Kantipur Engineering College

Two new ferrocenyl substituted triphenylamine based donor–acceptor dyes <bold>D1</bold> and <bold>D2</bold> were synthesized and used as sensitizers for dye sensitized solar cells (DSSCs).

Device A<bold>VC117</bold>:PC₇₁BM (THF cast), device B<bold>VC117</bold>:PC₇₁BM (pyridine/THF) and device C<bold>VC117</bold>:PC₇₁BM (pyridine/THF) with thermal annealing.

A conjugated acceptor-donor-acceptor (A-π-D-π-A) with the Zn-porphyrin core and the di-cyanovinyl substituted thiophene (A) connected at meso positions denoted as was designed and synthesized. The optical and electrochemical properties of were investigated. This new porphyrin exhibits a broad and intense absorption in the visible and near infrared regions. Bulk-heterojunction (BHJ) solution processed organic solar cells based on this porphyrin, as electron donor material, and PC71BM ([6,6]-phenyl C71 butyric acid methyl ester), as electron acceptor material, were fabricated using THF and a pyridine-THF solvent exhibiting a power conversion efficiency of 3.65% and 5.24%, respectively. The difference in efficiencies is due to the enhancement of the short circuit current J(sc) and FF of the solar cell, which is ascribed to a stronger and broader incident photon to current efficiency (IPCE) response and a better balanced charge transport in the device processed with the pyridine-THF solvent.

A novel BODIPY-porphyrin triad is prepared with two BODIPY molecules covalently attached <italic>via</italic> a 1,3,5-triazine molecule to a free-base carboxyphenyl <italic>meso</italic>-substituted porphyrin.

The global use of solar photovoltaic system is accelerating rapidly due to the ever-decreasing cost and improvement on cell efficiency. The variation in solar irradiance, weather condition and temperature cause the non-linear output V-I characteristics of solar cell. To get maximum power from the solar array instantaneously, there is a need of an efficient maximum power point tracker. This study has implemented the incremental conductance MPPT algorithm in hardware and compared the result with software model using MATLAB/SIMULINK. This algorithm automatically changes duty cycle of the converter connected to the solar panel, so that the panel voltage becomes equal to the voltage equivalent to maximum power point at any level of insolation, temperature or load conditions. It is found that the software modeled MPPT algorithm increased the PV output by 38 % while it was 33% in hardware implementation. The prioritizer embedded in the system optimizes the use of solar power rather than utility grid.

The gravitational water vortex power plant (GWVPP) is a new type of low head turbine system in which a channel and basin structure is used to form a vortex, where the rotational energy from the water can be extracted through a runner. This study is focused on the optimization of the runner to improve the efficiency of theGWVPP. Computational fluid dynamics (CFD) analysis iscarried out on three different runner designs with straight,twisted and curved blade profiles. ANSYS CFX was used to analyze the fluid flow through the channel, basin,turbine hub and blade, and results were used to evaluatethe efficiency of each of the runner designs. The CFDanalysis showed curved blade profile to be the mostefficient profile, with a peak efficiency of 82%, comparedto 46% for the straight blade runner and 63% for thetwisted blade version. An experimental test of the turbinesystem was carried out to validate the runner analysis, in ascale version of the GWVPP. The testing showed that therunner behaved as predicted from the CFD analysis, andhad a peak efficiency point of 71% at 0.5m head.

Porphyrin dyads, covalently linked by 1,3,5-triazine, either with one or two carboxylic acid groups, were synthesized and used as sensitizers in DSSCs, resulting in power conversion efficiency higher than 5%.

Devices based on <bold>BDT(CDTRH)2</bold>:PC₇₁BM processed from CF and DIO (3% v)/CF exhibit PCE of 4.58% and 6.07%.

We report the synthesis, characterization, and optical and electrochemical of properties of two novel molecules DRT3-BDT (1) and DTT3-BDT (2), comprising the same BDT central core (donor) and different end capped acceptor units, i.e. rhodanine with ethyl hexyl and thiazolidione with ethylhexyl, respectively, linked via an alkyl-substituted terthiophene (3 T) π-conjugation bridge. The electrochemical properties of these small molecules indicate that their energy levels are compatible with energy levels of PC71BM for efficient exciton dissociation. These molecules have been used as electron donors along with PC71BM as an electron acceptor, for the fabrication of solution processed "small molecule" bulk heterojunction (BHJ) solar cells (smOPV). The device prepared from optimized 1:PC71BM(1:1) processed cast from DIO (3%v)/CF solvent exhibited a power conversion efficiency of 6.76% with Jsc = 11.92 mA cm(-2), Voc = 0.90 and FF = 0.63. The device with 2:PC71BM under the same conditions showed a lower PCE of 5.25% with Jsc = 10.52 mA cm(-2), Voc = 0.86 and FF = 0.56. The AFM, TEM and PL quenching measurements revealed that the high Jsc is a result of the appropriate morphology and exciton dissociation. The performances were compared for the devices based on two small molecules. The higher Jsc for device 1 was attributed to its better nanoscale phase separation, smooth surface and higher carrier mobility in the 1:PC71BM blend film. Moreover, the higher value of FF for the 1:PC71BM based device was ascribed to a good balance between the electron and hole mobilities.

Stone mastic asphalt (SMA) mix amended with sugarcane bagasse fibers (SBFs) as a stabilizer was investigated experimentally in the present study. The SBF content in the combination mix was varied (0, 0.15%, 0.30%, 0.45% and 0.60%) to produce different mixtures. The prepared mixtures were then assessed for different properties, such as the Marshall stability, tensile strength, and draindown, which are crucial for the performance of the mix in flexible/bituminous pavements. The experimental results revealed enhanced stability and flow value, greater filling of voids in the bitumen, and reduced air voids in the prepared SMA mix amended with SBF. In addition, improvement in the tensile strength of the SMA mix amended with SBF was observed. An increase in the tensile strength ratio of the amended SMA mix was also observed. The susceptibility to the entry of moisture into the pores of the SMA mix was assessed by performing the draindown test, which revealed that the amended SMA mix exhibited reduced draindown due to the presence of SBF. Therefore, it was inferred that the inclusion of SBF into the SMA mix was beneficial for enhancing the stability and tensile strength of the mix and achieving higher resistance to moisture penetration. The present study demonstrated that the use of sugarcane bagasse as an active ingredient in the SMA mix would result in improved performance of flexible pavements while also reducing the waste burden of landfills and aiding in effective waste management.

Signature has been one of the widely used verification biometrics out there. Handwritten signatures are used in cheques, forms, letters, applications, minutes, etc. The Signature of every individual is unique in nature, that is why it is essential that a person’s handwritten signature be uniquely identified. Signature Verification is a widely used method for authenticating any individual during absence. Human verification is prone to inaccuracy and sometimes indecisiveness. This paper presents an investigation of using Convolutional Neural Network (CNN) for Writer-Dependent models in signature verification. Random distortions were generated in genuine images using an autoencoder to get forged signatures, which were passed to the classifier during training. The paper details all the pre-processing steps carried out on the image and shows various test results for changing the number of training sets of images. The average test accuracy for Persian dataset is 83% when the system was trained with 22 genuine images. There was a decrease of 9.4% in accuracy when the model was trained with 9 genuine images.

This paper presents a microstrip patch antenna consisting of a rectangular patch with four circular slots incorporated into patch for C-Band Satellite Communication. The proposed antenna achieves an impedance bandwidth of 4.7 GHz (3.6 GHz to 8.3 GHz). Maximum reflection coefficient is achieved at 4.4 GHz i.e. -32 dB. Good reflection coefficient and radiation pattern characteristics are obtained in the frequency band of interest. The antenna has a planar rectangular geometry with microstrip line feed and symmetrical circular slots. The proposed antenna is designed on low cost FR-4 substrate. The antenna is designed and simulated using High Frequency Structural Simulator (HFSS). The several factors affecting the bandwidth and reflection coefficient of the microstrip antenna such as length of ground plane, substrate and radiating patch dimensions are discussed in the paper.

Agriculture, as a fundamental aspect of human existence, faces challenges in crop selection, impacting resource allocation and productivity. This project addresses these challenges by proposing a stable system employing a soft voting classifier ensemble method. The ensemble comprises Naive Bayes, Support Vector Machine (SVM), Decision Tree, and Random Forest classifiers, offering personalized crop recommendations. Feasibility analysis encompasses technical, operational, economic, and scheduling aspects, ensuring practicality and efficacy. Development follows an incremental model, emphasizing continuous enhancement through feedback. Results indicate accuracies for individual classifiers (’Decision Tree’: 98.38%, ’Random Forest’: 98.90%, ’Naive Bayes’: 98.14%, ’SVM’: 98.50%), with an ensemble accuracy of 98.99%. Cross-validation confirms robustness. Evaluation metrics such as recall, precision, and F1 score demonstrate that the soft voting ensemble outperforms individual classifiers, highlighting its effectiveness in optimizing crop selection processes in agriculture and facilitating improved resource management and productivity.

Road Traffic Congestion Mapping is a system that enables people find information about the traffic congestion in the city. One of the major problems in the city centers like Kathmandu is the traffic jam in the roads. This project provides a feasible solution to the users in finding less congested path on the road to their destination. The system collects traffic congestion data from the roads and makes it available to users via Openstreet map. The surveillance cameras installed at the roads give continuous input to our system which then counts the number of vehicles in the road in a span of time to determine the congestion in the road. The system implements Background subtraction and thresholding for detection of vehicles from the image input received from the cameras. The congestion is plotted in Openstreet map, for example red line for highly congested road, blue line for mildly congested road and green line for free flow of vehicles in the road. Once this information is obtained, one can easily find the alternate path to their destination.

In this paper, various multiple input and multiple output (MIMO) detection techniques have been compared on the basis of BER performance and complexity. Maximum likelyhood (ML) detection method has shown optimal solution in MIMO systems compared to conventional detection techniques. However, higher receiver complexity leads to use of lower complexity techniques such as zero forcing (ZF) and minimum mean square error (MMSE) having relatively poor performance. Successive interference cancellation combined with ZF or MMSE has much improved performance but vulnerable to error propagation. Ordered succesive interference cancellation with MMSE (MMSE-OSIC) has reduced error propagation probability and gave the better performance. A new detection technique sphere detection (SD) based on Schnorr-Euchner enumeration has provided ML solution with much less computational complexity. For simulation, Rayleigh channel model has been considered.

In this paper the three new narrow bandgap D–A conjugated copolymers P1, P2 and P3 based on different weak donor fused thiophene-imidazole containing derivatives and the same benzothiadiazole acceptor unit were synthesized by Stille cross-coupling polymerization and characterized by 1H NMR, elemental analysis, GPC, TGA, DSC. These copolymers exhibit intensive absorbance in the range 350–900 nm and the optical bandgap lies in the range of 1.50–1.61 eV, which corresponds to the maximum photon flux of the solar spectrum. The electrochemical bandgap derived from cyclic voltammetry varies within the limits 1.47–1.65 eV and is approximately very close to the optical bandgap. The highest occupied molecular orbital (HOMO) energy level of all copolymers is deep lying (−5.24 eV and −5.37 eV and −5.25 eV for P1, P2 and P2, respectively) which shows that copolymers have good stability in the air and assured a higher open circuit voltage (Voc) for polymer BHJ solar cells. These copolymers were used as donors along with PC71BM and the BHJ polymer solar cells based on P1:PC71BM, P2:PC71BM and P3:PC71BM processed from chloroform (CF) solvent with 3 v% DIO as an additive showed an overall PCE of 4.55%, 6.76% and 5.16%, respectively.

Sister Robots - ‘Food and Medicine Carrier Robot’ are manually controlled robots which can be used in Corona Testing Hospital where health representatives or medical professionals need to carry food and medicine from the kitchen of the hospital to the patient isolation room. Robotics Association of Nepal [RAN] along with Team Robonauts, Team Kantibir and Team Oztec have developed the different versions of Sister Robots. The first version of the Robot - Sister V.1 is now working at the provincial hospital of Province no. 3, Hetauda Hospital. The Robot was built in support of Nepal Engineers’ Association [NEA] and handed over to Honorable Chief Minister, Province No.3 Dormani Poudel. Sister V.1 can help frontline health professionals communicate with COVID-19 suspected/infected patients in the isolation room, deliver essential foods and medicines as well. Sister V.1 took 20 days for its completion. The development team further developed second version of Sister Robot [Sister V.2]. First piece of Sister V.2 has been completed and 5 more are on the verge of development. Armed Police Force [APF] Hospital has made a trial of Sister V.2. Sister V.2 has been upgraded with a double 12V 8000MAH battery system for higher power performance, dual FPV camera system for Top 90 degree rotation back view and Front 180 degree bottom view. Food and Medicine can be easily delivered and COVID-19 can sanitize their hands with automatic sanitizer as well. Phase 1 product development has been completed and Phase 2 small scale production is currently running. The phases are defined based on the upgrade of the Robots as the team believes in iteration of the robot and customization as well. Mass scale production of Sister Robots on phase 3 Version 3 [Sister V.3] will consist of different lines of Robots with different specifications. More cost effective technologies will be introduced and automation will be a priority of Robot development. Research Center for Applied Science and Technology (RECAST) and Sagarmatha Engineering College collaborated with RAN for 6 more robots to develop financially. Phase 3 focuses in bringing Sister V.3 at each and every hospital in Nepal through collaborative partnerships and safety campaigns.

Abstract In today's cybersphere, cryptography plays a vital role in various fields. Image encryption is an integral part for securing information because of its vast application areas such as military (defense), multimedia, healthcare and so forth. In this article, an image encryption algorithm for both grayscale and color image is proposed based on Tangential Delay‐Ellipse Reflecting Curve System (TD‐ERCS) chaotic map system and deoxyribonucleic acid (DNA) coding. Chaotic map is used to scramble the pixel positions; to achieve confusion and for creation of mask image, and DNA coding is used for changing the pixel values; to achieve diffusion. Upon experimental analysis, proposed work achieved significantly high mean square error and low peak signal to noise ratio, almost zero correlation, high number of pixel change rate and unified averaged changed intensity values, and resistance to noise and data loss attacks. In addition, the decryption is possible without loss in quality of image.

A power conversion efficiency of about 5.26% has been achieved for the device based on<bold>P1</bold>:PC71BM.

The increasing sophistication of deepfake technologies has raised significant concerns regarding the authenticity of audio files. To address this challenge, we propose a deepfake audio detection system that employs a CNN-BiLSTM hybrid model for identifying synthetic speech. The system processes audio files by converting them into Mel-Spectrograms, which effectively capture the unique spectral features distinguishing real voices from fake ones. The processed data is then fed into the CNN-BiLSTM model, which leverages the strengths of Convolutional Neural Networks (CNNs) for spatial pattern recognition and Bidirectional Long Short-Term Memory Networks (BiLSTMs) for capturing long-term dependencies in the temporal sequence of the audio data. The model, trained on dataset, achieves an accuracy of 95%, effectively detecting subtle irregularities indicative of deepfake audio. The system provides users with a comprehensive analysis, including a confidence score and detailed insights into the authenticity of the audio, offering an effective tool for distinguishing real from fake audio. Our system combines cutting-edge machine learning technology with a user-friendly interface, making it both highly effective and accessible for practical applications in deepfake detection. Unlike existing systems that rely solely on either CNN or RNN architectures, our approach integrates both to enhance detection accuracy, particularly for complex and subtle manipulations. Additionally, we introduce a confidence scoring mechanism and insight analysis that provides users with transparent reasoning behind each detection.