A Fault Diagnosis Method of Packaging Production Line Based on State Data

Abstract

The present invention provides a method for fault diagnosis of a packaging production line based on state data, comprising: (1) collecting static state data generated during the operation of the equipment; (2) sequentially determining each fault name N i (3) collect the dynamic state data generated during the operation of the equipment; (4) compare all the current equipment state data with all the fault characteristics in the feature database to obtain the comparison Eigen matrix A kt , according to the comparison feature matrix A kt Obtain the current fault name; (5) compare all current equipment state data with all fault features in the feature database, and obtain the fault feature F in the current state t ;Analysis of fault characteristics F t , get the fault characteristic F t A collection of possible failures. The fault diagnosis method of the invention has low maintenance cost and is of great significance to improving the operation reliability of the packaging machinery and promoting the development of intelligent packaging machinery.

Description

technical field [0001] The invention belongs to the field of fault detection, and in particular relates to a state data-based fault diagnosis method for a packaging production line. Background technique [0002] With the development of science and technology, mechanized production has penetrated into all walks of life, among which the packaging industry has also realized mechanized production. Packaging machinery can use machinery to directly package all products, or package them in parts, which greatly improves the work efficiency of the packaging industry. Since the development of the packaging industry, the packaging process has been greatly simplified, but the main packaging processes, such as sealing, wrapping, and filling, still play an important role. [0003] As an important part of the modern packaging industry, packaging machinery is an indispensable and important equipment in the process of product production and processing. The Zhejiang provincial government st...

AI Technical Summary

Problems solved by technology

[0008] For packaging machinery, at present, enterprises mainly use preventive maintenance, post-event maintenance, predictive maintenance, condition monitoring maintenance, improvement maintenance, maintenance prevention and other methods to carry out work. There are generally problems such as difficult maintenance, high cost, and low efficiency. According to our analysis of Zhejiang Provincial packaging, printing, circuit breaker and other packaging production line typical application enterprise research and analysis, compared with a single equipment, the average maintenance time increased by 35%, and the maintenance cost increased by 30%

The reason is that because packaging machinery integrates machinery, electricity, hydraulics, control, communication, sensing, data, and software, and involves many fields such as machinery, artificial intelligence, electronics, communication, and information, the failure modes and characteristics are diverse and complex. , and the operation process is dynamic and changeable, but the enterprise still adopts traditional diagnosis and maintenance theories and methods to determine, without comprehensive consideration of multidisciplinary factors such as information, communication, network, software, artificial intelligence, management, etc. from the perspective of the system, lacking Theories and methods of scientific and reasonable condition monitoring and fault diagnosis

The structure of the packaging production line is complex, and the current fault diagnosis methods for equipment have the following problems: ① The operating status of each component of the equipment is not analyzed based on the operating status data; ② The technical threshold is high and the maintenance cost is high

Images