Pinned Repositories
Doc2Edi
Extracting Data from Document PDF and Converting to EDI211 Files Using GCP and Google Document AI
LLM-GPT-langflow-
⛓️ LangFlow is a UI for LangChain, designed with react-flow to provide an effortless way to experiment and prototype flows.
MAS-Technique----Analytical-Optimization-of-Grid-Supports-by-Converters
supporting the grid voltage and proper operation of the grid-connected converters (GCCs) under a wide range of grid voltage conditions have become major requirements. An analytical study is very useful for evaluating the supporting capability of the available control strategies in GCCs. This paper analytically studies, then modifies the supporting capability of three existing strategies. The contribution of this paper is two-fold: first, analytical expressions of instantaneous active/reactive powers oscillation and maximum phase currents are formulated and used to conduct several comparisons among different strategies. Second, based on the obtained formulas for the maximum phase currents, maximum allowable support (MAS) control schemes are proposed under unbalanced voltage conditions. The MAS control schemes have two important objectives: obtaining maximum active or reactive power delivery and simultaneously respecting the maximum phase currents under the unbalanced condition. The proposed equations can further estimate the maximum depth of the faulted voltage where each strategy is still able to satisfy the voltage support requirements imposed by the grid codes. The proposed expressions can also help all techniques to provide their maximum voltage or frequency support under the pre-set maximum phase current limitations. Different selected simulation and experimental tests are carried out for comparing the strategies, and validating the effectiveness of the proposed MAS equations.
Multi-Objective-Optimization-of-Distributed-Energy-Systems-under-Grid-Faults
Recently, riding through grid faults and supporting the grid voltage by using grid-connected converters (GCCs) have become major requirements reflected in the grid codes. This paper presents a novel reference current generation scheme with the ability to support the grid voltage by injecting a proper set of positive/negative active/reactive currents by using four controlling parameters. Analytical expressions are proposed to obtain the optimal values of these parameters under any grid voltage condition. The optimal performances can be obtained by achieving the following objectives: first, compliance with the phase voltage limits, second, maximized active and reactive power delivery, third, minimized fault currents, and fourth reduced oscillations on the active and reactive powers. These optimal behaviors bring significant advantages to emerging GCCs, such as increasing the efficiency, lowering the dc-link ripples, improving ac system stability, and avoiding equipment tripping. Simulation and experimental results verify the analytical results and the proposed expressions.
My-Machine-Learning-Projects-and-Papers
These are the published papers of the projects that are based on different Machine Learning algorithms and Optimization methods.
New-Method-to-Optimize-the-placement-of-SSSCs-in-57-Bus-IEEE-Power-Test-System
The optimised allocation of FACTS devices is a complex problem in large-scale power systems. The optimal placements of FACTS devices can have a determinant effect on the performance of the entire electrical network. In this paper, the proper allocation of SSSCs is proposed. Sensitivity analysis is used in order to model SSSCs effects on the power system. The advantage of this modelling is to accelerate the optimization process. By using SA for modelling the SSSC, optimization needs only one load flow in the initial state. According to this method, there is no need for exact modelling of SSSCs. Due to its high speed, another advantage of this method is its ability to cover all possible solution areas and all candidate places for SSSCs. Additionally, the number of SSSCs is not fixed and it can vary during optimization which is very beneficial. This approach is based on the objective function including transient stability and active power transmission capacity indices. The proposed method is applied on a typical 6-bus network and the 57-bus IEEE test system. The results show the superior performance of this method.
Optimised-Wind-Energy-Response-to-Grid-Faults
Recently, supporting the grid voltage and proper operation of the grid-connected converters (GCCs) under a wide range of grid voltage conditions have become major requirements. An analytical study is very useful for evaluating the supporting capability of the available control strategies in GCCs. This paper analytically studies, then modifies the supporting capability of three existing strategies. The contribution of this paper is two-fold: first, analytical expressions of instantaneous active/reactive powers oscillation and maximum phase currents are formulated and used to conduct several comparisons among different strategies. Second, based on the obtained formulas for the maximum phase currents, maximum allowable support (MAS) control schemes are proposed under unbalanced voltage conditions. The MAS control schemes have two important objectives: obtaining maximum active or reactive power delivery and simultaneously respecting the maximum phase currents under the unbalanced condition. The proposed equations can further estimate the maximum depth of the faulted voltage where each strategy is still able to satisfy the voltage support requirements imposed by the grid codes. The proposed expressions can also help all techniques to provide their maximum voltage or frequency support under the pre-set maximum phase current limitations. Different selected simulation and experimental tests are carried out for comparing the strategies, and validating the effectiveness of the proposed MAS equations.
OPTIMIZATION-of-the-APFs-Placement-Based-on-Instantaneous-Reactive-Power-Theory-by-GENETIC-ALGORITHM
In electrical distribution systems, a great amount of power are wasting across the lines, also nowadays power factors, voltage profiles and total harmonic distortions (THDs) of most loads are not as would be desired. So these important parameters of a system play highly important role in wasting money and energy, and besides both consumers and sources are suffering from a high rate of distortions and even instabilities. Active power filters (APFs) are innovative ideas for solving of this adversity which have recently used instantaneous reactive power theory. In this paper, a novel method is proposed to optimize the allocation of APFs. The introduced method is based on the instantaneous reactive power theory in vectorial representation. By use of this representation, it is possible to asses different compensation strategies. Also, APFs proper placement in the system plays a crucial role in either reducing the losses costs and power quality improvement. To optimize the APFs placement, a new objective function has been defined on the basis of five terms: total losses, power factor, voltage profile, THD and cost. Genetic algorithm has been used to solve the optimization problem. The results of applying this method to a distribution network illustrate the method advantages.
Smart-Monitoring-of-Deformations-on-Transformers-using-Image-Processing-and-UWB-Transceivers
In this paper, a novel method based on image processing is proposed for detecting the presence and location of mechanical deformations on an actual power transformer winding. A vertical imaging setup is used to obtain a two-dimensional (2-D) image of the transformer winding using synthetic aperture radar (SAR) imaging method and Kirchhoff migration algorithm. The main goal of the image processing method is to detect the location of the radial deformation on the transformer high-voltage (HV) winding. Two different deformations are applied on the winding under the test in this paper. The first one is a modeled bulgymechanical deformation and the second one is an actual concave buckling made on the actual transformer HV winding. The experimental results for different cases show the effectiveness of the proposed method to detect the presence and location of the deformation on the transformer winding.
WeBell-App-Sentiment-Analysis
MLEngFlaskProject_candidate The purpose of this project is to test the candidate's ability to build a simple Flask application.
masoudshab's Repositories
masoudshab/Multi-Objective-Optimization-of-Distributed-Energy-Systems-under-Grid-Faults
Recently, riding through grid faults and supporting the grid voltage by using grid-connected converters (GCCs) have become major requirements reflected in the grid codes. This paper presents a novel reference current generation scheme with the ability to support the grid voltage by injecting a proper set of positive/negative active/reactive currents by using four controlling parameters. Analytical expressions are proposed to obtain the optimal values of these parameters under any grid voltage condition. The optimal performances can be obtained by achieving the following objectives: first, compliance with the phase voltage limits, second, maximized active and reactive power delivery, third, minimized fault currents, and fourth reduced oscillations on the active and reactive powers. These optimal behaviors bring significant advantages to emerging GCCs, such as increasing the efficiency, lowering the dc-link ripples, improving ac system stability, and avoiding equipment tripping. Simulation and experimental results verify the analytical results and the proposed expressions.
masoudshab/OPTIMIZATION-of-the-APFs-Placement-Based-on-Instantaneous-Reactive-Power-Theory-by-GENETIC-ALGORITHM
In electrical distribution systems, a great amount of power are wasting across the lines, also nowadays power factors, voltage profiles and total harmonic distortions (THDs) of most loads are not as would be desired. So these important parameters of a system play highly important role in wasting money and energy, and besides both consumers and sources are suffering from a high rate of distortions and even instabilities. Active power filters (APFs) are innovative ideas for solving of this adversity which have recently used instantaneous reactive power theory. In this paper, a novel method is proposed to optimize the allocation of APFs. The introduced method is based on the instantaneous reactive power theory in vectorial representation. By use of this representation, it is possible to asses different compensation strategies. Also, APFs proper placement in the system plays a crucial role in either reducing the losses costs and power quality improvement. To optimize the APFs placement, a new objective function has been defined on the basis of five terms: total losses, power factor, voltage profile, THD and cost. Genetic algorithm has been used to solve the optimization problem. The results of applying this method to a distribution network illustrate the method advantages.
masoudshab/New-Method-to-Optimize-the-placement-of-SSSCs-in-57-Bus-IEEE-Power-Test-System
The optimised allocation of FACTS devices is a complex problem in large-scale power systems. The optimal placements of FACTS devices can have a determinant effect on the performance of the entire electrical network. In this paper, the proper allocation of SSSCs is proposed. Sensitivity analysis is used in order to model SSSCs effects on the power system. The advantage of this modelling is to accelerate the optimization process. By using SA for modelling the SSSC, optimization needs only one load flow in the initial state. According to this method, there is no need for exact modelling of SSSCs. Due to its high speed, another advantage of this method is its ability to cover all possible solution areas and all candidate places for SSSCs. Additionally, the number of SSSCs is not fixed and it can vary during optimization which is very beneficial. This approach is based on the objective function including transient stability and active power transmission capacity indices. The proposed method is applied on a typical 6-bus network and the 57-bus IEEE test system. The results show the superior performance of this method.
masoudshab/MAS-Technique----Analytical-Optimization-of-Grid-Supports-by-Converters
supporting the grid voltage and proper operation of the grid-connected converters (GCCs) under a wide range of grid voltage conditions have become major requirements. An analytical study is very useful for evaluating the supporting capability of the available control strategies in GCCs. This paper analytically studies, then modifies the supporting capability of three existing strategies. The contribution of this paper is two-fold: first, analytical expressions of instantaneous active/reactive powers oscillation and maximum phase currents are formulated and used to conduct several comparisons among different strategies. Second, based on the obtained formulas for the maximum phase currents, maximum allowable support (MAS) control schemes are proposed under unbalanced voltage conditions. The MAS control schemes have two important objectives: obtaining maximum active or reactive power delivery and simultaneously respecting the maximum phase currents under the unbalanced condition. The proposed equations can further estimate the maximum depth of the faulted voltage where each strategy is still able to satisfy the voltage support requirements imposed by the grid codes. The proposed expressions can also help all techniques to provide their maximum voltage or frequency support under the pre-set maximum phase current limitations. Different selected simulation and experimental tests are carried out for comparing the strategies, and validating the effectiveness of the proposed MAS equations.
masoudshab/Optimised-Wind-Energy-Response-to-Grid-Faults
Recently, supporting the grid voltage and proper operation of the grid-connected converters (GCCs) under a wide range of grid voltage conditions have become major requirements. An analytical study is very useful for evaluating the supporting capability of the available control strategies in GCCs. This paper analytically studies, then modifies the supporting capability of three existing strategies. The contribution of this paper is two-fold: first, analytical expressions of instantaneous active/reactive powers oscillation and maximum phase currents are formulated and used to conduct several comparisons among different strategies. Second, based on the obtained formulas for the maximum phase currents, maximum allowable support (MAS) control schemes are proposed under unbalanced voltage conditions. The MAS control schemes have two important objectives: obtaining maximum active or reactive power delivery and simultaneously respecting the maximum phase currents under the unbalanced condition. The proposed equations can further estimate the maximum depth of the faulted voltage where each strategy is still able to satisfy the voltage support requirements imposed by the grid codes. The proposed expressions can also help all techniques to provide their maximum voltage or frequency support under the pre-set maximum phase current limitations. Different selected simulation and experimental tests are carried out for comparing the strategies, and validating the effectiveness of the proposed MAS equations.
masoudshab/My-Machine-Learning-Projects-and-Papers
These are the published papers of the projects that are based on different Machine Learning algorithms and Optimization methods.
masoudshab/Smart-Monitoring-of-Deformations-on-Transformers-using-Image-Processing-and-UWB-Transceivers
In this paper, a novel method based on image processing is proposed for detecting the presence and location of mechanical deformations on an actual power transformer winding. A vertical imaging setup is used to obtain a two-dimensional (2-D) image of the transformer winding using synthetic aperture radar (SAR) imaging method and Kirchhoff migration algorithm. The main goal of the image processing method is to detect the location of the radial deformation on the transformer high-voltage (HV) winding. Two different deformations are applied on the winding under the test in this paper. The first one is a modeled bulgymechanical deformation and the second one is an actual concave buckling made on the actual transformer HV winding. The experimental results for different cases show the effectiveness of the proposed method to detect the presence and location of the deformation on the transformer winding.
masoudshab/Doc2Edi
Extracting Data from Document PDF and Converting to EDI211 Files Using GCP and Google Document AI
masoudshab/docker-compose-ghost-quickstart
:ghost: Ghost blogging platform quick start with docker compose
masoudshab/Optimal-allocation-of-FACTS-using-equivalent-impedance-models-of-VSCs
Flexible AC transmission system (FACTS) devices such as static synchronous series compensators, static synchronous compensators, and unified power flow controllers which use voltage source convertors (VSCs) can noticeably improve different characteristics of power systems. Therefore, finding their optimal allocation is a vital issue. To find the optimized placement of different FACTS devices, it is necessary to calculate the effects of such devices in the whole system. Sensitivity analysis has been recently proposed as a fast and reliable method to find the effects of voltage source convertors on system variables. In this paper, a novel approach is introduced to optimize the allocation of different FACTS devices based on the equivalent impedance model obtained from sensitivity analysis. This method requires neither the exact modeling of voltage source convertors nor more than one load flow, so it is much faster than conventional methods. Due to its simplicity and fastness, it can cover all possible locations, continuous sizing, type, and number of compensators. Therefore, its performance and accuracy are considerably high. The proposed method based on a novel objective function, which includes voltage profile, transmission capability, and stability, is applied to a typical 6-bus and the 30-bus IEEE test systems. Furthermore, genetic algorithm is implemented to solve the optimization problem. The test results illustrate the effectiveness of the introduced method. Copyright © 2014 John Wiley & Sons, Ltd.
masoudshab/Stability-Improv-in-Large-Scale-Power-Systems-based-on-Smart-Coefficient-Algorithm
Flexible AC transmission systems (FACTS) can be used to improve power system performance. These devices can improve system parameters; hence the maximum potential of the transmission system can be used. Unified power flow controller (UPFC) is one of the FACTS devices which can simultaneously control the bus voltage and real and reactive power flow in transmission systems. However, their excessive cost causes the optimal choice for the number and the location of these devices. This paper proposes a method for optimized UPFC allocation based on smart coefficients algorithm (SCA) to specify the location, number and input values by minimizing the voltage phase of system buses. The proposed SCA remarkably improves the accuracy and performance of traditional optimization processes in large scale networks. This new method is applied to the 118-bus IEEE standard system. The results of traditional and new optimization algorithm demonstrate the great improvement in optimization process using SCA.
masoudshab/Ladder-Network-Parameters-Determination-with-Supervised-Machine-Learning
In order to accurately model the transient behavior of transformer windings, we need models with a large number of nodes and parameters. This paper proposes amodel that has fewer nodes and can accurately predict the behavior of a transformer in a wide range of frequencies. In the proposed method, based on the terminal measurements, dominant resonances are determined, and it is experimentally shown that the winding has hidden resonances. Using this idea, we suggest the use of a section ladder network, which has a minimum number of nodes and can accuratelymodel the behavior of the transformerwinding. The parameters of this model are determined by minimizing the error function by using the genetic algorithm. The close agreement between the simulation and measurement results on the windings of a 20/0.4-kV and 1600-kVA transformer verifies the accuracy of the proposed method.
masoudshab/LLM-GPT-langflow-
⛓️ LangFlow is a UI for LangChain, designed with react-flow to provide an effortless way to experiment and prototype flows.
masoudshab/WeBell-App-Sentiment-Analysis
MLEngFlaskProject_candidate The purpose of this project is to test the candidate's ability to build a simple Flask application.
masoudshab/ChatGPT-LLM-haystack
:mag: Haystack is an open source NLP framework to interact with your data using Transformer models and LLMs (GPT-4, ChatGPT and alike). Haystack offers production-ready tools to quickly build complex question answering, semantic search, text generation applications, and more.
masoudshab/Deep_and_Machine_Learning_Projects
This Repository contains the list of various Machine and Deep Learning related projects. Related code and data files are available inside this folder. One can go through these projects to implement them in real life for specific use cases.
masoudshab/GPT-LLM-MOSS
An open-source tool-augmented conversational language model from Fudan University
masoudshab/GPT-picoGPT
An unnecessarily tiny implementation of GPT-2 in NumPy.
masoudshab/GPT-project--vall-e
PyTorch implementation of VALL-E(Zero-Shot Text-To-Speech), Reproduced Demo https://lifeiteng.github.io/valle/index.html
masoudshab/litellm
Call all LLM APIs using the OpenAI format. Use Azure, OpenAI, Cohere, Anthropic, Ollama, VLLM, Sagemaker, HuggingFace, Replicate (100+ LLMs)
masoudshab/machine-learning-engineering-for-production-public
Public repo for DeepLearning.AI MLEP Specialization
masoudshab/Machine-Learning-Ng
Arthur Samuel (1959). Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed. • Tom Mitchell (1998) Well-posed Learning Problem: A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.
masoudshab/Optimal-Location-and-Sizing-of-STATCOM
In this paper, a novel approach is introduced to optimise the placement of STATCOMs according to a new objective function including voltage profile of different system points (VP) and transmission capability indices. STATCOM can improve the power system parameters if they have been optimally allocated. Using equivalent impedance modelling of voltage source converters (VSCs) and sensitivity analysis in order to calculate the required indices help improve the speed of the calculations and the convergence rate of the optimization problem solving. Genetic Algorithm (GA) is used to solve the optimization problem. The proposed method is applied on a test system with 7 bus bars. The test results can clarify the effectiveness of the proposed method.
masoudshab/OSProj-digits-recognizer-kubeflow
Sample MLOps Workflow: Recognizing Digits with Kubeflow
masoudshab/Prac-git
This is a prac repo to stat working with git
masoudshab/Voltage-Improvment-by-Optimal-UPFC-Allocation-in-Large-Scale-Power-Systems
Because of excessive costs of UPFCs, the number and the location of these devices should be determined optimally. This paper proposes a method of optimization of UPFC allocation based on specific coefficients algorithm (SCA) to obtain their optimal number, location and sizing. The proposed SCA noticeably improves the accuracy and performance of traditionally used optimization processes especially in large scale networks. This method is applied to the 118-bus IEEE standard system. The results of ordinary and new optimization algorithm have shown the great improvement in optimization process using SCA.