A method and system for intelligent prediction of faults of filling and packaging machine

Abstract

The invention discloses an intelligent prediction method and system for faults of a filling and packaging machine, comprising the following steps: collecting historical vibration signals and real-time vibration signals of a transmission system of the filling and packaging machine; dividing the vibration signals into training sets, and using sliding time windows to divide samples, Normalize the samples; build a DALCNN model, initialize the parameters of the network randomly, and train the network with the training data; use the trained model to predict the fault of the real-time signal, and output the time when the transmission system will fail next time. Make full use of the time information of vibration signals to accurately extract features, avoiding the problem of manual feature selection and inaccurate feature extraction; it solves the difficulty in obtaining full life cycle data under all working conditions, the distribution of training data and real-time signals is different, and model failure prediction The problem of poor effect improves the accuracy and achievability of failure prediction in actual production.

Description

technical field [0001] The invention relates to the field of failure prediction of filling and packaging machines, in particular to an intelligent prediction method and system for failure of filling and packaging machines. Background technique [0002] Filling and packaging machine refers to the mechatronics equipment that packs the product through filling, sealing and other processes, making the product safe and beautiful, and improving the added value of the product. As the main core mechanism of the packaging machine, its transmission system is composed of key components such as driving motors, couplings, rolling bearings, and ball screws, which are responsible for generating and transmitting the power required by the equipment. It is subject to cyclic alternating loads for a long time and is prone to failure. If it is not detected and replaced in time, it will seriously affect the production of the packaging line. In the actual packaging production, the packaging machin...

AI Technical Summary

Problems solved by technology

In the actual industrial production scene, it is difficult to obtain the full life cycle signal of the filling and packaging machine transmission system, and the feature extraction is inaccurate. The long and short time memory network (LSTM) combined with the convolutional neural network (CNN) is used to extract and analyze the vibration signal. Prediction, using the idea of ​​domain adaptation to add domain adaptive loss to the loss function of the model to improve the fault prediction effect of the model when there is less training data to solve the above problems

Images