Research On Model Identification Of Leather Cutting Machine’s Motion Control Testing Platform

With the increasingly development of equipment and manufacturing industries, thecorresponding technology of motion controlling has been steadily progressing, mainly incomputer numerical control (CNC) machine tools, microelectronic equipments and otherindustries. In the industries of clothing and footwear, the efficiency of labor has been greatlyimproved with the appearance of leather cutting machine which has been widely used all overthe world. With the advantages of stable operation, fast cutting speed, good quality of cuttingleather samples, less affected by human factors and high-usage of the leather, CNC leathercutting machine can save costs of production, shorten the processing time, reduce the manuallabor, at the same time can reduce labor intensity and so on. It can also reflect the level ofdevelopment of China’s clothing and footwear industry automation. At present, although ourcountry CNC leather cutting machine research and development has made some achievements,but there is still a large gap compared with foreign advanced level. Because of a lack of keytechnologies, the manufacturers of our country have great dependence on foreign products. Thereis an urgent need for independent research and development of CNC leather cutting machines.In order to facilitate the design of controller and error compensation algorithm and thedevelopment of the CNC leather cutting machines with high operating speed and high cuttingprecision subsequently, the designing scheme of “PC+motion control card+controlled object”is proposed in this paper. The servo driver, linear motor of CNC leather cutting machine areserved as controlled objects in the control system. The identification algorithm of least square isused to complete the identification of the controlled objects, thus providing conditions forindependent research and development of motion controller and its control algorithm, alsoproviding the foundation and basis for optimizing control.Firstly, the characteristics and effects of motion control card, servo driver and linear motorwhich are hardware parts of CNC leather cutting machine are introduced in this paper, and themotion control card is the core of the whole system. Secondly, the experimental platform is builtby using the independent research and development motion control card. The experiment ofcontrolled objects identification and data collection programs which are used as data sources for controlled object identification are designed on the basis of achieving the basic control functionsof CNC leather cutting machine. Finally, based on the three control models of ARX, OE andARMAX, system identification of the controlled objects is implemented by using the algorithmof least squares. The mathematical models of controlled objects under different control modelsare obtained, also the predictive control and analysis of the prediction error based on thesemodels, in order to further understand the characteristics of the controlled objects, and lay thefoundation to achieve better control strategies. Identification results can be used to adjust themotion control card parameter optimization, prediction, and fault detection system forecasting.