Research On Current Ripple Noise Reduction Algorithm And Its Application

The electric window changes the original mechanical lifting structure,provides one key lifting and object anti-pinch functions,which greatly improves the using experience.The anti-pinch function is related to the life safety of drivers and passengers.At present,the mainstream anti-pinch control method in the market is to use Hall sensors to detect the motor motion state.In order to reduce the cost of hardware and improve integration,the demand for using the motor current ripple signal to realize the anti-pinch function is increasing.However,the current ripple signal of the motor is generated by the commutator and brush sliding contact of different stators in the working process of the DC brush motor.The signal contains a lot of noise,if directly used for the anti-pinch function,it will easily cause misjudgment.At present,there is little research on the denoising algorithm of motor current ripple signal,so this research has important theoretical and application value.In order to effectively extract the motor ripple signal to make anti-pinch judgment and save details at the same time,this paper studies the joint signal denoising algorithm of wavelet threshold denoising and variational mode decomposition optimized by sparrow search algorithm,and applies it to the own anti-pinch test system.The main research work is as follows:(1)The paper proposes an improved wavelet threshold denoising algorithm.First,we use two equations to establish a simulation model of DC motor to learn its working characteristics.By the way,we analyse the principle of current ripple of DC brush motor and compare the different anti-pinch algorithms.Then,the necessity of using wavelet threshold denoising algorithm for ripple signal and the influencing factors are analyzed.After introducing the calculate method of sample entropy,this paper proposes an improved wavelet threshold denoising algorithm.Finally,we use signal-to-noise ratio and root mean square error to measure the denoising effect.(2)The paper proposes a variational mode decomposition using sparrow search algorithm to optimize parameters,working with the improved wavelet threshold denoising algorithm.The variational mode decomposition algorithm is selected to decompose the ripple signal which contains DC component.The component will weaken the advantages of the first improved algorithm.Since the number of mode decomposition and penalty factor need to be set in advance,the sparrow search algorithm is selected to optimize variational mode decomposition,and the correlation coefficient is introduced to evaluate the noise level of each mode component.For high noise modes,the improved wavelet threshold denoising algorithm is used after decomposition,so as to improve the denoising effect of ripple signal containing DC component.(3)The software and hardware of window anti-pinch test system are designed and implemented.After analyzing the demand of system function,the paper completes the hardware circuit board,the MCU program and anti-pinch test software,which can display the result of time-frequency analysis of ripple signal,decomposition and denoising effect,as well as data saving,with a graphical interactive interface.