Industrial manufacturing has become increasingly mechanized in recent years. The manufacturing productivity and costs of products are significantly impacted by machine cutting tools, which are a crucial component of industrial production. Tool breakage frequently happens suddenly and without warning in a practical manufacturing process, leading to an unusually unbalanced ratio of tool breakage samples to normal samples. Considering the need of current scenario of development of smart system in production units, the authors have proposed a deep learning-based model for prediction of damaged inserts which can give accurate results as compared to traditional techniques and manual inspection methods. The real time data of damaged tool and undamaged tools were collected, and the model training was done to predict defective inserts with high accuracy.
Top1. Introduction
With the advent of technological revolution and use of disruptive technologies the use of smart manufacturing techniques is of great importance to industries. The concept of smart manufacturing is helping business to enhance productivity and decrease the lead time required for manufacturing. Industry 4.0 can be used to accomplish smart design, smart machinery, smart monitoring, smart control, and lastly smart scheduling, which should raise the bar for automation. By leveraging advanced technologies such as AI and ML, smart manufacturing can help manufacturers identify and address quality issues in real-time, resulting in higher quality products. By automating processes and eliminating waste, smart manufacturing can help reduce production costs and increase profitability. It can also generate large amounts of data that can be used to optimize production processes, identify trends, and make informed business decisions. With technological improvements, there is a greater scope and demand for automated machining operations. Monitoring tool wear during machining operations is one of the essential jobs in automation. Tool failure can be caused by abrasion, adhesion, thermal stresses, or fracture . An abrupt brittle failure caused by extreme pressures, material flaws, sporadic cutting, or intense vibrations is referred to as a tool failure. Adhesive wear is caused by tool bits adhering to chips that produce build-up edges. The material softens as a result of increased temperature at the tool and task interface, which leads to thermal stresses that, in turn, distort tools through plastic deformation. In CNC machines, the use of inserts for cutting tools plays a vital role. A damaged insert if used can damage the product to be machined. Identification of defective inserts plays an important role in selection of cutting tools. Selection of inserts for cutting tools in CNC machines requires careful consideration of several factors, including the material being machined, the type of machining operation, the cutting speed and feed rate, the required surface finish, and the cost . By selecting the appropriate inserts for each machining operation, manufacturers can improve their efficiency, quality, and profitability. A defective cutting tool insert can damage the product and can lead to poor quality and effectiveness. The detection of defective cutting tools are done in industries by experienced persons working in the domain of CNC machining. Tool monitoring is an important work in industries as majority of times tool wear out during continuous production activities. Proper monitoring of the tools and inserts are essential to ensure quality of manufacturing. For tool monitoring direct methods like inspection tools are used to check the tool wear rate. The surface roughness of the inserts are checked to ensure the insert wear out rate. As compared to direct methods in current scenario indirect and contact less methods can give us more efficient results. Surface roughness is the word used to describe the waviness created on the work piece surface as a result of the feed marks that the cutting tool leaves behind when it processes the surface. In terms of machine vision, this difference in surface roughness is known as texture. As tool wear increases, this surface texture changes. By using image processing-based techniques, the shortcomings of the stylus approach to quantify surface roughness have been significantly reduced. For tool monitoring different intelligent systems are developed. Various industry 4.0 approaches are used for tool wear detection and monitoring. Here we have tried to use a simple mechanism for checking the defective tool inserts using CNN technique.
Here we have proposed a smart system to detect the defective tools. We have taken the images of different defective and good condition cutting tools and trained the computer using artificial neural network model to predict the condition of cutting tool whether it is defective or not defective based on its features captured in the image. We would be using a camera in the machine setup which will capture the image and predict whether the tool is defective or not defective. Here in the study we have taken milling cutter inserts of tungsten carbide, cemented carbide, ceramic and High speed steel materials.