Research And Application Of Motor Component Size Matching Method Based On KNN Algorithms

The three-phase asynchronous motor is mainly composed of a stator,a rotor,a front end cover and a rear end cover.The shape and size of each component after assembly directly determines the performance of the motor: if there is a situation in which the front and rear end bearing housings are not concentric,the motor It is prone to abnormal conditions such as excessive vibration,increased loss,increased electromagnetic noise,and sweeping of the hall;if the distance between the front and rear end bearing chambers exceeds the standard,abnormal conditions such as abnormal motor shaft rotation,shortened bearing life,and increased running noise may occur.Through the summary analysis of the production data,we found that most of the motor test anomalies are caused by large dimensional deviations after component assembly.Many companies are difficult to improve the quality of parts processing in the short term due to processing conditions.However,the low qualified rate of finished product assembly is a roadblock that restricts the improvement of production efficiency.As part of the research project of Factory Industry 4.0,we decided to abandon the previous manual screening mode and instead,after completing the dimensional measurement,input the data into the computer and filter,classify and assemble a batch(or batch)of parts through the program.Use the filtered paired co mponents.Reduce manual labor,improve efficiency,and improve accuracy.After research,we use the K-Nearest Neighbor(KNN)classification algorithm as the principle method for classifying and screening components.The idea of this method is that if the majority of the samples of the k most similar in the feature space(ie,the nearest neighbor in the feature space)belong to a certain category,the sample also belongs to this category.In the KNN algorithm,the selected neighbors are all objects that have been correctly classified.The method determines the category to which the sample to be classified belongs based on only the category of the nearest neighbor or samples.The KNN algorithm is more complicated in mathematics implementation.However,in the engineering field,because the data structure is relatively simple,when using the KNN method for classification decision,only need to correlate with a very small number of adjacent samples,so the KNN method is more suitable than other methods.This paper discusses the process of component classification and matching for a certain type of three-phase asynchronous motor: firstly,the parts that are absolutely qualified are matched and screened,and some parts are removed without matching;then the data of the remaining parts are randomly classified and matched with the standard.After comparing the model library,the most suitable matching scheme is selected;if there are still inappropriate accessories,the relevant data and prompt information are output and manually processed.This process has low requirements on equipment hardware,and has obvious effects on assembly accuracy improvement,and has high reference significance for related enterprises.