Research On Modeling Of Dye-uptake Rate Based On Data Driying For Reactive Dyes On Cotton Fibers

In recent years, China has taken the energy conservation and emission reduction as a basicnational policy, and taken the implementation of clean production, saving energy and reducingconsumption as the constraint condition of dyeing and printing industry. As an important speciesof natural fibers, cotton fiber has faced with awkward situations in dyeing with reactive dyes,such as high energy consumption and wastewater discharge. By establishing the mathematicalmodel of the relationship between process factors and the dyeing rate, the dyeing process can becontrolled via the model prediction and online feedback technology. It’s important to improveproduction efficiency, reduce production cost, improve product quality, and achieve energyconservation and emissions reduction.Firstly, we chose three reactive dyes which were used for color matching: Reactive Red3BE,Reactive Yellow3RE and Reactive Blue S3G. Secondly, the single-factor models betweenprocess factors and dye-uptake rate were established by BP Neural Network theory. Finally, thesemodels were used to predict the dye-uptake rate. The model validated that the relative error of thepredicted values and experimental values were within1.5%, so the model can accurately predictthe dye-uptake rate.First of all, we established the single-factor models by Grey System theory, and then themulti-factor models were established on combining the BP Neural Network theory and GreySystem theory. Finally, these models were used to predict the dye-uptake rate. The results showedthat these models could accurately predict the dye-uptake rate by the model verification andvalidation, and the relative errors of the predicted values and experimental values were within2.0%.Before heat preservation we chose two kinds of heating rate,2℃/min and1℃/min, toreach55℃when dyeing cotton fabric with reactive dyes. And then the dyeing rate curves weredrawn. After we got the instantaneous dye-up rate of each point in time through differentialderivative, we established the models about the instantaneous dye-up rate by BP Neural Networktheory. The results showed that on the premise of guarantee the quality of dyeing, dyeing rate in instantaneous model can be utilized to adjust the heating rate, so as to achieve the aim ofcontrolling dyeing rate to save dyeing time and improve efficiency.