Prediction of Optimum Cutting Parameters Using Intelligent Techniques
Mohamed Sabry El-Agamy;
Abstract
Competition and advances in manufacturing industries require developing faster and more cost efficient systems that can enhance automation; integration and speed of design; production and inspection systems operations. New approaches such as Machine Vision (MV) and Artificial Neural Networks (ANNs) play crucial role in modern industry. They are featured with their flexibility, speed and integration capability that contribute in developing and enhancing integration between MSSs including CAD, CAM, CAPP and CAI systems. This study is step forward to this integration.
The aim of this research is to develop an integrated multi-function neuro-vision system that can be used as manufacturing decision support in reverse engineering where products with certain specifications are required to be produced.
The methodology used is based on integration of both MV techniques and ANNs paradigm in one hybrid intelligent system for prediction of optimal machining conditions. The idea of using both approaches in one system is maximizing their advantages while reducing their limitations. Matlab and LabView software were used for developing required processing engine and user interfaces.
The scope of developed system (CAI-X) covers both rotational and non-rotational parts. The part image is used as main input to generate required outputs.
The first module (CAI-1) consists of two models for material classification and recognition according to microstructure and surface colour texture. The first model scope covers eighteen Aluminum, cast iron and copper alloys. The second module covers six main materials including steel, Aluminum, plastic, Cast Iron, copper and wood.
The second module (CAI-2) is used in identifying standard material size required to produce part with known size. CAI-2 detect part edges, calculate actual dimensions and generates required material size and select standard size after adding required tolerances for holding the part and finishing process.
The third module (CAI-3) is feature extraction and matching module designed that CAI-3 classifies part shape and recognizes twenty four common features that can be inspected by vision systems. complicated or interacted features are not included in this work such as blind holes and pockets.
The fourth module (CAI-4) is a surface roughness classifier that is used in classification of different finish grades obtained by three machining processes: vertical milling, turning and casting processes. Selection of these surfaces types stems from the little research in that area.
The fifth module (CAI-5) is a process planning module that generates essential manufacturing conditions for producing given part after receiving the outputs generated from the first four modules. CAI-5 output includes machining process; machine type; cutting tool type and material; cutting parameters including feeds, speed and depth of cuts; coolant type and; fixation method required to hold the part during different machining processes.
The sixth module (CAI-6) is used in inspection of surface defects in finished parts including cracks, rust, dents, smearing, flaking and fretting.
Finally, the seventh module (CAI-7) is used in inspection of assembly parts. This module performs measurement, counting, verification of feature presence, assembly direction and proper surface coating.
The main contribution of this thesis can be summarized in the following points; CAI-X represents an integration of CAPP and CAI functions in hybrid intelligent system. CAI-X can be used in reverse engineering applications for designing new parts based on existing one and inspection of actually produced ones to verify that their
The aim of this research is to develop an integrated multi-function neuro-vision system that can be used as manufacturing decision support in reverse engineering where products with certain specifications are required to be produced.
The methodology used is based on integration of both MV techniques and ANNs paradigm in one hybrid intelligent system for prediction of optimal machining conditions. The idea of using both approaches in one system is maximizing their advantages while reducing their limitations. Matlab and LabView software were used for developing required processing engine and user interfaces.
The scope of developed system (CAI-X) covers both rotational and non-rotational parts. The part image is used as main input to generate required outputs.
The first module (CAI-1) consists of two models for material classification and recognition according to microstructure and surface colour texture. The first model scope covers eighteen Aluminum, cast iron and copper alloys. The second module covers six main materials including steel, Aluminum, plastic, Cast Iron, copper and wood.
The second module (CAI-2) is used in identifying standard material size required to produce part with known size. CAI-2 detect part edges, calculate actual dimensions and generates required material size and select standard size after adding required tolerances for holding the part and finishing process.
The third module (CAI-3) is feature extraction and matching module designed that CAI-3 classifies part shape and recognizes twenty four common features that can be inspected by vision systems. complicated or interacted features are not included in this work such as blind holes and pockets.
The fourth module (CAI-4) is a surface roughness classifier that is used in classification of different finish grades obtained by three machining processes: vertical milling, turning and casting processes. Selection of these surfaces types stems from the little research in that area.
The fifth module (CAI-5) is a process planning module that generates essential manufacturing conditions for producing given part after receiving the outputs generated from the first four modules. CAI-5 output includes machining process; machine type; cutting tool type and material; cutting parameters including feeds, speed and depth of cuts; coolant type and; fixation method required to hold the part during different machining processes.
The sixth module (CAI-6) is used in inspection of surface defects in finished parts including cracks, rust, dents, smearing, flaking and fretting.
Finally, the seventh module (CAI-7) is used in inspection of assembly parts. This module performs measurement, counting, verification of feature presence, assembly direction and proper surface coating.
The main contribution of this thesis can be summarized in the following points; CAI-X represents an integration of CAPP and CAI functions in hybrid intelligent system. CAI-X can be used in reverse engineering applications for designing new parts based on existing one and inspection of actually produced ones to verify that their
Other data
| Title | Prediction of Optimum Cutting Parameters Using Intelligent Techniques | Other Titles | إستنباط محددات التشغيل المثلى باستخدام الاساليب الذكية | Authors | Mohamed Sabry El-Agamy | Issue Date | 2017 |
Recommend this item
Similar Items from Core Recommender Database
Items in Ain Shams Scholar are protected by copyright, with all rights reserved, unless otherwise indicated.