The Virtual Manufacturing Technology Of The Worsted Textiles Based On The Intelligent Prediction Models

With the promising application of the modern science and technology, the manufacturing technology with the automated and mechanized modes, which was ever satisfied when the category of the products with the huge yields was seldom varied, trends to the flexibility, systematization, and intelligence because the sorts of the products is frequently altered while their demand are often in a few quantity. The virtual manufacturing is one of the milestones in the modern advanced processing technologies, which has been widely researched and developed in the fields of the manufacturing industries such as machine, automobile, aviation etc. However, in textile industry of China, the latest science achievements are not widely used and researched. Especially, the advantages on raw materials and manual labor disappeared after China entered the WTO. At the same time, the variety of the species, small amount and quick consignment have become the fashion in the textiles trades. Therefore, it provides a fortunate chance for Chinese textile industries to greatly improve the textiles manufacturing through the modern advanced processing methods and information technology, which is also an important industry project eagerly required to be resolved now.Consequently, the virtual manufacturing technology (VMT) of worsted textiles has been researched and discussed based on the practical database from one representative wool mill using the intelligent tools or technologies such as artificial neural network (ANN), grey superior analysis (GS) and case-based reasoning (CBR) in the present thesis. It is possible for the manufacturers to realize the prediction and control of the quality, the adjustment of parameters, the design of the new products and the decision-making of the produce plans through VMT, which can improve the research level of the information and intelligence of China textile industry. Meanwhile, the basic theory and applied technology, which are utilized to optimize the VTM through the different methods, are valid and effective in this thesis. According to our work and efforts, three exciting achievements are shown as follows. 1. Building of the whole virtual manufacturing model of the worsted textilesFive key procedures, i.e., top dyeing, roving, spinning, weaving and finishing, in the course of from top dyeing to the end-use fabrics, are investigated and discussed, respectively, and the optimal prediction models have been established for the simulation of the corresponding manufacturing. Grey superior theory (GS), multilinear regress (MLR) and subjective experience analysis (SE) are compared and applied to optimize the input parameters of neural network. Each of four optimal models not only is respectively performed alone, but also combined together to simulate the whole textile processing, which makes probable to realize the virtual manufacturing.2. ANN models optimized by grey superior analysisThrough GS approach, the corresponding processing parameters of the prediction models can be sequenced and selected to input into model one by one according to the calculated degree of grey incidence. The optimal parameters (or group) extracted by GS are satisfied among GS, MLR, and SE after the mean accuracy (MA) and relative coefficient (R) are compared from each other.Through the combination of GS and ANN model, besides the optimization of the input parameters of ANN model, the correlative sensitive parameters that remarkably influence on the performance of the models and the quality can also be extracted. In fore-spinning, total drawing times and roving twist factor are regarded as the sensitive variables to roving quality; in spinning, draft ratio, roving weight, ring traveler and the nominal twist are high significant to spinning performance and yarn properties; back-beam height and filling count influence the weaving efficiency essentially; fiber diameter, warp count, filling twist, wash variable and steam parameter affect to physical properties of the cloth directly.3. Optimization and control of new product based on CBR and ANN modelsIn order to deal with the problems of the empiricism and variety of the new textile design, the most similar case used should be firstly retrieved according the main characteristic parameters of the textiles based on case-based reasoning algorithm. Then, the most similar case is simulated to verify the virtual processing by ANN models. If the results simulated are satisfied, it can be reused directly as the choice for actual processing; otherwise, the corresponding sensitive variables are adjusted in order to achieve the objection and to optimize the textile design. And lastly, the processing parameters can be renewed rapidly by means of this mode, and the high quality processing is as well as ensured.In addition, the deducing models, which are used to predict the process parameters according to the actual textile properties, including spinning parameters, material properties and finishing combination variables have been researched and built in order to optimize the combination of the resource and labor. The mean accuracy (MA) of the prediction models that deduce spinning draft ratio, ring traveler and spindle speeds exceed 97%; for fiber diameter, coefficient of variability of diameter, fiber length and coefficient of variability of length and short fiber content, MA also exceed 95%; for finishing combination variables, i.e., wash variable, cook variable and steam variable, their values are 95.59%, 87.43%, and 76.07%, respectively.In a word, the virtual manufacturing in worsted textile industry can be realized through the certain combination of GS, CBR and ANN, as well as the linkage with computer net technology. The flexibility in structure, intelligence of functions, and practicability and systematism of VMT are also to some extent executed.