Cotton Yarn Quality Prediction System Based On Neural Network

The quality of cotton yarn directly affects the economic benefits of cotton spinning enterprises,and there are many factors affecting the quality of cotton yarn,including the performance of raw cotton and the spinning process.In the same case of the spinning process,it is the raw cotton performance that plays a decisive role.Traditionally,through trial spinning to understand the yarn performance,although simple and effective,it will cause a lot of waste of resources,and can not meet the actual production needs of enterprises.How to use the artificial intelligence algorithm and advanced computer technology to accurately predict the quality of cotton yarn according to the large amount of historical processing data of enterprises will be of great significance for enterprises to reduce production costs and improve competitiveness.This paper firstly studies the basic process of spinning,understands the common raw cotton performance index and cotton yarn quality index,and comprehensively analyzes the main factors affecting the quality of cotton yarn,and finally determines the quality of cotton yarn through the performance of raw cotton.In order to solve the problem that the prediction accuracy of the traditional model is not high,the neural network algorithm is used to establish the cotton yarn quality prediction model.Select length,uniformity,specific strength,micronaire value,elongation,fineness,number of heteroplasms,impurity area,reflectance,yellow depth,maturity,fiber neps,inclusion,velvet rate,moisture regain,A total of 17 raw cotton performance indexes of sugar content and foreign fiber volume were used as input of the model;single yarn strength,strip dry CV,and breaking strength were used as the output of the model.A single BP neural network model was established and the advantages and disadvantages of the model were analyzed.In order to improve the shortcomings of BP neural network algorithm,genetic algorithm and particle swarm optimization algorithm are used to optimize BP neural network,and GA-BP neural network model and PSO-BP neural network model are established respectively.Experiments show that these two optimization algorithms can effectively improve the accuracy and stability of BP neural network model prediction.RBF neural network model and GRNN neural network model are proposed.Experiments show that using these two models is effective for cotton yarn quality prediction.Among them,the GRNN neural network model has a considerable degree of precision.The effect of using the grey correlation analysis algorithm to screen the raw cotton performance indicators to improve the accuracy of the model is verified.In order to facilitate the use of ordinary textile workers,this paper developed a cotton yarn quality prediction system based on neural network.The system needs analysis and design was completed,and finally the development of the system was realized using Python as the main programming language.The results show that the system has the characteristics of simple interface and convenient operation,which can effectively predict the quality of cotton yarn.