Visvesvaraya National Institute of Technology

Compressive Sensing (CS) is a new sensing modality, which compresses the signal being acquired at the time of sensing. Signals can have sparse or compressible representation either in original domain or in some transform domain. Relying on the sparsity of the signals, CS allows us to sample the signal at a rate much below the Nyquist sampling rate. Also, the varied reconstruction algorithms of CS can faithfully reconstruct the original signal back from fewer compressive measurements. This fact has stimulated research interest toward the use of CS in several fields, such as magnetic resonance imaging, high-speed video acquisition, and ultrawideband communication. This paper reviews the basic theoretical concepts underlying CS. To bridge the gap between theory and practicality of CS, different CS acquisition strategies and reconstruction approaches are elaborated systematically in this paper. The major application areas where CS is currently being used are reviewed here. This paper also highlights some of the challenges and research directions in this field.

A World Health Organization (WHO) Feb 2018 report has recently shown that mortality rate due to brain or central nervous system (CNS) cancer is the highest in the Asian continent. It is of critical importance that cancer be detected earlier so that many of these lives can be saved. Cancer grading is an important aspect for targeted therapy. As cancer diagnosis is highly invasive, time consuming and expensive, there is an immediate requirement to develop a non-invasive, cost-effective and efficient tools for brain cancer characterization and grade estimation. Brain scans using magnetic resonance imaging (MRI), computed tomography (CT), as well as other imaging modalities, are fast and safer methods for tumor detection. In this paper, we tried to summarize the pathophysiology of brain cancer, imaging modalities of brain cancer and automatic computer assisted methods for brain cancer characterization in a machine and deep learning paradigm. Another objective of this paper is to find the current issues in existing engineering methods and also project a future paradigm. Further, we have highlighted the relationship between brain cancer and other brain disorders like stroke, Alzheimer's, Parkinson's, and Wilson's disease, leukoriaosis, and other neurological disorders in the context of machine learning and the deep learning paradigm.

In this paper, a simplified nearest level control balancing method for modular multilevel converter is presented. The proposed method neither requires individual sorting of the submodule voltages nor the redundancy of the switching states. Once the sorting of the submodules is done on the basis of the number of the submodules to be switched on, the identifications of the submodules can be carried out throughout the stages of the implementation of the this method. The proposed method also does not require the individual submodule status in the gate pulse generation stage. The gate logic in the presented method can be implemented with the help of the switching states of the voltage levels. Those simplifications and removing of the some of the stages by the proposed balancing method may ease and lead to the less processor time at the implementation level. The pictorial presentation further helps in consolidating the understanding of the different stages of the method. Rigorous simulations are carried out for open and one of the prominent closed-loop applications, i.e., modular multilevel converter-based high-voltage direct current to demonstrate the validity and effectiveness of the proposed simplified balancing method under normal and emergency conditions.

The time of operation of overcurrent relays (OCRs) can be reduced, and at the same time, the coordination can be maintained, by selecting the optimum values of time multiplier setting (TMS) and plug setting (PS) of OCRs. This paper presents hybrid genetic algorithm (GA) - nonlinear programming (NLP) approach for determination of optimum values of TMS and PS of OCRs. GA has a drawback of, sometimes, converging to the values which may not be optimum, and NLP methods have a drawback of converging to local optimum values, if the initial choice is nearer to local optimum. This paper proposes a hybrid method to overcome the drawback of GA and NLP method, and determine the optimum settings of OCRs. The main contributions of this paper are - 1) systematic method for formulation of problem of determining optimum values of TMS and PS of OCRs in power distribution network as a constrained nonlinear optimization problem, 2) determining initial values of TMS and PS using GA technique and finding final (global optimum) values using NLP method, thus making use of the advantages of both methods (and at the same time overcoming the drawbacks of the methods).

The choice of activating agent for the thermochemical production of high-grade activated carbon (AC) from agricultural residues and wastes, such as feedstock, requires innovative methods. Overcoming energy losses, and using the best techniques to minimise secondary contamination and improve adsorptivity, are critical. Here, we review the importance and influence of activating agents on agricultural waste: how they react and compare conventional and microwave processes. In particular, adsorbent pore characteristics, surface chemistry interactions and production modes were compared with traditional methods. It was concluded that there are no best activating agents; rather, each agent reacts uniquely with a precursor, and the optimum choice depends on the target adsorbent. Natural chemicals can also be as effective as inorganic activating agents, and offer the advantages that they are usually safe, and readily available. The use of a microwave, as an innovative pyrolysis approach, can enhance the activation process within a duration of 1–4 h and temperature of 500–1200 °C, after which the yield and efficiency decline rapidly due to molecular breakdown. This study also examines the biomass milling process requirements; the influence of the dielectric properties, along with the effect of washing; and experimental setup challenges. The microwave setup system, biomass feed rate, product delivery, inert gas flow rate, reactor design and recovery lines are all important factors in the microwave activation process, and contribute to the overall efficiency of AC preparation. However, a major issue is a lack of large-scale industrial demonstration units for microwave technology.

Pneumonia causes the death of around 700,000 children every year and affects 7% of the global population. Chest X-rays are primarily used for the diagnosis of this disease. However, even for a trained radiologist, it is a challenging task to examine chest X-rays. There is a need to improve the diagnosis accuracy. In this work, an efficient model for the detection of pneumonia trained on digital chest X-ray images is proposed, which could aid the radiologists in their decision making process. A novel approach based on a weighted classifier is introduced, which combines the weighted predictions from the state-of-the-art deep learning models such as ResNet18, Xception, InceptionV3, DenseNet121, and MobileNetV3 in an optimal way. This approach is a supervised learning approach in which the network predicts the result based on the quality of the dataset used. Transfer learning is used to fine-tune the deep learning models to obtain higher training and validation accuracy. Partial data augmentation techniques are employed to increase the training dataset in a balanced way. The proposed weighted classifier is able to outperform all the individual models. Finally, the model is evaluated, not only in terms of test accuracy, but also in the AUC score. The final proposed weighted classifier model is able to achieve a test accuracy of 98.43% and an AUC score of 99.76 on the unseen data from the Guangzhou Women and Children's Medical Center pneumonia dataset. Hence, the proposed model can be used for a quick diagnosis of pneumonia and can aid the radiologists in the diagnosis process.

Injection of the wind power into an electric grid affects the power quality. The performance of the wind turbine and thereby power quality are determined on the basis of measurements and the norms followed according to the guideline specified in International Electro-technical Commission standard, IEC-61400. The influence of the wind turbine in the grid system concerning the power quality measurements are-the active power, reactive power, variation of voltage, flicker, harmonics, and electrical behavior of switching operation and these are measured according to national/international guidelines. The paper study demonstrates the power quality problem due to installation of wind turbine with the grid. In this proposed scheme STATic COMpensator (STATCOM) is connected at a point of common coupling with a battery energy storage system (BESS) to mitigate the power quality issues. The battery energy storage is integrated to sustain the real power source under fluctuating wind power. The STATCOM control scheme for the grid connected wind energy generation system for power quality improvement is simulated using MATLAB/SIMULINK in power system block set. The effectiveness of the proposed scheme relives the main supply source from the reactive power demand of the load and the induction generator. The development of the grid co-ordination rule and the scheme for improvement in power quality norms as per IEC-standard on the grid has been presented.

Globally, India is the leading producer of fruits. Fruits after consumption leave a peel which is a nuisance to the environment as a solid waste. In this article, commonly available large volume-fruit peels (FP) (viz. banana, orange, citrus, lemon and jackfruit) were investigated for surface, physical and chemical characteristics with a view to propose their valorization in detail. Each FP was characterized by proximate and ultimate analysis, porosity, particle density, bulk density, point of zero charge (pH pzc ), surface pH, surface charges, water absorption capacity, BET surface area, scanning electron microscopy, Fourier transform infrared spectroscopy and TGA/derivative of thermogravimetric. The BET surface area of FP is very less, between 0.60 and 1.2 m 2 /g. The pH pzc and surface pH values of orange peel (OP), citrus peel (CP), lemon peel (LP) and jackfruit peels (JFP) are in the range of 3-4. The pH pzc value and surface pH of banana peel (BP) is closer to 7. The order of surface acidity is OP > LP > CP > JFP > BP. From TG curves it is clear that FPs are stable below 150C. The results will be useful for rational design, when FP is used as a substrate for bioactive compounds, phenolic antioxidants, organic acids, enzymes, biofertilizer, production of energy and as adsorbents.

Depression is a leading cause of mental ill health, which has been found to increase risk of early death. Moreover it is a major cause of suicidal ideation and leads to significant impairment in daily life. Emotion artificial intelligence is a field of ongoing research in emotion detection, specifically in the field of text mining. The advent of internet based media sources has resulted in significant user data being available for sentiment analysis of text and images. This paper aims to apply natural language processing on Twitter feeds for conducting emotion analysis focusing on depression. Individual tweets are classified as neutral or negative, based on a curated word-list to detect depression tendencies. In the process of class prediction, support vector machine and Naive-Bayes classifier have been used. The results have been presented using the primary classification metrics including F1-score, accuracy and confusion matrix.

The positive features of neural networks and fuzzy logic are combined together for the detection of stator inter-turn insulation and bearing wear faults in single-phase induction motor. The adaptive neural fuzzy inference systems (ANFISs) are developed for the detection of these two faults. These faults are created experimentally on a single-phase induction motor in the laboratory. The experimental data is generated for the five measurable parameters, viz, motor intakes current, speed, winding temperature, bearing temperature, and the noise of the machine. Earlier, the ANFIS fault detectors are trained for the two input parameters, i.e., speed and current, and the performance is tested. Later, the three remaining parameters are added and the five input ANFIS fault detector is trained and tested. It observed from the simulation results that the five input parameter system predicts more accurate results

Human action recognition in video analytics has been widely studied in recent years. Yet, most of these methods assign a single action label to video after either analyzing a complete video or using classifier for each frame. But when compared to human vision strategy, it can be deduced that we (human) require just an instance of visual data for recognition of scene. It turns out that small group of frames or even single frame from the video are enough for precise recognition. In this paper, we present an approach to detect, localize and recognize actions of interest in almost real-time from frames obtained by a continuous stream of video data that can be captured from a surveillance camera. The model takes input frames after a specified period and is able to give action label based on a single frame. Combining results over specific time we predicted the action label for the stream of video. We demonstrate that YOLO is effective method and comparatively fast for recognition and localization in Liris Human Activities dataset.

The High power induction machines are designed at medium voltage (MV) rating for better performance. The multilevel inverters (MLI) are able to provide medium voltage with high quality output at low switching frequency as compared to conventional two-level inverter. In addition to this, MLI reduces dv/dt, switching losses and leakage current. In this paper, approaches to reduce and eliminate the common mode voltage (CMV) using five- level diode clamped multilevel inverter (DCMLI) are presented. The CMV spikes are also eliminated by shifting dead-time across the phase pole. A novel technique for the selection of switching states to synthesize the desire vector is proposed. This paper realizes the implementation of five-level diode clamped MLI for three phase induction motor. Experimental results demonstrate the feasibility of the proposed solution.

When a nodal demand is excessive as in a fire-flow condition or when a pump fails or a pipe breaks, a water distribution system (WDS) may temporarily become deficient and unable to satisfy all nodal demands. However, the prediction of the performance of a WDS under a temporarily-deficient condition is necessary for simulation-based reliability analysis and design of WDSs. Available methods for such prediction are reviewed herein. When the actual outlets are considered as demand nodes the methods which simultaneously consider the nodal flows and heads give fairly accurate and similar results. However, when the demands of secondary networks are assumed concentrated at the nodes of the primary WDSs, the prediction of the deficient-condition performance of a primary WDS is rather approximate. For reliability purposes, however, the method using parabolic head-discharge relationship (no flow at minimum head to required flow at desirable head), is the best for prediction of deficient-network performance.

Abstract A simple and scalable approach has been reported for V 2 O 5 encapsulation over interconnected multi-walled carbon nanotubes (MWCNTs) network using chemical bath deposition method. Chemically synthesized V 2 O 5 /MWCNTs electrode exhibited excellent charge-discharge capability with extraordinary cycling retention of 93% over 4000 cycles in liquid-electrolyte. Electrochemical investigations have been performed to evaluate the origin of capacitive behavior from dual contribution of surface-controlled and diffusion-controlled charge components. Furthermore, a complete flexible solid-state, flexible symmetric supercapacitor (FSS-SSC) device was assembled with V 2 O 5 /MWCNTs electrodes which yield remarkable values of specific power and energy densities along with enhanced cyclic stability over liquid configuration. As a practical demonstration, the constructed device was used to lit the ‘VNIT’ acronym assembled using 21 LED’s.

Leaf diseases on cotton plant must be identified early and accurately as it can prove detrimental to the yield. The proposed work presents a pattern recognition system for identification and classification of three cotton leaf diseases i.e. Bacterial Blight, Myrothecium and Alternaria. The images required for this work are captured from the fields at Central Institute of Cotton Research Nagpur, and the cotton fields in Buldana and Wardha district. Active contour model is used for image segmentation and Hu's moments are extracted as features for the training of adaptive neuro-fuzzy inference system. The classification accuracy is found to be 85 percent.

In this paper, we investigate the resource allocation problem for unmanned aerial vehicle (UAV)-assisted networks, where a UAV acting as an energy source provides radio frequency energy for multiple energy harvesting-powered device-to-device (D2D) pairs with much information to be transmitted. The goal is to maximize the average throughput within a time horizon while satisfying the energy causality constraint under a generalized harvest-transmit-store model, which results in a non-convex problem. By introducing the Lagrangian relaxation method, we analytically show that the behavior of all D2D pairs at each time slot is exclusive: harvesting energy or transmitting information signals. The formulated non-convex optimization problem is thus transformed into a mixed integer nonlinear programming (MINIP). We then design an efficient resource allocation algorithm to solve this MINIP, where D.C. (difference of two convex functions) programming and golden section method are combined to achieve a suboptimal solution. Furthermore, we provide an idea to reduce the computational complexity for facilitating the application in practice. Simulations are conducted to validate the effectiveness of the proposed algorithm and evaluate the system throughput performance.

This paper presents a reliable microgrid for residential community with modified control techniques to achieve enhanced operation during grid connected, islanded, and resynchronization mode. The proposed microgrid is a combination of solar photovoltaic, battery storage system and locally distributed generation (DG) systems with residential local loads. A modified power control technique is developed such that local load reactive power demand, harmonic currents, and load unbalance are compensated by respective residential local DG. However, active power demand of all local residential load is shared between the microgrid and respective local DG. This control technique also achieves constant active power loading on the microgrid by supporting additional active power local load demand of respective residential DG. Hence, proposed modified power control technique achieves transient free operation of the microgrid during residential load disturbances. An additional modified control technique is also developed to achieve seamless transition of microgrid between grid-connected mode and islanded mode. The dynamic performance of this microgrid during grid-connected, islanded, and resynchronization mode under linear and nonlinear load variations is verified using real-time simulator.

This paper presents a three-phase hybrid cascaded modular multilevel inverter topology which is derived from the proposed modified H-bridge module. This topology results in the reduction of number of power switches, losses, installation area, voltage stress and converter cost. For renewable energy environment such as photovoltaic (PV) connected to the microgrid system, it enables the tranformerless operation and enhances the power quality. This multilevel inverter is an effective and efficient power electronic interface strategy for renewable energy systems. The basic operation of single module and the proposed cascaded hybrid topology is explained. The ability to operate in both symmetrical and asymmetrical modes is analyzed. The comparative analysis is done with classical cascaded H-bridge and flying capacitor multilevel inverters. The nearest level control method is employed to generate the gating signals for the power semiconductor switches. To verify the applicability and performance of the proposed structure in PV renewable energy environment, simulation results are carried out by MATLAB/Simulink under both steady-state and dynamic conditions. Experimental results are presented to validate the simulation results.

The efficient and effective search for the optimum design solution of a water distribution network using genetic algorithms (GAs) is governed by several factors such as representation scheme, population size, hydraulic simulation model, fitness function, penalty method, GA operators, number of generations, and more importantly the size of the search space. This paper proposes a modified GA that uses basic operators along with their derivatives randomly. Further, a methodology based on critical path method is suggested to reduce the search space. A software tool, GA-WAT, based on the proposed methodology is developed and first tested and verified for its efficiency and effectiveness on two previously published single source networks. Later, it is applied to the optimal design of a larger, two-source hypothetical network. The results obtained indicate that the modified GA with reduction in search space proposed herein is more effective, especially for large practical networks.

It is a common practice around the world to stabilise black cotton soil using lime or cement to improve the strength of stabilised sub-base and subgrade soil. However, production of cement and lime is highly energy intensive. It is also reported to emit large quantity of CO2 into the atmosphere. Moreover, the global warming potential of fly ash (0.00526–0.027 kg CO2eq/kg) being a waste, is very low as compared to that of cement (0.82–0.948 kg CO2eq/kg) and lime (about 0.416 kg CO2eq/kg). Thus, in order to reduce the emission of greenhouse gases associated with lime and cement stabilisation, an experimental investigation is conducted to study the feasibility of using fly ash geopolymer to stabilise black cotton soil. The experimental investigation was carried out by varying fly ash content from 5% to 20% and treating the samples with a lower concentration of 5M NaOH solution. Tests were conducted in the laboratory to obtain the unconfined compressive strength and California bearing ratio and resilient modulus values of the stabilised samples. Moreover, microstructural analysis using X-ray diffraction, scanning electron microscope, energy dispersive spectroscopy and Fourier transformed infrared spectroscopy were conducted to see insight into the material behaviour. It is concluded from the present study that the fly ash-based geopolymer could be used for stabilisation of black cotton soil for highway subgrade and sub-base preparation.