The Development of Cotton-yarn-quality Predicting System
Download 400.27 Kb. Pdf ko'rish
|
2011, Paxta-kalava sifatini bashorat qilish tizimini ishlab chiqish
1. The Neural Network Construction Module
Developments According to the requirements of network mode mentioned above, this part was developed by Matlab neural network toolbox, in which the Newcf. Function was called to design the BP network; finally the designed network was packaged into the same function file named net.m together with the data pre-designed program and the network training program. During the construction and training process, the network was activated by transferring the external data information and the design of the network nodes of hidden layer into the internal of function file through a call to this function file, after that the network training was completed. 2. The Modules Developments on Network Training and Yarn Quality Prediction These two parts were developed with the technique of VB and Matlab merging programming, the running of Matlab application program was controlled by the master control program of VB and the data transfer was carried out between the environments of VB and Matlab. A. The Realization Base of VB and Matlab Merging Programming ActiveX Automation protocol is a protocol that allows one application program (the control side) to control another (the server side). As VB could supports the ActiveX automation control side protocol and Matlab was able to support automated server-side protocol, therefore when the ActiveX connection established between them. The Matlab command could be called directly in VB environment, and then the realization of data transfer between the two programs followed. B. The Network Training Module Developments The function this module require to be achieved was completing the network training and the retention of the trained model on the basis of transferring the input messages from the network training interface in VB environment to the network models in Matlab environment as well as making them activated. The realization processes were as follows: First, a Matlab. application object named MT was created in the VB environment so that the ActiveX connection could be achieved between VB and Matlab ; then the activation and training of the network could be realized on the basement of transferring the input messages from the network training interface in VB environment to Matlab environment by MT.Putfullmatrix method, the call for Matlab by MT.Execute method to implement the net.m function files and the transfer of data information and the setting to the network hidden layer nodes to the network through function parameters latter ; After training, the trained network model was saved into model base, which was available to be called by yarn quality prediction module through a call for Matlab program to implement the "SAVE" order by MT.Execute method [6] . C. The Yarn Quality Predicting Module The function this module to be realized was transferring the fiber quality index value from the quality-predicting interface in VB environment to the trained network model that loaded in Matlab environment to realize the prediction and not only returning the prediction results to VB environment but also displaying them on quality prediction interface, the above functions was realized as follows: after the Matlab application object named MT being created in the VB environment, the trained network model the user specified was loaded to Matlab environment through a call for Matlab to implement the "LOAD" command by MT.Execute method;Then fiber index value was transferred into Matlab environment from the VB environment by MT.Putfullmatrix method and passed to the loaded network mode after being pre-processed, then the yarn quality was predicted through a call for the neural network emulation function "SIM". Finally the predicting results obtained were processed, transmitted in the VB environment by MT.Getfullmatrix method and displayed on the quality-predicting interface [7] . V. M ODEL L IBRARY C ONSTRUCTING Cotton yarns with different count were collected from several cotton mills within a year, along with the corresponding raw cotton information indicators. After being analyzed and processed, the data could be used to train the network models constructed by this system, and then all the trained models were saved in model database for the yarn quality prediction. VI. T HE I NSTANCE AND A NALYSIS The model"JC18.2T" was called from the model library to predict the quality of the JC18.2T yarn produced by a factory. The properties of the raw cotton the yarn used were summarized in Table 1, while both the predictive values and the measured values were shown in Table 2; As it could be seen in Table 2, the yarn tenacity and its evenness CV%were predicted with the relative error of less than 4% by the trained model. Since the predicting accuracy was in an acceptable range, the predicting results would be useful for guiding yarn -spinning practice. TABLE I. THE PROPERTIES OF COTTON FIBER Download 400.27 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling