Research On Detection Technology Of Auto Brake Disc Automatic Balancing Machine

Rotor dynamic unbalance is one of the main causes of rotating machinery failure.The dynamic balance precision of the automobile brake disc has an important impact on the braking safety and performance of the vehicle.In order to ensure the dynamic balance precision of the brake disc to meet the requirements,it is necessary to use the balancing machine to detect the unbalance of the brake disc.Therefore,the key technology research on the detection station of the balancing machine is the key to the accurate detection of the brake disc.In view of the main technical problems of balancing machine detection station,respectively from the structure design and analysis of balancing machine detection station components,balancing machine detection station servo system controller design and balance machine detection system of digital signal processing method three aspects.First of all,according to the working requirements of the balancing machine,the need to detect the structure of the components to achieve the unbalance signal torsional component and translational component separation,and signal acquisition should avoid resonance interference by mechanical structure.In this paper,the key structure of the balancing machine detection system is designed,and the mechanical characteristics of the pendulum structure are analyzed,and the feasibility of the static couple separation of the pendulum vibration system is verified theoretically.The finite element modal analysis of the key components of the vibration system of the pendulum frame is carried out,and it is proved that the structure is not easy to resonance under external excitation,and meets the working requirements.At the same time,the static analysis of the box structure is carried out to ensure that the strength and reliability of the working structure meet the requirements.Subsequently,as the premise of accurate and stable extraction of unbalanced signals,the rotor speed should be kept stable during the sampling time and the time for the detection spindle to reach a stable speed should be shortened.In this paper,PID control is combined with BP neural network,and the fitness function of BP neural network is optimized by using PSO algorithm(Particle Swarm Optimization).The controller of detection station servo system is designed,and the control performance of detection servo system is improved.The simulation experiment is completed in Matlab environment,which verifies the superiority of the designed controller compared with PID controller and BP-PID controller.Finally,the key to the dynamic balance detection problem is to accurately collect the unbalance information from the vibration signal of the balancing machine,so consider the characteristics of the vibration signal of the balancing machine,it is necessary to use the appropriate signal processing method.In this paper,a dynamic balance vibration signal processing method based on CEEMDAN(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise)is proposed.The method combines orthogonal coefficient method,wavelet soft threshold method and ACF(Autocorrelation Function),which can effectively remove the independent IMF(Intrinsic Mode Function)and reconstruct the signal.The proposed method is compared with EEMD(Ensemble Empirical Mode Decomposition)and EMD(Empirical Mode Decomposition).The simulation results show that compared with EEMD and EMD,the proposed method can effectively reduce the reconstruction error of simulation signals.In the experiment,the real machine signal is sampled and the actual unbalance information is extracted,and several algorithms are compared and analyzed.It is proved that CEEMDAN is superior to EMD and EEMD in the application of dynamic balance detection problem.