Research And Implement Of Monitoring System For Special Experiment Mechanism Based On Digital Twin Technology

Special wind tunnel test mechanism(referred to as test mechanism)is an important device for complex aerodynamic test of aircraft.Under narrow space constraints and complex aerodynamic loads,the test mechanism is prone to collision and fracture accidents.It is urgent to develop advanced monitoring system to realize collision prediction and mechanical response monitoring under complex aerodynamic loads.In order to solve this problem,this paper studies the monitoring system of test institutions based on the idea of digital twinning.The main research contents are as follows.1)The actual demand of the monitoring system of the test mechanism is systematically analyzed,and a monitoring method of the test mechanism based on digital twinning technology is proposed.Based on the study of design objectives and design principles,the general line of system research is formulated,and Unity is selected as the digital twin basic platform to provide visualization support in monitoring research.2)In terms of collision prediction,aiming at the problem of low detection accuracy and failure to realize collision prediction in traditional bounding box detection,a doublelayer collision prediction algorithm for twin objects of test mechanism based on current frame and prediction frame is proposed.The algorithm includes the first-layer bounding box collision prediction and the second-layer risk triangle one-sided collision prediction,and detects the current frame collision while predicting the next frame collision,so as to realize the review of the prediction results.According to the environmental constraint characteristics of the test mechanism,a HOLE enclosure was designed to solve the problem that the common enclosure box does not apply to the hole structure.According to the velocity direction of the dynamic object,the risk triangle one-sidedness is selected to reduce the calculation amount of the second layer primitive detection of the algorithm.Experiments show that the proposed collision prediction algorithm can effectively improve the efficiency of graph element detection and effectively predict the occurrence of collision.3)In the mechanical response monitoring of the mechanism,in view of the complex wiring of the bridge method in the original stress monitoring,it is impossible to obtain the overall deformation and stress distribution of the parts,a real-time monitoring method of the deformation and stress of the test mechanism based on the simulation of the deformation and stress regression model is proposed.According to the stress analysis of the whole mechanism,the weak links of the mechanism are obtained,and the data of surface feature points are obtained based on APDL program to form the neural network data set.Then,the GSO-BP neural network is used to fit the finite element simulation data to form the mechanical response regression model.The test results show that the stress deformation and stress distribution of the monitoring platform strut are close to the simulation results of ANSYS finite element software.The maximum deformation error is only 0.125 mm,and the stress distribution is basically the same.Compared with ANSYS,the calculation time is significantly reduced,which can achieve the mechanical property monitoring requirements.4)The integration and implementation of the monitoring system for special test institutions are completed.Based on Unity digital twinning platform,the mechanism collision prediction and mechanical response monitoring module are integrated and the monitoring system interface is realized.The integrated system can effectively realize collision prediction and mechanical response monitoring of mechanism under complex aerodynamic loads,and improve the monitoring level of mechanism.