Department of Electrical Engineering
Permanent URI for this communityhttps://repository.cuilahore.edu.pk/handle/123456789/18
Browse
107 results
Search Results
Item Adaptive Nonlinear Gain Approximation of a Robust Nonlinear Controller to Handle Uncertain Disturbances for UAV Systems Using Neural Networks(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Waqas Ahmad; CIIT/FA23-REE-003/LHR; Dr. Mirza Tariq Hamayun; LHR TP 10045The quadrotor unmanned aerial vehicle (UAV) has been a part of scientific studies owing to its applications in military, agriculture, infrastructure and construction, and environmental monitoring and conservation. However, its control design is challenging due to its nonlinear underactuated dynamics, external disturbances, parametric uncertainties, and the aerodynamic variations encountered during flight operations. Sliding mode control (SMC) is the most widely used control technique for handling nonlinear dynamics in the presence of external disturbances and modeling uncertainties. However, conventional SMC requires a priori knowledge of the upper bound on the overall disturbance at all times. Usually, these upper bounds on disturbances are derived from past data, environmental conditions, and empirical modeling; therefore, a conservative bound is used in SMC design, leading to high control effort and draining more energy from the system. To address this issue, adaptive sliding mode control (ASMC) strategies based on a neural network (NN) are presented in this dissertation. Neural networks are recognized as the leading approach to intelligent computation and are highly effective in handling nonlinearity, fault tolerance, adaptation, and continuous online learning. In this research, NN-based ASMC strategies are presented for a quadrotor UAV to improve flight performance by online estimation of adaptive modulation gain in response to variations in disturbance magnitude. Different neural network schemes are employed, including feed-forward neural networks (FNNs) using backpropagation (BP) and Levenberg-Marquardt (LM) algorithms for optimization, and a radial basis function neural network (RBFNN) with a Gaussian activation function, to adaptively estimate the modulation gains. Furthermore, the stability analysis of the proposed controllers is proven using Lyapunov theory.Item Implementation of Machine Learning Models for Predicting Hydrogen Production from Renewable Energy(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Adnan Ayub; CIIT/FA23-REE-002/LHR; Dr. Muhammad Yaqoob Javed; LHR TP 10044The world continues to be largely reliant on fossil fuels, such as coal, oil and natural gas, which are major contributors of greenhouse gases, air pollution and climate change caused by the release of gases like CO2, NO o and SO2. Even though renewable energy sources such as solar and wind provide a cleaner substitute, their nature as intermittent and weather-dependent sources become a big problem in terms of large scale and long-term energy storage making traditional battery systems economically impractical. Consequently, the utilization of the surplus renewable energy through the process of water electrolysis to create green hydrogen has become a viable and alternative way of storing energy in the long term. This work suggests a two-stage machine learning-based predictive model of hydrogen production when using renewable energy, based on real-world working data of a 40.5 MW grid-connected photovoltaic (PV) power facility. The initial step involves predicting photovoltaic power output from meteorological variables using multiple regression and deep learning models, such as Support Vector Regression (SVR), Random Forest, Decision Tree, and Gated Recurrent Units (GRU). The second stage is to incorporate the predicted solar energy into a hybrid electrochemical model of hydrogen production, and the same machine learning models serve as data-driven correction models to more effectively predict hydrogen yield. It is a two-stage method that integrates both physical modelling and machine learning to model nonlinear system behaviour and real-world losses of operation. The findings show that SVR was always better in both phases than the other models with an R 2 value of 0.965 in photovoltaic power prediction and 0.968 in hydrogen production prediction. The given framework minimized the mistake in the production of hydrogen annually to about 3 percent, which is much better than theoretical models and deep learning alternatives. Moreover, a Sobol based global sensitivity analysis revealed that Global Horizontal Irradiance (GHI) made the greatest contribution to the uncertainty in the hydrogen production process, then AC power output and temperature at the module. The outcomes of these studies support the premise that the two stages proposed framework is a viable, precise, and large-scale solution to real-time forecasting, optimization of the system, and successful grid integration of green hydrogen system.Item Energy Aware Path Planning and Guidance for Non- Holonomic Robots in a Manufacturing Workshop(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Rimsha Rehman; CIIT/FA23-REE-005/LHR; Dr. Mujtaba Jaffery; LHR TP 10046Automated Guided Vehicles (AGVs) are now a core part of modern manufacturing workshops, enabling fast, reliable, and flexible material movement while reducing dependence on manual handling. Despite their importance, most existing path planning approaches still focus mainly on minimizing distance or travel time. These strategies overlook a crucial factor, energy consumption, which directly affects operational cost, battery health, sustainability goals, and overall system efficiency. Addressing this limitation, this thesis presents an Energy Aware Path Planning (EAPP) framework specifically designed for single-load, non-holonomic AGVs used in structured workshop environments. The proposed framework models the workshop layout as an undirected graph, where nodes represent workstation points or intersections and edges represent feasible routes that comply with AGV kinematic and turning constraints. Unlike traditional planners, our approach integrates a physics-based energy model into the A* algorithm, allowing each edge to be evaluated not only by its geometric length but also by its expected energy usage, considering acceleration, deceleration, rolling resistance, turning angles, and standby power. This enables the AGV to prioritize paths with fewer turns, even when they are slightly longer, ultimately reducing total energy consumption. The method is implemented and validated using MATLAB simulations, where both distance-based A* and the proposed energy-aware A* are compared under realistic AGV parameters. The results show that the EAPP framework significantly lowers energy consumption and maintains competitive travel times, while still producing feasible and safe navigation paths within constrained workshop networks. By directly embedding energy considerations into the path planning process, this work contributes to more sustainable intralogistics and provides a practical, scalable solution suitable for real-time AGV navigation and industrial deployment.Item Estimation of Wake Effect in Wind Farms using Machine Learning Algorithms(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Muhammad Usama Riaz; SP22-REE-001; Dr. Aamer Bilal Asghar; LHR TP 9501The need for renewable energy is always growing, making wind farm efficiency opti mization a vital area of study. Wind turbines that are dispersed throughout the landscape on wind farms use the kinetic energy of the wind to generate electricity. The idea that a wind turbine running in the wake of another turbine lowers wind speed after the rotor and reduces power generation is introduced by these massive wind farms with several turbines. The flow field in the wake of first row turbines is characterized by strong wind velocity deficit and high turbulent intensity as well. For this reason, a further downstream turbine in a wind farm can capture lesser wind energy than the first row turbine. Most importantly wake effect modeling is done in this study since it is important in determining the real en ergy production of a wind farm enhanced by this feature. In this study, machine learning algorithms are used to determine the wake effect of a wind turbine in a wind farm. Although the wind turbines appear simple in their working principle, their control is actually complex and difficult to maintain operational and structural integrity and performance. Each turbine produces a ‘wake’ when it takes wind energy; this is a major factor in the overall dynamics of the turbines. These wakes, which are areas of turbulent flow and decreased wind speed behind the turbines, are a major barrier to efficient wind energy harvesting. Moreover, the extension of the mechanical life of the turbines in not only increase efficiency but also sus tain turbulence in the wake. The present research work, specifically focuses on estimating wake effects in wind farms for machine learning algorithms. The data used for this aim is archival data of wind farm recordings, but in this case, the data is computationally syn thesized. The goal is to improve accuracy and efficacy while reducing wake estimating loss compared to traditional methods. Utilizing machine learning models such as Artificial Neural Networks (ANN), Random Forest Regression, Decision Tree Regression, Support Vector Machines (SVM)andtheadaptive neuro-fuzzy inference system (ANFIS) which are used to capture complex architecture and forecast accuracy. The outcomes of this research have the potential to revolutionize wind farm optimization by providing a more adaptive and faster wake estimation process. Among the models tested, the Random Forest Model performed the best with an outstanding R2 score of 0.99, demonstrating its effectiveness in accurately estimating wake effects. In comparison, the Decision Tree Regression also showed strong performance with R2 score of 0.97, although it exhibited a slightly higher MSEandRMSEItem Estimation of Syngas Production in Biomass Using Machine Learning Algorithms(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Nabeel Arshad; FA22-REE-006; Dr.Aamer Bilal Asghar; LHR TP 9499Renewable energy technologies can be beneficial for the economic development of any country on the planet. Also, with the present situation in the energy sector, the high cost of fuel, the use of renewable energy resources has seemed to have gained importance. The energy resources that pertain to solar, biomass, and wind energies are clean energy friendly to the environment. In the modern world, fossil fuels meet the requirements of the transportation sector in large volume and bring several negative impacts viz., air pollution, noise, and global warming. Besides, the fast decline of below-ground petroleum resources that arise with the rapid consumption of fossil fuels is accompanied by another grave problem for the transportation sector. Usually, the gasifier's temperature is greater than 1000°C. Before coal is put into the gasifier, it must go through a process called pyrolysis, a sequence of intricate physical and chemical reactions that occur without oxygen or air and gradually increase in temperature from 150 to 700 degrees Celsius. High molecular weight charcoal and volatile substances, such as CO, H2, H2O, CO2, and CH4, are the process byproducts. In this research work, machine learning algorithms such as Artificial Neural Network (ANN), Random Forest (RF), GBR, and XGB will be used to increase the efficiency of biomass gasification using AI-based machine learning algorithms. A dataset of biomass converting into syngas and other useful products is obtained to estimate the output of syngas, H2, CH4, and composition of different gases based on input parameters, namely carbon, temperature, sulfur, oxygen, nitrogen and ash, etc. The algorithms have an advantage in their capacity to train a model using a dataset consisting of numerical values and achieve an optimal fit point by incorporating efficiency parameters. Root Mean Square Error (RMSE) is a frequently used metric to estimate the precision of predictions. The RMSE values will be computed for each scenario. The approach will be experimented with under temperature variations, and other relevant constraints.Item Landslide Monitoring Using State of the Art Machine Learning Techniques with Image Processing of Space Borne Remote Sensing Imagery Data(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Mohammad Mazan; FA21-REE-003; Dr. Muhammad Farooq-i-Azam; LHR TP 9366Landslides present significant worldwide difficulties, leading to profound repercussions such as property destruction and human death. As a result, the methods we currently have, including light detection and ranging (LiDAR) and on-site assessments by experts to look at things like topography, ground fractures, slope stability, and changes in vegetation, are not good enough to accurately predict and prevent these events. The rising frequency of landslides worldwide, intensified by alterations in weather patterns and human actions, underscores the necessity for sophisticated alert and surveillance tools. Landslides can result in devastating outcomes, such as fatalities, displacement of individuals, and significant economic repercussions. Technological advancements, namely in machine learning, remote sensing, and image processing, provide promising options to improve landslide prediction and mitigation efforts. This work aims to address the critical issue of evaluating and predicting landslides, with the objective of creating a dependable approach. We extract landslide causative factors from remote sensing imagery data through image processing, compile them into a single dataset, and then use this dataset to train machine learning algorithms. The landslide causative factors included in our dataset are the Topographic Wetness Index (TWI), Peak Ground Acceleration (PGA), Stream Power Index (SPI), Terrain Ruggedness Index (TRI), Normalized Difference Vegetation Index (NDVI), curvature, elevation, faults, rainfall, geology, solar radiation, aspect, landcover, slope, streams, and roads. We tested how well different machine learning algorithms could predict the risk of landslides. These included Logistic Regression (LR), Random Forest (RF), XGBoost (XGB), AdaBoost, Long Short-Term Memory (LSTM) networks, Deep Neural Networks (DNN), NGboost, and 1D Convolutional Neural Networks 1D (CNN 1D). The results demonstrate significant accuracy scores for each model, with the LSTM network emerging as the most efficient, showcasing an accuracy of 0.91 and an amazing Area Under the Curve (AUC) score of 94%. The success of the LSTM can be due to its proficiency in managing sequential and time-dependent data, which is essential for comprehending and predicting the dynamic characteristics of landslides. The LSTM’s ability to utilize temporal patterns improves the effectiveness of early warning systems, representing a notable progress in landslide prediction approaches.Item Power Potential Assessment of Wave Energy and Integration with other Renewable Energy Resources for Coastal Areas(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Muhammad Shaharyar Haider; FA21-REE-005; Dr. Sobia Baig; LHR TP 9367Climate change is one of the most signicant challenges faced by the world. As the hu man population continues to grow, so too does the demand for energy. Fossil fuels being the primary source of energy produce greenhouse gases which adversely affect our climate. The United Nations Sustainable Development Goals (UNSDGs) emphasize the urgent need for reliable, sustainable, and clean energy. Wave energy is a promising form of renewable energy that harnesses the power of ocean waves to produce electricity. While it is under utilized and research on it is lacking in comparison to other renewable energy resources, it can contribute signicantly towards achieving the UNSDGs by providing a clean source of electricity for coastal areas. This research work assesses the potential of wave energy at a selected coastal location, aiming to evaluate its feasibility as a complementary energy source alongside other renew able energy sources. Historical wave parameter data spanning 15 years is obtained for the selected location and processed. Using Python, the dataset comprising of sea state param eters in each month of the year is then tted with probability density functions. The Monte Carlo simulation is used to generate synthetic wave scenarios. These simulations incorpo rate randomness in the wave parameters to assess the variability of wave energy generation. The MATLABbased tool for simulating wave energy converters, WEC-Sim, is used to as sess the capabilities of a converter in harvesting the resource based on the monthly average of wave parameters obtained from the simulations. In addition, a hybrid wave and solar energy system is developed using MATLAB Simulink to integrate the two renewable re sources. The analysis revealed signicant seasonal variations in wave power. At the selected location, the yearly average wave power was observed to be 376 kW/m, with the winter months from October to March showing a high availability of wave power. The month of January has the highest theoretical wave power potential of 675 kW/m. The sum mer months showed signicantly lower power potential with the month of July showing the lowest potential of 140 kW/m. This suggests seasonal variability in wave energy resource as sea conditions change and impact generation. The WEC-Sim simulations revealed that the wave energy converters only capture a fraction of the wave power. The commercially available and tested wave energy converters are analyzed and the most suitable converter is selected for use at the selected location. The research concludes that wave energy is a vi able energy source with negligible greenhouse emissions, and is capable of reducing yearly energy costs by up to 18250 USD per meter of energy capture from wavefronts. These ndings underscore the importance of integrating wave energy into the broader renewable energy portfolio to achieve the UNSDGs, especially in remote coastal communities where grid infrastructure is challenging to set up and maintainItem Output Observer-based Sensor Fault Detection in Twin Rotor Multiple Input Multiple Output System(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Muhammad Salik Bilal; FA20-REE-010; Dr. Mirza Tariq Hamayun; LHR TP 9368Complex air vehicles are challenging to analyze due to their highly nonlinear behavior and considerable cross-coupling interaction. The control of a Twin Rotor Multiple Input Multiple Output System (TRMS) is difficult due to significant cross-coupling and nonlinear dynamics that resemble to helicopters. A beam that can freely rotate in both the vertical and horizontal planes on its base makes up the TRMS. The main rotor and the tail rotor are its two rotors. Both of these rotors are driven by DC motors. Gyroscopic disturbances and sensor faults in the rotor motors during rotation can affect the TRMS's stability and input tracking. In this work, an optimal controller is designed to ensure stability and reference tracking. Furthermore, an observer is designed to estimate the system outputs for sensor fault detection in TRMS subjected to deterministic disturbance and norm-bounded uncertainty in system matrix (A) using Linear Matrix Inequalities (LMIs) technique for TRMS. The effectiveness of reference tracking and estimation of the system outputs for sensor fault detection has been investigated through simulation environment.Item An Efficient Fault Detection Method for Grid Connected Solar Photovoltaic System(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Zain Rasool; FA21-REE-001; Dr. Muhammad Yaqoob Javed; LHR TP 9369Photovoltaic (PV) module faults have harmful effects on both the efficiency of power generation and overall safety. Among these faults, current mismatch is the most common type, leading to a decrease in output current and causing distinct steps in the (current and voltage) I-V representative curves as well as multiple spikes in the P-V curves. Consequently, the power output of PV units is significantly impacted. This research delves into the scrutiny of faulty PV units in real-world PV power sites, specifically focusing on current inequality faults resulting from partial shadowing, hot spots, and cracks. There are various techniques used to detect the current faults. These techniques are ground fault detection and interruption (GFDI), over current protection (OCP), Insulation monitoring devices (IMD), and Arc fault current interruption (AFCI). Other than these devices various classification algorithms have also been developed which can be employed to classify the detected PV faults while the system is running. In this research the dataset from previous research is used to train regression tree, SVM, and logistic regression classifiers. Amongst these classifiers, regression tree classifier has presented an accuracy of up-to 99%, while the previous research presented an accuracy of 98%. This research distinguishes between different fault features within the I-V curve steps and proposes computational analytics and statistical techniques for diagnosing PV unit mismatch faults. The inclusion of PV system reduces the carbon footprint paving ways to green energy, in addition to saving fuel and generation costs on a yearly basis. As such this research aligns itself with the Sustainable Development Goals (SDGs) set by the United Nations.Item Design and Development of an Intelligent Control System for Multifunctional Agricultural Simulator(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Muhammad Afzaal; FA22-REE-005; Dr. Fawad Azeem; LHR TP 9370