| As the structural parts connected by bolts become larger and larger,the loss of production after the bolt breaks is also increasing.Therefore,higher requirements must be ensured for bolted connections.The traditional manual inspection method is gradually replaced by the current online monitoring of bolt pre-tightening force,but there are still certain shortcomings: the degree of bolt looseness cannot be accurately predicted;the amount of invalid data transmission is large.Aiming at the above problems in the current bolt pre-tightening force monitoring system,this paper designs a set of real-time bolt pre-tightening force monitoring system based on the Internet of Things technology.Through the self-designed intelligent gateway,the data of the bolt pre-tightening force monitoring equipment is collected wirelessly,and the data is reported to the cloud platform for analysis and processing through 4G.At the same time,the equipment status monitoring,remote control,failure warning and some visual management functions can be realized by designing the Web client.In order to accurately predict the bolt loosening time in advance,a fault prediction scheme is designed based on the BP neural network algorithm.Aiming at the problem that the BP neural network algorithm is easy to fall into the local optimal solution,a particle swarm algorithm is proposed to iteratively optimize its weights and thresholds.Aiming at the shortcomings of the standard particle swarm algorithm,two strategies are designed to optimize by designing random inertia weights and dynamically adjusting learning factors.Finally,the test shows that the root-mean-square error of the improved PSO-BP neural network algorithm prediction accuracy is reduced by 47.9% compared with that before the improvement,indicating that the improved algorithm has higher stability and prediction accuracy.In order to reduce the problem of invalid data transmission in the remote operation and maintenance system,the layered data processing mechanism is designed to optimize data transmission links in this paper.Starting from the device layer,platform layer,and application layer,design corresponding working mechanisms to optimize data transmission according to the characteristics of data interaction between each layer.Strategies such as repeated data filtering,batch data reporting,symmetrical monitoring points,and resumable transmissions are designed to reduce the amount of invalid data transmission.The test result shows that compared with the transparent transmission method,the data transmission volume is reduced by about 61% through layered data processing mechanism. |
