Straightening adaptive optimization method and system, storage medium and computing equipment

Abstract

The invention discloses a straightening self-adaptive optimization method and system, a storage medium and computing equipment, and the method comprises the steps of collecting data in a three-point bending straightening process of a slender shaft, preferentially selecting potential data which does not reach an optimal expected value in deformation of the slender shaft but is larger in bending error fluctuation, and forming a new database through integration; developing a BP neural network, and using a new database for training and learning, so that the three-point bending intelligent straightening machine achieves efficient and high-precision automatic straightening, has the advantages of being high in self-learning capacity, high in self-adaptation level and excellent in performance, fully considers potential data, and is scientific and accurate in prediction result; wide application prospects are realized in the fields of robots, high-end equipment, electronic appliances and the like.

Description

Technical field[0001]The present invention belongs to the technical field of training machine learning, and specific relates to a pro-self-adaptive optimization method, system, storage medium, and computing device.Background technique[0002]The elongated shaft parts are one of the most common parts of the machinery industry, widely used in automotive, ship boat, aviation, aerospace, oil and other industrial fields. With the development of the industry, the accuracy requirements of these parts are also increasing, such as the air engine spindle in certain processes, requires the straightness accuracy than 0.3mm · m-1. These parts have experienced multi-channel complex steps such as cutting and heat treatment from the fuel to the finished product, which is easy to produce bending deformation, and many smaller bending deformations (≤ 1 mm) are difficult to distinguish, and these bending modifications Not timely processing will affect subsequent processing, and even have a large amount o...

AI Technical Summary

Problems solved by technology

These parts have gone through multiple complex processes such as cutting and heat treatment during the process from blank to finished product, and are prone to bending deformation. Many small bending deformations (bending amount ≤ 1mm) are difficult to distinguish with the naked eye, and if these bending deformations Failure to deal with it in time will affect subsequent processing, and even generate a large amount of waste products, causing great losses

[0003] For the alignment of the straightness of slender shaft parts, traditional factories rely entirely on the experience of workers to determine, and the alignment accuracy cannot be guaranteed. This method of straightness alignment relying on workers' experience can no longer meet the requirements of rapid industrial development.

In recent years, although some automatic straightening equipment has gradually appeared in the domestic market, it has not solved the problems of straightness measurement accuracy and straightness parameter calculation accuracy, and some equipment does not have self-learning ability and low intelligence. , the price is generally higher and other disadvantages

Images