STEP-NC based intelligent nonlinear process planning method

Abstract

The invention relates to an STEP-NC based intelligent nonlinear process planning method. The method comprises the steps of determining a machining operation method corresponding to a component machining feature type through a pre-trained BP neural network model; sequencing all machining steps in the machining operation method based on pre-defined machining step sequencing principles to obtain a reasonable machining step sequence; and in allusion to each machining step in the machining step sequence, selecting resources matched with the machining step, and optimizing each machining step and process parameters of each machining step by adopting a chaotic genetic algorithm to obtain an optimal machining process plan. According to the method, the chaotic algorithm, the genetic algorithm and the BP neural network are organically combined and applied to process optimization for the STEP-NC, thereby being capable of performing efficient, accurate and intelligent logical reasoning, effectivelysolving a complex process planning problem, and having important significance for the further study of the STEP-NC theory.

Description

technical field [0001] The invention relates to an intelligent nonlinear process planning method based on STEP-NC. Background technique [0002] STEP-NC describes the processing object in a manner based on manufacturing features, takes object-oriented processing steps as the basic unit of the processing flow, and describes the processing technology of manufacturing features in each processing step through processing operations, so process planning is the implementation of STEP-NC is an important step, and it is also the key to realize open, intelligent and networked STEP-NC numerical control system. [0003] Scholars at home and abroad have done a lot of meaningful research on process planning based on STEP-NC. For example, Suh et al. from Pohang University of Science and Technology in South Korea developed a workshop programming system PosSFP. In PosSFP, a graphical process flow chart is used to represent the process flow. The basic node is the process step and the AND-OR ...

AI Technical Summary

Problems solved by technology

Xu Xun and others from the University of Auckland in New Zealand considered the material hardness and processing depth as the input of the neural network, and used the BP neural network to optimize the milling process parameters. Such constraints, often can not get the optimal process parameters

Shakeri et al. automatically generate a sequence of steps by establishing a rule table that can reflect the priority relationship between processing steps. The optimization of the sequence of steps is based on the sum of the priority values ​​of all steps in the sequence of steps. The higher the sum of the priority values, the closer the work step sequence is to the optimal value, but this method needs to formulate a large number of rules and the algorithm is more complicated, and the versatility is not strong

Zhang Chengrui of Shandong University and others proposed an online planning mode based on step-level planning-feature-level planning-part-level planning, and respectively gave the optimization model of step-level milling consumption, the screening method of feature-level process routes, and the heuristic-based The work step sorting method, but the heuristic algorithm is based on local search, it is difficult to get the global optimal solution

Tian Xitian and others from Northwestern Polytechnical University took the auxiliary time as the optimization goal and used genetic algorithm to solve the processing step sequencing problem based on STEP-NC. However, the genetic algorithm is sensitive to the initial value, and it is easy to fall into the local extreme value, so that it is often impossible to obtain the optimal Routing

[0004] To sum up, the methods based on STEP-NC technology at home and abroad have certain limitations, such as poor intelligence and low efficiency.

Images