Research And Design Of A Fiber Ceramic Sleeve Classification System Based On DSP

In modern society, fast and accurate communication is essential. The fiber-opticcommunications with the communication capacity, transmission distance, good anti-jammingperformance and long life, etc., which makes fiber-optic communication has broad market. Thefiber tube, in fiber optic connectors, plays a vital role for optical fiber communication, and themarket demand for fiber-optic tube is also increasingly rising. The fiber optic tube needs to bere-classified according to the inner diameter size because of the production process and otheraspects. Currently, artificial classification is not only low efficient, but high labor costs, whileskilled workers is even more difficult to find, which restrict the development of enterprises. Ahigh efficient automated classification system is what the enterprise needs.In this context, this paper proposes to study about the accurate, rapid automaticclassification for the fiber tube. The paper builds the hardware control platform based on DSP,which controls three stepper motors, and detects the friction between the fiber tube and the rodwith the pressure sensors, then gets the classification result according to the detection data.Meanwhile, the system provides a good human-computer interaction design, which is easy tooperate. The system parameter can be set according to the actual need, which improves theusability of the system.Papers process the friction data detected with the weighted least squares method. Theprinciple that the friction between the fiber-optic tube aperture and standard stick varies linearlywas investigated. After the analysis of the actual friction data collected between the differentkinds of standard fiber-optic tube and standard detection stick, the standard friction model ofeach kind of the fiber tube aperture with the weighted least squares method was got, and thefriction model database was achieved by the calculate of the characteristic of the standardfriction model, the relation between all kinds of standard fiber-optic tube and the friction modeldatabase was established. A classification method was presented to compare the frictioncharacteristics of actual data collected with the friction model database. The method wasevaluated on the DSP, the classification was tested. The experimental results indicate that theclassification method is feasible, can detect18fiber tubes per minute with accuracy of up to99%, and improve the accuracy and speed of detection.