最后，通过深入分析和测试对比上述3种预测方法对各输出因子的预测效果，基

Finally, through in-depth analysis and test comparison of the prediction effects of the above three prediction methods on each output factor, based on the characteristics of each yarn index and the amount of spun yarn data, a simple and efficient yarn quality based on multi-dimensional fusion is constructed Forecasting model.

Finally, by in-depth analysis and testing, we compare the predictive effects of each output factor by the above three prediction methods, and construct a simple and efficient image of yarn quality prediction based on the characteristics of each yarn indicator and the data volume of fine yarn varieties.

Finally, based on the characteristics of each yarn index and the amount of spinning variety data, a simple and efficient yarn quality prediction model based on multi-dimensional fusion is constructed.<br>