{"id":3128,"date":"2021-08-06T18:59:39","date_gmt":"2021-08-06T22:59:39","guid":{"rendered":"https:\/\/cecas.clemson.edu\/manufacturing-lab\/?page_id=3128"},"modified":"2025-03-21T12:44:45","modified_gmt":"2025-03-21T16:44:45","slug":"ankit-agarwal","status":"publish","type":"page","link":"https:\/\/cecas.clemson.edu\/manufacturing-lab\/people\/ankit-agarwal\/","title":{"rendered":"Ankit Agarwal"},"content":{"rendered":"

<\/div><\/div><\/span>

Ankit Agarwal<\/h1><\/span>
<\/div><\/div><\/div>

Postdoctoral Researcher
\nDepartment of Automotive Engineering
\nClemson University International Center for Automotive Research<\/h3>\n<\/div>
<\/a><\/a><\/a><\/div><\/div><\/div><\/div><\/div>
<\/span><\/div><\/div><\/div><\/div>

My decision to choose industry-oriented research as a professional career in the long term is based on moral values, outlook, and personal experiences. I always had a passion for learning how machines work and the evolution of new technologies from school days. This motivated me to choose Mechanical Engineering as a discipline during undergraduate studies. The laboratories of the undergraduate program were the most exciting part and helped me to develop an interest towards manufacturing processes and systems. I spent considerable time appreciating automation trends and CNC machine tools during undergraduate and postgraduate studies. Considering long-term career interests in the domain of advanced manufacturing, I opted for a doctoral degree program in the area of CNC Machine Tools and CAD\/CAM at Indian Institute of Technology Jodhpur, India after postgraduate studies. The doctoral degree program helped me to transform from an enthusiastic student to a thoughtful researcher and appreciate the multidisciplinary nature of current research trends. I have developed good knowledge base in the domain of advanced machining, geometric tolerances, and application of machine learning in manufacturing. In subsequent years of professional career, I look forward to contribute meaningfully in the areas integrating smart manufacturing and data sciences.<\/p>\n<\/div><\/div><\/div><\/div><\/div>

<\/div><\/div><\/span>

Research Work<\/h1><\/span>
<\/div><\/div><\/div>

HG-XAI: Human Guided eXplainable Artificial Intelligence<\/h3>\n

The HG-XAI integrates human expertise with the CNN-based Efficient-Net-b0 model to identify tool wear states efficiently. The work employs two XAI concepts to integrate human expertise effectively with the Efficient-Net-b0 model. Firstly, the feature map depicting the region of an input image contributing the most to the identified wear state is extracted through the last convolutional layer using the Gradient Class Activation Maps (Grad-CAM) technique. Secondly, classification probabilities \ud835\udc43 = [\ud835\udc431, \ud835\udc432, \ud835\udc433, \ud835\udc434] of all wear states are recorded from the Softmax layer of the Efficient-Net-b0 model. The maximum classification probability (\ud835\udc5a\ud835\udc4e\ud835\udc65(\ud835\udc43 )) value is used to decide the human involvement in identifying the tool wear state. If the \ud835\udc5a\ud835\udc4e\ud835\udc65(\ud835\udc43 ) is greater than the predefined threshold value, the classification result of the Efficient-Net-b0 model is considered as the final tool wear state. Alternatively, HG-XAI will seek the assistance of human expertise in deciding the tool wear state. The HG-XAI displays the captured image, feature map, and probability values of the two highest-ranked wear states to the human for making the final decision.<\/p>\n<\/div>

<\/span><\/div><\/div><\/div><\/div>

Design-as-a-Service (DaaS)<\/h3>\n

Cloud-based Design and Manufacturing (CBDM) has emerged as an enabler for product realization by integrating various service-based models. However, the existing framework does not thoroughly support the innovation ecosystem from concept to product realization by formally addressing economic challenges and human skillset requirements. Our work considers the augmentation of the Design-as-a-Service (DaaS) model into the existing CBDM framework for enabling systematic product innovations. The DaaS model proposes to connect skilled human resources with enterprises interested in transforming an idea into a product or solution through the CBDM framework. The model presents an approach for integrating human resources with various CBDM elements and end-users through a service-based model.\u00a0 It is established that the DaaS has the potential for rapid and economical product discovery and can be readily accessible to SMEs or independent individuals.<\/p>\n<\/div>

<\/span><\/div><\/div><\/div><\/div>

Contrived Tool Wear Methodology<\/h3>\n

The stochastic nature of the tool wear makes it difficult to model and requires a significant number of cutting tests and machining of a large volume of material that incurs high cost and time. To avoid such time-consuming and cost-intensive cutting tests, the methodology is developed to generate tool wear artificially using a grinding process. The tools are worn by taking several passes over a grinding wheel in a controlled environment. The performance of contrived and naturally worn tools is compared by analyzing various parameters such as process force, wear topography and chip formation, which shows a good agreement. Also, it is realized that the contrived wear method enables a consistent starting point while studying any wear stage of the worn-out tool, thereby decoupling the stochastic nature of tool wear.<\/p>\n<\/div>

<\/span><\/div><\/div><\/div><\/div>

Evolution of Tool Wear in Machining of Inconel 718<\/h3>\n

In recent years, machine vision techniques are becoming predominant for various manufacturing applications, such as identifying machine setup abnormalities, tool status monitoring, and machined surface analysis. An image recognition-based method is developed to measure the flank wear width\/area during trochoidal milling of Inconel 718. The proposed method is implemented as an automated computational program, and a series of experiments are performed to analyze the progression of the tool flank wear area over the volume of material removed. The developed image processing method is able to evaluate the flank wear width\/area accurately and efficiently.<\/p>\n<\/div>

<\/span><\/div><\/div><\/div><\/div>

Stochastic Machining<\/h3>\n

The concept of stochastic machining aims to introduce stochastic nature to the toolpath. The toolpath strategy involves random movement of the tool over the surface. It is hypothesized that the random motion of the tool avoids resonance by continuously varying the direction of cutting force and thereby reduces chatter. Also, the tool often passes over already machined regions of the workpiece and enables rapid dissipation of heat. The stochastic toolpath strategy devised herein is applied to generate toolpath for 3-axis ball-end milling of 2-Dimensional (2D) and free form surfaces.<\/p>\n<\/div>

<\/span><\/div><\/div><\/div>
<\/span><\/div><\/div><\/div><\/div>

Machining of Thin-walled Components<\/h3>\n