The main contents and conclusions of this paper are as follows:(1)A smart washer(SW)sensor based on piezoelectric material is proposed,and the signal-to-noise ratio of the received signal under environmental noise is improved by combining the time-reversal algorithmWhen the traditional sensor is coupling with the monitored bolted structures,it will bring some adverse effects on the structure,such as weakening the bolt stiffness,increasing the mass of joints,et al.To eliminate the adverse effects,this paper proposes a smart washer sensor to sense the change of the preload.The smart washer is fabricated by embedding the piezoelectric material into the washers,which are the accessory structure of bolt joints.Combined with the time-reversal algorithm,a monitoring method of bolt looseness with noise robustness is proposed.The experiment is carried out by using two smart washers:one emits the signals and another senses the signals.The acoustic mutuality of bolt nodes,the signal-to-noise ratio of the received signal under different excitation frequencies and noise levels are studied.For the data processing,the wavelet energy method is applied to extract the energy in a sensitive frequency band as the bolt loosening index.The experimental results show that the time-reversal algorithm can significantly improve the signal-to-noise ratio of the received signal under the noise and low signal amplitude,and further expand the applicability of the bolt loosening monitoring method based on the amplitude or energy in the noise environment.(2)Based on the series/parallel model of piezoceramic electro-mechanical impedance(EMI),a distributed bolt looseness monitoring method is proposed to effectively reduce the number of sensors and the measuring times.There are a large number of bolts at large-scale structural bolt joints.It is an urgent problem that how to use a small number of sensors to monitor bolt looseness at different positions.Since the time-reversal algorithm is an active sensing approach that needs two sensors to implement,to reduce the number of sensors,based on the EMI method,a distributed bolt looseness monitoring method based on the series/parallel model of electro-mechanical impedance is proposed.Firstly,the monitoring process of a bolt looseness is carried out by using the electro-mechanical impedance of one smart washer.Secondly,in the case of the series/parallel connection,the coupling electro-mechanical impedance(EMI)curves of different piezoceramic sensors can be obtained through a single-channel sweep.Based on the principle of the equivalent circuit,the feasibility of the model is analyzed.The results of theoretical analysis show that the series and parallel circuits of piezoelectric sensors have the opposite effect on the piezoelectric impedance and admittance signal,that is,the series circuit method has the superposition effect on the real part of impedance signal of each piezoelectric sensor and the inhibition effect on the admittance signal.The parallel method has the superposition effect on the admittance signal of each piezoelectric sensor and the inhibition effect on the resistance signal.According to the theoretical results,the impedance signal in the series circuit is selected and applied in the verification experiment of a distributed bolt looseness monitoring approach.For the signal processing of a series/parallel circuit,a method named 3dB bandwidth-based root mean square deviation(3dB-RMSD)is proposed.The experimental results show that the distributed bolt looseness monitoring method based on piezoelectric series/parallel circuit can not only reduce the number of sensors,but also accurately predict the degree and position of bolt looseness.(3)A high-resolution pre-load monitoring method for the initial looseness of bolts is proposed by using the coda wave interference(CWI)The goal of bolt monitoring is to give early warning of bolt pre-load looseness,which is to give a warning when a slight looseness of bolt pre-load occurs.Since the aforementioned methods are not sensitive to the initial looseness of bolts,to overcomes this issue,a high-resolution bolt looseness monitoring method based on coda wave interference is proposed.By using stretching method to calculate the time shift of coda wave under different pre-load forces,the looseness degree of bolts can be measured.First of all,the feasibility of the proposed method is analyzed theoretically by using the theory of waveform summation.Secondly,the high sensitivity characteristic of the coda wave is verified by numerical simulation.Finally,based on relevant experiments,a DRBP index is proposed to quantitatively evaluate the resolution of the coda wave interference method.The experimental results show that the CWI-based bolt looseness monitoring method has a high sensitivity to the change of pre-load,and its ultimate resolution is up to 0.323%.Therefore,the proposed CWI-based bolt looseness monitoring method can effectively monitor the slight changes of bolt pre-load,which is expected to solve the problem of bolt looseness at the initial stage.(4)The sensitivity of direct wave and coda wave to the change of bolt pre-load is studied.A method of bolt pre-load monitoring based on direct wave and coda wave in the long-time service is proposed.The pre-load force of bolt may undergo a process from small perturbation to large loss in the long-term service.According to the characteristics of bolt pre-load looseness in different stages,we proposed a pre-load monitoring method using direct wave and coda waves.Firstly,the sensitivity of direct wave and coda wave to the change of pre-load is studied by changing the loading interval of loading forces.The resolutions of the two methods are presented by calculating the time shifts of the direct wave and the coda wave.The experimental results show that the ultimate value of the DRBP of direct wave is 1.681%.Therefore,the direct wave-based method is suitable for the case where the change of pre-load is greater than 1.681%.The DRBP range of coda wave within 0.323%-1.980%,which means that the method based on coda wave is suitable for the condition that the change degree of pre-load force is between 0.323%-1.980%.To sum up,the CWI-based approach is suitable for the case of a small change of bolt preload,for example,the initial looseness of bolts.While the time shift of direct wave can reflect the large pre-load looseness of bolt pre-load.By integrating the direct wave with the coda wave,the change of the bolt pre-load force in the whole service cycle can be monitored. |
PDF Full Text Request