On-line Identification Technology Of Ball Screw Wear State Based On Servo Drive Signal

Numerical Control Machine Tools play an important role in the automotive industry.The performance of CNC machine tools directly affect the quality of the product and manufacturing cost of automobiles.It is generally through the ball screw and other key transmission parts that complete the preset motion to complete the machining operation.With high-speed reciprocating interaction between nut,ball and screw,the wear and tear of these parts is likely to occur.The continuous deterioration of the screw will lead to intensified vibration,high failure rate of the machine tool,low machining accuracy and bad consistency of products.It is of great theoretical and practical value to carry out the monitoring and predictive maintenance of ball screw wear,which is also an effective method to avoid these problems.Atpresent,the research of the ball screw wear condition monitoring is at the experimental research stage.Aiming at on-line identification of ball screw wear state,the paper analyzes and summarizes the latest research results of ball screw wear monitoring at domestic and foreign.On knowledge of the traditional monitoring method,the problems include the following such as inconvenient installation,high cost,complicated application process and so on.This research systematically studies key technology of on-line monitoring of ball screw wear based on the motor current signal and servo axis tracking error.The main research work of this paper is as follows:(1)In order to solve the problem of on-line monitoring of the wear state of ball screw.Firstly,the feed drive system transmission performance is analyzed.that the performance with the continued deterioration of the screw will gradually become worse,the machine tool feed drive shaft vibration will be intensified,and the fluctuation of servo axis tracking error will be increased;Secondly,This paper proposed the ball screw wear deterioration monitoring method through the vibration characteristics and tracking error fluctuation characteristics,the feasibility of which is verified through experiments;Finally,the on-line monitoring method is put forward according to the ball screw wear state characterization(2)In order to solve the problem of on-line monitoring the ball screw wear deterioration parameter with the current signal,this paper studya new method to identify the feed-drive system natural frequency based on servo motor current signal.Firstly,the relationship between the vibration response and the current response is analyzed.Moreover,Based on the experimental modal analysis(EMA)and the operational modal analysis(OMA)method the current signal identification results are analyzed.The experimental results show that the current signal identification results are accurate,and the current signal has the same sensitivity as the vibration signal.It provides a method to extract the characteristic values of the ball screw’s wear state by the current signal monitoring.(3)In order to solve the problem of current signal feature extraction of the on-line monitoring ball screw wear deterioration,thepaper proposed wavelet packet energy eigenvalue to monitor the ball screw wear.And the natural frequency of feed system transmission mechanism is identified according the current signalto determine the wavelet packet energy decomposition layer.In order to reduce the interference factors in the signal analysis,the application of frequency conversion principle would automatically intercept the time domain current signal with ball screw wear sensitive information.The experimental results show that the current sensor has the advantages of the following: convenient installation and sensitive signal;The sensitive domain current signal can be intercepted accurately and automatically;The interference factors can be reduced;The extracted characteristic value is sensitive to the deterioration of the feed drive system ball screw.(4)In order to solve the problem of the recognition of ball screw wear state and suppression of the machining vibration.This paper established a rapid identification model of ball screw wear state base on support vector machine,and according to the workpiece position information of the 3-axis coordinate measurement the classification criteria of feed drive system wear state is established;The wavelet packet energy extracted by the current signal and the variance eigenvalue extracted by the following error are used as model input parameters to complete the training and parameter optimization of the recognition model of ball screw state,and the model was then validated.In order to suppress the machining vibration caused by the ball screw wear,the parameter adjustment rules are established according the field regulation experience and fuzzy control theory;The appropriate adjustment parameters can be provided for the different ball screw wear state to improve the quality of the parts.Finally,this paper builds a platform monitoring device for the ball screw wear test to give experimental verification.The experimental results show that the ball screw wear monitoring,wear state recognition and parameter adjustment can well suppress the vibration.Through deeply study of moditoring of the feed drive system wear condition,A corresponding monitoring system is built,as well as the servo parameters control strategy,which can effectively monitor the health of ball screw machine,avoid serious failure of machine tool,improve use rate and the economic efficiency of the factory,reduce scrap rate,and improve the quality consistency of the products.Regarding the above problems,the relevant theoretical and experimental researches are carried out,in order to improve the service performance and service life of CNC equipment.