Juanjuan Qin, Baokang Zhang, Qi Wu, Ning Sun, and Maoqi Cao
Support vector machine, statistical process control, control chart patterns recognition, superposition method
Statistical process control is an effective quality control method, and its wide application has important practical significance in modern manufacturing industry. This paper studies a support vector machine classifier model based on the superposition method, which is suitable for pattern recognition of control chart of statistical process control and can effectively improve the intelligence level of quality monitoring in production. We also design a data preprocessing method based on sample information overlay to improve the recognition rate of sample features in response to the problem of unclear abnormal features in observed data. Through the comparison between the statistical characteristics and shape characteristics of the sample data, a two-layer support vector machine classifier model is established. Using the algorithm of Fruit fly optimisation to optimise the parameters of the two-layer support vector machine classifier can improve the recognition accuracy of the model. The effectiveness and superiority of the proposed model are verified by simulation experiments.
Important Links:
Go Back
