Browsing by Author "Ramachandran, A."

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Body-in-White Joint Stiffness Sensitivity Analysis
    (2019) Ramachandran, A.; Reddy, H.; Chavali, T.; Hukar, A.; Somasekharan, J.
    The major objectives of a car design are reducing costs, maximizing performance, and improving fuel economy. The total mass of a car has a direct effect on all these objectives. Around 25% of the total mass of a car is typically accumulated in its body in white (BIW). Thus, reducing the total mass of the BIW while satisfying the target stiffness is of utmost importance in the early stages of design. Reducing the total mass of the BIW involves the identification of potential locations for mass reduction and stiffness improvement. The joints of the BIW are often the most critical locations that decide the overall stiffness of the BIW. Understanding the contribution of each joint toward the overall stiffness is thus of paramount importance toward improving the stiffness of the BIW.This paper describes a new approach of identifying the contribution of each joint in a BIW toward the overall stiffness of the BIW. The joints in the BIW are parametrized and Altair Optistruct [1] tool is used to find optimum value of each parameter. The contribution of different joints to the overall stiffness is different for different loads. This methodology was applied to a simplified BIW model for a combination of two load cases-static torsion and bending. The relative joint sensitivities of the joints in BIW have been identified and verified based on the results. The entire methodology has also been automated in a preprocessor to reduce the total time involved. � 2019 SAE International. All rights reserved.
  • No Thumbnail Available
    Item
    Body-in-White Joint Stiffness Sensitivity Analysis
    (SAE International, 2019) Ramachandran, A.; Reddy, H.; Chavali, T.; Hukar, A.; Somasekharan, J.
    The major objectives of a car design are reducing costs, maximizing performance, and improving fuel economy. The total mass of a car has a direct effect on all these objectives. Around 25% of the total mass of a car is typically accumulated in its body in white (BIW). Thus, reducing the total mass of the BIW while satisfying the target stiffness is of utmost importance in the early stages of design. Reducing the total mass of the BIW involves the identification of potential locations for mass reduction and stiffness improvement. The joints of the BIW are often the most critical locations that decide the overall stiffness of the BIW. Understanding the contribution of each joint toward the overall stiffness is thus of paramount importance toward improving the stiffness of the BIW.This paper describes a new approach of identifying the contribution of each joint in a BIW toward the overall stiffness of the BIW. The joints in the BIW are parametrized and Altair Optistruct [1] tool is used to find optimum value of each parameter. The contribution of different joints to the overall stiffness is different for different loads. This methodology was applied to a simplified BIW model for a combination of two load cases-static torsion and bending. The relative joint sensitivities of the joints in BIW have been identified and verified based on the results. The entire methodology has also been automated in a preprocessor to reduce the total time involved. © 2019 SAE International. All rights reserved.
  • No Thumbnail Available
    Item
    Word Sense Disambiguation using Bidirectional LSTM
    (Institute of Electrical and Electronics Engineers Inc., 2019) Rakshith, J.; Savasere, S.; Ramachandran, A.; Akhila, P.; Koolagudi, S.G.
    Word Sense Disambiguation is considered one of the challenging problems in natural language processing(NLP). LSTM-based Word Sense Disambiguation techniques have been shown effective through experiments. Models have been proposed before that employed LSTM to achieve state-of-the-art results. This paper presents an implementation and analysis of a Bidirectional LSTM model using openly available datasets (Semcor, MASC, SensEval-2 and SensEval-3) and knowledge base (WordNet). Our experiments showed that a similar state of the art results could be obtained with much less data or without external resources like knowledge graphs and parts of speech tagging. © 2019 IEEE.