Abstract:

Objective In the process of textile production, constant yarn tension is essential to ensure product quality, and excessive tension during yarn transmission will cause sudden stress change or even yarn breakage and will affect fabric formation, resulting in inelastic fabric structure and seriously affecting the quality of products. In order to detect yarn tension efficiently and conveniently, a method based on transverse vibration frequency non-contact detection of yarn tension is developed.
Method The transverse vibration characteristics of a moving yarn were studied by combining theoretical modeling, numerical analysis and experimental verification. The equation of yarn transverse vibration was derived based on chord vibration theory, and the yarn vibration frequency was obtained. The yarn tension was further calculated according to the relationship between frequency and tension. A yarn tension measurement method based on transverse vibration frequency measurement was proposed, and an open structure yarn transmission experimental platform was designed and built for experimental verification. The measured values were fitted to the generalized regression neural network model.
Results When the linear density is constant, the yarn vibration frequency increases with the increasing velocity; the greater the linear density, the smaller the yarn vibration frequency. First, the geometric parameters of the moving yarn are determined, and the fitting formula among the yarn motion speed, vibration frequency and tension is established. Then, the yarn vibration frequency and velocity data were obtained experimentally, and the vibration frequency and motion velocity were substituted into the fitting formula to obtain the yarn tension. Finally, the calculated yarn tension was fitted using the GRNN model to obtain the final tension results. The data generalized regression neural network fitting curve using the vibration frequency for measuring the yarn tension is on the same overall trend as the tension trend measured by the tension sensor(Fig. 5 and Fig. 6). In the actual working state, the yarn vibration frequency will be disturbed by other external factors such as machine vibration, and the yarn tension measured by the vibration frequency will also produce large fluctuations, so that the measurement results show a dispersed state is not conducive to the real-time control of the tension. Due to the uncertainty of the geometric parameters of the yarn, the error of the fitting formula, and the accuracy of the vibration frequency obtained by the high-speed camera, there are some errors in measuring the yarn tension based on the vibration. However, the measurement accuracy of the method can also be further improved, such as using a more accurate signal processing technology to process the vibration displacement signal. At the same time, although this method can effectively measure the yarn tension, but still has a certain application range. As the velocity increases, the system vibration frequency gradually decreases and disappears at the critical velocity. Therefore, in order to detect the vibration frequency of the system vibration, the yarn movement speed is less than the critical speed of the instability.
Conclusion The yarn vibration equation of the string vibration theory is established, and then the yarn vibration frequency is calculated by the fitting formula between the yarn tension, motion speed and vibration frequency; then the result is fitted through the generalized regression neural network model. The accuracy and reliability of the measurement of the yarn tension and the fitting algorithm are verified. At the same time, the influence of yarn density and motion speed on vibration frequency is also considered, and it has strong applicability for yarn with different line density and different motion speed.

Key words: yarn tension, non-contact testing, lateral vibration, frequency, high speed camera