AI Applications in Sheet Metal Forming

Sheet metal components are indispensable in a great variety of products from large to micro sizes due to the high strength to weight ratio and relative ease of forming them. Design of sheet metal forming dies, however, is complex and largely experience-based. This chapter presents an overview of applying artificial intelligence (AI) tools and techniques in designing and planning of sheet metal progressive dies. Human skills and expertise are still deemed necessary, although advanced AI tools have complemented the design process well.

This chapter presents classification and representation of shape features in sheet-metal parts. Shape features in a sheet-metal part model can be associated with volume subtraction from base-sheet (e.g., piercing/blanking operation), deformation/modification of base-sheet or forming operation on base-sheet. The shape features in a sheet-metal part model are classified as (i) Volumetric features and (ii) Deformation features. These features can also be classified as ‘2-dimensional (2D) features’ (volumetric features) and ‘3-dimensional (3D) features’ (deformation features) as a result of modification and forming of base-sheet. The thickness is constant for a sheet-metal part. Hence, the shape features in a sheet-metal part are also referred as constant thickness features. The representation, classification, and extraction procedures of the sheet-metal features are based on topology and geometry. The novelty is that the proposed classification and representation of sheet-metal features is based purely on shape entities and therefore it is possible to automatically extract the features from any sheet-metal part model. This enables the use of the proposed classification and representation to be unambiguous and application independent and to handle equivalences between feature labels and their representations among applications. The definition presented for a feature can also be extended to include application specific information.

To automate the process of die design, firstly all design features of sheet metal parts are to be extracted automatically from drawing files by a computer-aided system. After feature extraction, next important activity is manufacturability assessment of sheet metal parts. Traditional process of manufacturability assessment of sheet metal parts involves calculations and decisions, which have to be made on the basis of experience and practice codes without the computer aids. In the present chapter, an automatic feature recognition system is described. The system initially extracts the basic entities from the 3-D CAD model and recognizes various design features of flat parts, bending parts, and deep drawn parts. The system is coded in AutoLISP language. The system displays the details of all design features of part in the prompt area of AutoCAD software. The system has been installed on Autodesk AutoCAD software. The present chapter also describes a knowledge-based system (KBS) for manufacturability assessment of sheet metal parts. Knowledge obtained from published literature, die designers, and process planners has been analyzed, tabulated, and incorporated into a set of production rules of the IF–THEN variety. The system is coded in the AutoLISP language and user interface is developed using visual basic (VB). The system output includes recommendations on the suitability of design features of the part for required manufacturing operations. The knowledge base of this system can be modified depending upon the capabilities of a specific shop floor. The low cost of the system makes it affordable for process planners working in small-and medium-size sheet metal industries.

In sheet metal industries, blanking dies are considered as basic types of stamping dies. Die design is a highly experience-based task. There is no computer-aided system available for automation of design of blanking dies. This chapter describes a knowledge-based system for design of sheet metal blanking dies. The developed system can be used to analyze different die design techniques and complete the design task base on an optimum design configuration. A knowledge base for the selection of a most suitable design from 14 different designs has been constructed. The developed system uses parametric design approach for design and finally engineering drawings in 2-dimensions (2-D) and 3-dimensions (3-D) can be generated automatically. The system is tested through sample runs. The system was built on the AutoCAD platform with Visual Basic, AutoLISP programming languages and MS-Access for the database.

This chapter describes a knowledge-based system (KBS) developed for design of deep drawing die for axisymmetric parts. The production rule-based KBS approach of artificial intelligence (AI) is utilized for development of the proposed system. The overall system is organized in 4 subsystems and 27 modules. System modules are coded in AutoLISP language and user interface is created using Visual Basic 6.0 and interface with AutoCAD software. The proposed system is capable to automate all major activities of design of deep drawing die such as process planning, design of strip-layout, selection of die components, and modeling of die components and die assembly. The system is user interactive, flexible, and has low cost of implementation.

This work is concerned with the process design of multistage deep drawing, where an integrated artificial intelligence (AI) approach is presented with a special focus on box-shaped parts. This approach combines three AI tools, namely part shape recognition, expert system for process design governing rules, and search and optimization via dynamic programming. Validation and final selection of optimized process plans are done using finite element analysis with full account of the formability limits of the material used. The main advantage of the proposed integrated approach is its capability of generating valid, optimized process plans in a relatively short time compared to traditional approaches. Two case studies are presented for demonstrating its effectiveness.

In this present study, the Knowledge-Based System (KBS) for non-axisymmetric deep drawing products with elliptical shape was constructed by using process sequence design. The system developed in this work consists of four modules. The first one is recognition of shape module to recognize non-axisymmetric products. The second one is a three-dimensional (3-D) modeling module to calculate the surface area for non-axisymmetric products. The third one is a blank design module to create an oval-shaped blank with the identical surface area. The forth one is a process planning module based on the production rules that play the best important role in the knowledge-based system for manufacturing. The production rules are generated and upgraded by interviewing with field engineers. Especially, the drawing coefficient, the punch and die radii for elliptical shape parts are considered as main design parameters. The KBS constructed in this study would be very useful to reduce lead-time for manufacturing and improve an accuracy of products.

The present chapter describes an expert system for automatic design of compound dies. The knowledge base of this system is constructed through coding of more than 1500 production rules of ‘IF-THEN’ variety in AutoLISP language. The system is structured in 22 modules. User interface is created using Visual Basic (VB) and AutoCAD software. The proposed system automates all major activities of design of compound die. The system finally generates the drawings (2-D and 3-D) of die components and die assembly of compound die automatically in the drawing editor of AutoCAD software. These drawings can be directly used for die manufacturing. The system can be implemented on a PC having VB and AutoCAD software and therefore its low cost of implementation makes it affordable for small scale enterprises.

The present describes the research work involved in prediction of life of compound die using Artificial Neural Network (ANN). The parameters affecting the life of compound die are investigated through Finite Element Analysis (FEA) and the critical simulation values are determined. Based on FEA results, S–N approach is used for calculation of number of cycles of compound die. The number of cycles gives the number of sheet metal parts that can be produced on compound die before its failure. The proposed ANN model is tested successfully on different compound dies designed for manufacturing various industrial sheet metal parts.

This chapter describes a knowledge-based system (KBS) for automatic design of bending dies. The system is developed using the production rule based approach of artificial intelligence (AI). The overall system is organized in 3 subsystems and 19 modules. System modules are coded in visual basic 6.0 language. The system is integrated with inventor CAD software. The proposed system automates all activities of design of bending die and finally gives outputs in form of drawings of die components and die assembly. It eliminates the dependency on domain experts for die design. The system is easy to operate and its knowledge base can be modified and/or updated on the advancement in technology in future.