Gear hobbing method for technological parameter self-learning optimization in machining process

Abstract

The invention discloses a gear hobbing method. The method is characterized in that in the gear hobbing process, self-learning optimization of gear hobbing technological parameters is carried out according to the following specific steps of 1, constructing a gear hobbing effectiveness evaluation model, 2, generating a gear hobbing technological parameter optimization group and 3, achieving self-learning optimization of the gear hobbing technological parameters. The method has the advantages that in the gear hobbing process, an improved back-propagation neural network and a differential evolution algorithm are used for improving the hobbing technological parameters, self-learning optimization of the technological parameters can be achieved, and better technological parameters are sought; the optimized technological parameters are stored in a technological living example set, and effective data support can be provided for new gear hobbing problems.

Description

technical field [0001] The invention relates to gear processing technology, in particular to a processing method for optimizing process parameters during gear hobbing. Background technique [0002] In actual gear hobbing, different batches of gears have different basic parameters and different processing requirements, so the processing parameters must be adjusted. In the same batch of gear processing, the basic parameters of the gears are the same, and the processing requirements are the same. However, as the processing progresses, various conditions such as hob micro-wear and thermal deformation will occur, and adjustments are also required. At present, no matter whether the process parameter decision is made manually by querying the manual or through software, there are many parameters and adjustment is troublesome. Most of the process personnel do repetitive work, and few can achieve self-learning adjustment of process parameters. The adjustment is inefficient, resulting ...

AI Technical Summary

Problems solved by technology

The hobbing process parameters have their particularity, and the output of a single neural network often appears unstable, that is, the generalization ability is poor, and the above method cannot be directly applied to the self-learning optimization of process parameters in the process of gear hobbing. Research in this area is lacking

Images