Veermata Jijabai Technological Institute
Recent publications
Bio inspired textured surfaces are used in diversified engineering applications namely in tribology, biomedical, antifouling, anti-icing, micro fluidics, anti-bacterial surfaces, solar cells, etc. These textured surfaces are categorized as dimpled textured surfaces (negative) or pillared (positive) textured surfaces. Presently, laser machining is used to create dimpled textured surfaces. However very limited number of micro manufacturing process are available which can fabricate positive textured surface consisting of pillars of varying cross-sectional areas. Therefore, the primary objective of this work is to explore the reverse micro EDM process (R-MEDM) to fabricate pillared (positive) textured surface even on metallic surface. Earlier studies in R-MEDM had reported challenges involved in preparation of electrode used in the process. It is very difficult to machine thousands of micro-holes on plate which act as electrode in R-MEDM process. These holes will be replicated on work surface during processing. The work presented here addressed solutions to the challenges of preparation of electrodes and improves the process by using modified bubble assisted flushing method. Detail experimental investigation has been carried out in this work to evaluate the effect of process parameters on material removal rate, explore the use of low-cost sieves as electrodes followed by contact angle measurement on a fabricated textured surface. It is found that sieves are low-cost solution to successfully fabricate hydrophobic textured surface. Passing bubble enhances the debris flushing efficiency. Improvement of material removal rate by 20–25% and contact angle by 15–18% is realized using a bubble flushing method. The array of square rods of 200 µm side and 250 µm lengths were successfully fabricated on 6 mm bulk rod of brass with a good dimensional accuracy.
U.S. General Service Administration (GSA), 2013 guideline were used in the progressive collapse analysis of seismically designed three-dimensional special moment resistant frame. The analysis was carried out to assess the vulnerability of low-rise, mid-rise and high-rise three-dimensional special moment resistant frame with six bays in either direction. All the structures considered for this study were designed according to the Indian Standard codes. To evaluate the behaviour of the structure, linear static analysis (LSA), non-linear static analysis (NLSA), linear dynamic analysis (LDA) and non-linear dynamic analysis (NLDA) have been carried out by using Alternate Path Method (APM) for column removal at the critical locations based on these guidelines. The results show negligible displacement versus time history when compared between LDA and NLDA. It also reveals that the removal of the internal column leads to less vibration when compared to the other two locations during sudden removal. The height of the structure will largely affect the behaviour of the structure during any sudden column removal.
The usage of social networking platforms for interaction and meeting has grown significantly. However, as social networks continue to expand, Twitter has emerged as a vital social media for real life and has become a major source of spam. To address the aforementioned problems, spam identification on social media becomes an increasingly crucial task. The features that are present in the high dimension data of social networks cannot be used effectively by the existing approaches. In order to filter the spam information in social media, a Chimp Sailfish Optimization-based Deep Neuro Fuzzy Network (ChSO-based DNFN) is proposed. The proposed method effectively performs well under high dimensional data in real platform environment using deep learning classifier. It is more robust and generates optimal result and also reduces the computational complexity problems. Additionally, the proposed approach demonstrated improved performance in terms of metrics like precision, recall, and F-measure, which were measured using a 5k continuous dataset and yielded values of 0.894, 0.903, and 0.898, respectively.
This letter proposes a control algorithm as a stepping stone for the stabilization and control of structured and unstructured systems discussed in the literature. The foundation behind this proposed controller design theory utilizes an invariant target manifold giving rise to a non-degenerate form, through which the definition of certain passive outputs and storage functions provide the control law for stabilizing the system. The proposed control framework is developed by integrating the notions of immersion and passivity hence labeled as the “Passivity and Immersion (P&I) approach”. Immersing the system dynamics into the user-defined lower order target dynamics and using the concept of attractivity of the manifold based upon the passivity theory results in encompassing various design paradigms under a single roof of the P&I approach. The benchmark examples are solved to show the effectiveness of the proposed P&I approach in confirming how the various design paradigms can be unified.
The precise prediction of groundwater level is essential for water reserve management. In this study, the two intelligence models, viz, Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS), for predicting groundwater levels in Pravara River seven groundwater stations such as Sangamner, Deolali, Shrirampur, Rahata, Sonai, Taklibhan, and Loni are examined. To forecast groundwater level, 19 years' (2000–2018) data sets were used as target data including rainfall and temperature as input data. The results confirmed that ANN and ANFIS models can precisely predict groundwater level. The ANFIS model, which had R2 = 81.70, RMSE = 1.389, NS = 0.817, and NMSE = 0.193, outperformed the ANN model, which had an R2 = 76.32, RMSE = 0.539, NS = 0.584, and NMSE = 0.159. By comparing the observed and predicted groundwater levels at the Sangamner, Deolali, Shrirampur, Rahata, Sonai, Takali, and Loni groundwater monitoring stations using the best ANN and ANFIS model, it can be seen that the two values are nearly close to each other. The unknown value at the Sonai, Taklibhan, and Loni station from 2000 to 2014 is predicted using the Inverse Distance Weighting (IDW) interpolation method. The inference drawn from this study would be beneficial to groundwater management.
The virtual derivatives computation and successive derivations of virtual inputs in an adaptive backstepping controller cause the explosion of complexity. Moreover, the feedback linearization has poor robustness features and necessitates exact estimation of the feedback control law’s coefficients. Due to measurement noise, the model-based estimation techniques for identifying uncertainties result in inaccurate gradient and Hessian calculations. Such limitations lead to model and measurement uncertainties that prevent effective stabilization and control of nonlinear systems. Machine learning-based data-driven approaches offer effective tools for identifying dynamical systems and uncertainties with minimal prior knowledge of the model structure. Therefore, the contribution of this research is two-fold: First, the general controller design theory is proposed which utilizes the idea of an invariant target manifold giving rise to a non-degenerate two form, through which the definition of certain passive outputs and storage functions leads to a generation of control law for stabilizing the system. Since the above concepts are linked with the Immersion and Invariance (I&I) design policy and the passivity theory of controller design, the proposed methodology is labeled as the “Passivity and Immersion (P&I) based approach”. Second, the proposed P&I approach is integrated with a Bayesian nonparametric approach, particularly the Gaussian Process for stabilization and control of the partially unknown nonlinear systems. The effectiveness of the proposed methodologies has been evaluated on an inverted pendulum using MATLAB in the presence of input-output uncertainties.
The present work deals with steady state air flow analysis of electric motor having 20 hp rating running at 1450 rpm. The motor is being used to run the belt pull system to drive the exhaust fan in the industry. Air flow analysis of electric motor is carried out to predict temperature distribution over the motor. The modeling of the complete motor is done in CATIA. Meshing of whole geometry is done in ICEM CFD14.5. Results are obtained using FLUENT. In this research only copper losses and iron losses are taken into consideration based on the output obtained from electrical simulation software. The copper and iron losses are found out from electromagnetic analysis using Motorsolve software. Losses are treated as heat source or input to find out temperature distribution. To improve the accuracy, the computational fluid dynamics (CFD) analysis is performed by considering the air flow around casing and fins and thermal generation due to the losses. It is observed that there is a significant rise in temperature on casing and on fins for 30 °C ambient temperature.
Proportional-Integral-Derivative (PID) controller is widely used across various industrial process control applications because of its straightforward implementation. However, it can be challenging to fine-tune the PID parameters in practice to achieve robust performance. The paper proposes a model-based reinforcement learning (RL) framework to tune PID controllers leveraging the probabilistic inference for learning control (PILCO) method. In particular, an optimal policy given by PILCO is transformed into a set of robust PID tuning parameters for underactuated mechanical systems. The robustness of the devised controller is verified with simulation studies for a benchmark cart-pole system under server disturbances and system parameter uncertainties.
Biomass is the most versatile feedstock for renewable energy and chemical production. Biochemical techniques such as fermentation and biomethanation have been extensively developed for converting biomass into bioethanol, biogas, and high‐value platform chemicals. However, the techno‐economic feasibility of the various biochemical techniques for the production of a range of biofuels and chemicals has not been fully consolidated in a review. This paper reviews the techno‐economic studies of biochemical conversion of biomass in a comparative fashion between feedstocks, treatment methods, and product types. The review starts with an overview of various biomass treatment approaches and the need for pre‐treatment for processing second‐generation feedstocks. This is followed by a review of the main biochemical conversion processes, offering insights into process stages, product yields and quality, as well as commercialisation prospects and challenges. The various techno‐economic aspects of biomass conversion via biochemical techniques, such as conversion efficiency, production capacity, minimum selling price, capital cost, unit production cost, and profitability metrics, are critically reviewed. It was found that bioethanol and biogas production is the most commercially viable products from the biochemical processing of biomass. The production of other biofuels and chemicals such as biobutanol, biohydrogen, furfural, volatile fatty acids, succinate, levulinic acid, and sugar alcohols via biochemical techniques are still largely limited by low conversion, frail microbial strains, cost of enzymes, and separation and refining challenges. Overcoming these technical bottlenecks, as well as feedstock price and supply security, are crucial to enhance the overall techno‐economic attractiveness of biochemical processes for fuels and chemical production from biomass resources. This article is protected by copyright. All rights reserved.
The reconfigurable manufacturing system (RMS) meets the challenges of dynamic customer demands, technological advancements, and reducing lead time, among other things. It was necessary to have a framework that can assist in increasing RMS adoption as well as evaluating its performance. The present study seeks to develop a hybrid framework for prioritizing performance metrics of RMS that helps the designers of the manufacturing system in decision making. A total of 31 enablers for RMS were identified through a literature review, the weight of each enabler was computed by the fuzzy-AHP (analytic hierarchy process) method, and the fuzzy-TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) method was used to prioritize 22 performance metrics of RMS. According to the findings of the presented study, smart factory enablers have the highest weightage among all of the main criteria enablers, followed by strategy and policy enablers. The prioritization of performance metrics reveals that the top three most important performance metrics for RMS are lead time, reconfiguration time, and product flexibility. The feasibility and appropriateness of the framework was tested through a case analysis in the manufacturing organization. The framework developed has a high capacity to assist designers during the adoption of the RMS and will facilitate the identification of the relevant parameters. The authors believe that researchers and professionals will find this study as a ready reference for the stepwise adoption of RMS. The study presented here is likely the first to present a hybrid framework for RMS in which a set of enablers and performance metrics are presented together.
Epilepsy is a major threat to society regarding the treatment time, cost, and unpredictable nature of the disease, imposing an urgent need for intelligent analysis. Electroencephalogram (EEG) is a commonly deployed test for detecting epilepsy that analyses the electrical activity of an individual's brain. In this work, an optimized deep sequential model is proposed to improve the seizure classification performance based on a hybrid feature set derived from EEG signals. A novel hybridized algorithm called Battle Royale Search and Rescue optimization (BRRO) is proposed for optimizing the deep learning model. Also, a proposed hybrid feature set utilizes the empirical mode decomposition (EMD), variational mode decomposition (VMD), and empirical wavelet transform (EWT). This feature set is created to capture the discriminative temporal property of the dataset. The proposed method is validated using publically available datasets. The results manifest that the proposed optimized algorithm provides better results in comparison to other existing alternatives.
For effective online monitoring to assess thermal performance and life expectancy, Top-oil temperature (TOT) and Hot spot Temperature (HST) should be accurately esti-mated. In general, the thermal-electrical analogy is used for the first-order model based on the Resistance-Capacitance (RC) circuit structure to approximate the evolution of thermal performance. Traditionally, the TOT model parameters are identified by minimizing the error between estimated and actual values using the input-output data. The Gradient-based estimation (Least-square minimization) guarantees the convergence of the parameter estimation error to zero only when the Persistence of Excitation (PE) condition holds for regressor signals. As the choice of input-output data used for parameter identification is crucial, the Design of Experiment (DoE) is generally performed in the laboratory to satisfy PE conditions. Therefore, the TOT model parameter estimation problem is reformulated from a system identification perspective by exploring the finite-time estimators (FTE) that accurately estimate the TOT model parameters for non-PE data from real-time operating trans-formers without DoE. The experimental analysis is carried out on MATLAB to demonstrate the identified PE problem, effect of DoE, and the performance of finite time estimators for real-time scenario-based non-PE data in thermal modeling of the transformer.
In this paper, we analyze recent disturbances in the generation-deficit Mumbai distribution network which imports almost the same amount of power that it generates. A new I2-analysis is used which has several advantages. In this approach, the alternating sinusoidal current waveform is squared and converted into positive pulses averaged over a cycle, which mimic particles, and which constitute power displacement in the power network. Magnitudes of pulses are used in lieu of phasor vectors to determine electrical quantities such as line-flows and voltages from the accumulated magnitudes in the circuit elements. The procedure for computation takes an entirely different route and is suitable for analyzing disturbances in the context of the systems mentioned above. The example includes a recent blackout in the Mumbai region.
During the machining process, coolant is utilized to remove chips and tiny abrasive particles created during the machining process as well as to lessen heat concentration and friction between tools and chips. The machining performances, such as tool life, surface roughness, cutting forces, retention of mechanical properties of the work material, etc., are also desired to be retained or improved at the same time. This presented research work’s main goal is to investigate and analyze the impact of coolant at 0 °C on input machining parameters when turning SS304 (an austenitic stainless steel of the 300 series with high corrosion resistance) on a CNC lathe and to optimize the input variable factors, such as feed rate, cutting speed, and depth of cut for the best machining conditions, and each input cutting parameter is given a weight using the analytic hierarchy process (AHP) technique. A novel experimental setup is created to decrease the temperature of emulsion coolant and to use it in control conditions during machining operation. To research and assess the impact on the workpiece surface roughness, forces produced during actual cutting operations, the rate of tool wear, and the rate of material removal, twenty-seven sets of experiments using the partial factorial design approach are devised and carried out. Prioritizing the many optimal solutions accessible for this work is done using the technique for the order of preference by similarity to ideal solution (TOPSIS) and grey relation grade (GRG) approaches. Further, the surface finish of the workpiece after machining, rate of tool wear, cutting force generated during machining, and material removal rate from the workpiece were compared with traditionally/conventionally used input parameters with newly obtained optimized parameters through this work. Approximately a 30% improvement is observed in output parameters compared with using traditional parameters, and was close to the 50% of the result obtained through cryogenic machining. The work piece’s chip morphology along with tool wear was observed in form of SEM images, and it supports the claim of the surface finish and tool wear. The material removal rate was physically observed during machining. SEM pictures were used to physically validate the changes in tool wear. It has also been shown that keeping the coolant temperature at 0 °C significantly improves a number of work quality and machining characteristics. This method offers a substitute for cryogenic machining, making it useful for the manufacturing sectors.
Most of recent constructed buildings have a common feature, the ground storey remains open, and there are no partition walls in ground storey. It is highly flexible in the ground storey, the relative horizontal displacement in the ground storey is much higher than what each of the storeys above it has. This highly flexible ground storey is also called soft storey. Soft storey (relatively weak ground storey) building shows a higher chance of collapsing during earthquake because of flexibility. The total horizontal force in ground storey is much higher than its upper storeys. Higher flexibility directly results in higher displacement in lower floors of such framed buildings. In such buildings due to open storey the moments & shear force in columns is on higher side as compared to other framed buildings. As the study suggest it is highly risky and need at most care in designing such buildings with open ground storey. For all the new RC frame buildings it is recommended that there should be no such sudden reduction in stiffness. ⁴ To reduce such effect it is recommended to have masonry walls or RC shear walls in ground storey. Infill helps in achieving a more flexible structure, and gives a strut like action between columns and beams (force is transferred from one node to another). In the present study the high rise building is studied by considering presence of infill walls and shear wall. The response spectrum analysis is performed for building with infill arrangement and RC shear wall on open ground storey.
Verifying image authentication is necessary when images support evidence for essential purposes such as law enforcement and forensic investigation. The self-embedding fragile watermarking method is a powerful technique for confirming exact content authentication and restoration. This paper proposes the integer wavelet transform(IWT) based watermarking approach for accurate authentication and restoration. Here, watermark bits are calculated from each block of 2×2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2\times 2$$\end{document} pixels. The authentication bits are inserted into the block’s three LSBs, while restoration bits are inserted into the mapping block. Restoration bits are calculated from the five MSBs of the cover image using IWT. In contrast, authentication bits are calculated from secret keys, MSBs and, the location of pixels. The experimental outcomes prove that the PSNR and NCC of the reconstructed image in the suggested scheme are significantly high analyzed with the existing methods.
The usage of the internet as a fast medium for spreading fake news reinforces the requirement of computational utensils in order to fight for it. Fake videos also called deep fakes that create great intimidation in society in an assortment of social and political behaviour. It can also be utilized for malevolent intentions. Owing to the availability of deep fake generation algorithms at cheap computation power in cloud platforms, realistic fake videos or images are created. However, it is more critical to detect fake content because of the increased complexity of leveraging various approaches to smudge the tampering. Therefore, this work proposes a novel framework to detect fake videos through the utilization of transfer learning in autoencoders and a hybrid model of convolutional neural networks (CNN) and Recurrent neural networks (RNN). Unseen test input data are investigated to check the generalizability of the model. Also, the effect of residual image input on accuracy of the model is analyzed. Results are presented for both, with and without transfer learning to validate the effectiveness of transfer learning.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
1,950 members
Mahesh Shirole
  • Computer Engineering and Information Technology
Sandeep Sambhaji Udmale
  • Computer Engineering and Information Technology
Pramila M. Chawan
  • Computer Engineering and Information Technology
Surendra Bhosale
  • Department of Electrical Engineering
Jaypal Baviskar
  • Department of Electrical Engineering
H R Mahajani Marg, Matunga, 400019, Mumbai, Maharashtra, India
Head of institution
Prof. Dhiren Patel