Sharif Digital Repository

Broadband low-noise amplifiers (LNAs) find widespread application in many communication and measurement systems, and extensive research efforts have been devoted to improve their performances. The main efforts include the system bandwidth enhancement for increasing the data rate, the system noise reduction for improving sensitivity of receivers, and lowering the power consumption for increasing battery life of handheld systems. There is a steady quest to develop novel circuit architectures and design methodologies that can pave the way for performance enhancement of overall system.In this thesis, the design and implementation of a broadband low-noise amplifier circuit is investigated. The... 

The goal of this master thesis is to develop a new accurate numerical algorithm for digital differential relay to protect a power transformer. This work employs specially developed software to simplify its input signal processing task which helps to quickly and accurately detect and clear the transformer internal faults. The available commercial transformer digital differential relays lack the required sensitivity to detect turn to turn faults when a few numbers of turns (less than 10% of the winding) are involved. This work presents a new equivalent circuit for transformer when a turn to turn fault occurs on either winding. This equivalent circuit helped to develop an accurate algorithm for... 

The work starts with comparison of 3D-wound core transformers against the conventional transformers using the published research results. The past works on transformer short circuit force calculation are reviewed. An algorithm is developed to design 3D-wound core distribution transformers. To determine the maximum short circuit forces, currents in different type of short circuits have been calculated. The worst case in different core configurations and different type of winding connections is determined. Different analytical methods are investigated for transformer short circuit force calculation. Two-dimensional and three-dimensioal finite element methods (using Comsol software) employed to... 

This thesis is concerned with the performance of heavy equipments like electric substation transformers and bushings under earthquake loading. It begins with the description of the transformers, attached bushings, and some typical examples of substation failures in earthquake loading. This is followed by the field test and development of numerical models of transformer and two different types of seismic isolations, Natural rubber bearings (NRB) and high damping rubber bearings (HDRB) and in order to find the appropriate isolator, responses of fixed base and isolated transformers are compared. Important observations were made on the seismic responses of the transformer and its bushing.... 

Nowadays, non-linear loads comprise an important part of electrical grids. One of the most essential sources of nonlinear loads is switching power supplies, which mainly include power electronic rectifiers. When power outage occurs, Uninterruptible Power Supplies (UPS) must be able to supply these loads. The nonlinearity of the load has a significant effect on the Total Harmonic Distortion (THD) of the output voltage, which may not be easily reduced, especially in three-phase inverters. Furthermore, the presence of isolating transformers in these power supplies limits the control of harmonics and intensifies the nonlinear effects of the system.In this Master Thesis, a disturbance observer is... 

One of the most critical equipments in power grids are power transformers, which are vulnerable to damage due to the possibility of faults in its various components. A very common category of such faults is turn-to-turn fault in its winding. This fault initiates by the aging of the winding insulation with a short circuit of several adjacent loops to each other and if it is not detected quickly by the transformer protection, it will spread rapidly and can lead to irreparable failure in the transformer. Traditionally, turn-to-turn fault should be detected by mechanical pressure relays, however, due to the poor performance of these relays in the rapid detection of this type of fault, in the... 

Recent approaches in the field of semantic image segmentation rely on deep networks that are trained by pixel-level labels. This level of labeling requires a lot of time for the labeler person; because these networks require large training datasets to achieve optimal accuracy and the lack of data at the labeled pixel level causes a significant drop in their performance. In order to overcome this problem, weakly supervised segmentation approaches have been proposed. In these approaches, weaker labels such as image-level labels, bounding boxes, scribbles, etc. have been introduced to train the networks.In this thesis, a method for segmentation of kidney and kidney tumor in CT scan images based... 

Over the past two decades, the interest in exploring consumers' behavior has risen to popularity. Nowadays, it is necessary for marketers to reach out to the right consumers at the right time and with the right message. Even though the positive impact of tracking consumers throughout their purchase journey and its managerial implications have been emphasized several times by academics, There is still a lack of practical research on how to discern the consumers' stage in the purchase journey from a holistic point of view. In this study, we develop several machine learning models, including RNNs (GRU and LSTM), Transformers, and XGBoost, by utilizing historical data of Yektanet (the leading... 

Sample efficiency and systematic generalization are two long-standing challenges in sequential decision-making problems, especially, in reinforcement learning settings. It is hypothesized that involving natural language in conjunction with other observation modalities in decision-making environments can improve generalization due to its compositional and open-ended nature, and sample efficiency due to the concise information summarized in relatively short linguistic units. By exploiting this information and the compositional structure of the language, one can achieve an abstract and factored understanding of the environment and the task at hand. To do so, it is necessary to find the proper... 

The purpose of ranking Persian web pages is to retrieve the highest number of relevant documents related to the search query of Persian language users, with the minimum number of documents returned from the web. Information retrieval is one of the key issues in search engines. In this study, billions of documents were collected from Persian web pages, and due to infrastructure limitations, a few hundred million documents were indexed in a database like Elastic. Now, considering the user’s actual query, relevant documents should be retrieved from the indexed documents. To achieve this goal, a large Persian language model was needed. Existing large language models for Persian were not usable... 

Transformers are vital components of power networks, responsible for converting electricity from one voltage level to another. They are designed to operate under sinusoidal conditions, but the increasing use of non-linear loads and power electronic devices has caused a spike in harmonic pollution in power networks. These harmonics result in increased losses in transformers, leading to a reduction in capacity, accelerated aging, and a shortened life. The associated costs incurred by these issues include reduced energy sales for electricity companies and premature replacement costs. In this thesis, it is assumed that the transformer is affected by harmonics and ambient temperature, and other... 

Existing face forgery detection methods mainly focus on certain features of images, such as features related to image noise, local textures or frequency statistics of images for forgery detection. This makes the extracted representations and the final decision depend on the data in the database and makes it difficult to detect forgery with unknown manipulation methods. Solving this challenge, which is called the generalization challenge in artificial intelligence literature, has become the main goal of researchers in this field. In this thesis, the focus is on extracting effective features for success in forgery detection and preventing the performance of the forgery detection network from... 

Today, the production of colloquial texts in messengers, search engines, and question and answer systems has increased significantly, while text documents in other fields have a formal tone and style. Thus, there is a need for a system to convert these texts from colloquial form to the formal style. Attention to this need in non-Persian languages has also been recently and seriously felt, but almost at the time of writing, an efficient system has not been offered, and this issue requires more work in Persian than in languages such as English. In general, transferring texts from one form to another falls into the category of natural language processing applications and is called "style...