A FEATURE- AND KNOWLEDGE-BASED SEQUENCE PLANNING SYSTEM FOR THE MANUFACTURE OF SHEET METAL COMPONENTS USING PROGRESSIVE DIES

  • Yang Yang

Student thesis: Phd

Abstract

Sheet metal components play an important role in industry, and progressive dies are widely used for the mass production of sheet metal components. Sheet metal components manufactured using progressive dies are usually complex since they contain several features requiring different stamping operations; therefore, sequencing these operations is important because it influences the design of the die (e.g., the number of tool stations) and hence its size and cost. The sequence of operations is usually associated with features because they are the basic elements for the design and manufacture of any sheet metal component. The sequence of operations is also knowledge-intensive because heuristic and empirical knowledge are essential. In this research work, a heuristic method is developed to recognise features from sheet metal components. This method presents a two-stage algorithm which avoids the drawback in existing feature recognition algorithms, such as fixed feature patterns, requiring considerable pre-processes. The first stage of this algorithm extracts chains of thickness faces and sets of surface faces bounded by inner loops in the component. In the second stage, an extracted chain/set of faces is classified by, first, determining its topological and geometric characteristics and then comparing its characteristics with pre-set attributes for each feature. The recognised features are used as input to an algorithm specifically developed for retrieving similar sheet metal components, which can be reused or to help generate the sequence plans. In this algorithm, the main structure of a component is simplified and represented using a tree structure, and feature-related information of the component is represented as node attributes. A unique shape index is used for representing any given component, which reduces the computational efforts of retrieving similar candidates. Furthermore, a formalised procedure for calculating the degree of similarity between two compared components is given. Since the comparison is performed considering the general shape of the component, the bends and their directions, feature types, dimensions, and location, the retrieval results are explainable to the design and manufacturing engineers. The retrieved results are also useful to users, e.g., the sequence plan associated with the previously-proven component can be regarded as a reference for designing a new plan. The recognised features are also used as input to a feature- and knowledge-based sequence planning system. This system establishes a scheme that maps sheet metal features into their corresponding stamping operations, including shearing, bending, and forming. The mapped operations are initially placed in one of two categories and then the operations in one category are clubbed together using heuristic rules. The essential characteristics of each group are analysed to establish fuzzy relationships between any selected group and the remaining groups. The final membership value vector obtained from the fuzzy set theory is used to sequence these groups. The developed system has been tested with several representative components provided by our industrial partner, which contain features ranging from a simple cut-out to a lance with cut-out, and the generated plans bear a good degree of similarities with the near-optimal plans used in practice.
Date of Award31 Dec 2022
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorSrichand Hinduja (Supervisor) & Robert Heinemann (Supervisor)

Keywords

  • Sheet Metal
  • Progressive Die
  • Feature Recogntion
  • Sequence Planning
  • Model Retrieval

Cite this

'