Fabric quality prediction method based on KNN improved PSO-BP algorithm

The invention discloses a fabric quality prediction method based on a KNN improved PSO-BP algorithm, and belongs to the technical field of textiles, and the method comprises the steps: obtaining data of a certain textile mill within a preset time through a collection device; selecting samples of process parameters and fabric weave parameters causing fabric defects as an original sample set; performing normalization processing on the original sample set; classifying the normalized sample set by adopting a KNN algorithm, and dividing the classified data into a training set and a test set according to a preset proportion; constructing a fabric quality prediction model; giving an initial weight and a threshold value to the fabric quality prediction model by using a particle swarm algorithm; training the fabric quality prediction model through the training set; and inputting the test set into a trained fabric quality prediction model, predicting a fabric quality grade, judging whether the fabric quality grade meet