Machining task fine-grained monitoring method based on improved Gaussian hidden Markov model

Abstract

The invention relates to a machining task fine-grained monitoring method based on an improved Gaussian hidden Markov model, and the method comprises the steps: firstly adding a pre-training mechanismand a migration mechanism on the basis of the Gaussian hidden Markov model, and constructing the improved Gaussian hidden Markov model; and decoding a power signal acquired in the machining process through the improved Gaussian hidden Markov model trained in the steps, and judging a machining task executed by machining equipment corresponding to the power signal, thereby realizing monitoring of the machining task. On the basis of a variant Gaussian hidden Markov model of a traditional hidden Markov model, a pre-training mechanism and a migration mechanism are added to the variant Gaussian hidden Markov model, and the accuracy and robustness of the algorithm are improved.

Description

technical field [0001] The invention relates to the field of mechanical cutting, in particular to a fine-grained monitoring method for machining tasks based on an improved Gaussian hidden Markov model. Background technique [0002] With the development of modern industry, information technology and network technology have been widely used in production workshops, and digitalization has become the main direction of future industrial development. Undoubtedly, digital technology has significantly improved the production management, resource scheduling, and exception handling of the production line. The construction of digital workshops is also of great significance to the transformation and upgrading of enterprises and adapting to the market. Machining production occupies a major position in the production forms of all workshops, so the monitoring of machining tasks is one of the key points in the construction of digital workshops. [0003] Machining task monitoring refers to ...

AI Technical Summary

Problems solved by technology

At present, the technologies related to machining task monitoring are mainly divided into three categories. One is to use the method of machine vision to monitor machining tasks, but this method is not suitable for the scene of processing complex parts.

The second is to use radio frequency identification (RFID) to monitor machining tasks, but this method can only achieve coarse-grained processing task monitoring

However, the implementation of this type of method faces many difficulties: it is difficult to select appropriate parameters, and the operating parameters selected for different types of work, different equipment, and different tasks are not the same; it is not easy to obtain parameters, and the acquisition of optical, thermal, and mechanical parameters needs to be done outside the machine tool. The installation and debugging process is cumbersome and increases the complexity of the system; the robustness of the method is not strong, and the change of process parameters will lead to a significant drop in monitoring accuracy. If you want to improve the accuracy, you need to re-perform pattern recognition

Images