Sharif Digital Repository

This study presents a framework for safe decision-making in autonomous lane-changing using Deep Reinforcement Learning (DRL). The primary objective is to reduce risks and accidents on highways while improving the safety and efficiency of autonomous driving systems. For this purpose, DQN and PPO algorithms were utilized as the core decision-making components. Additionally, a probabilistic risk assessment based on Bayesian theory was employed to analyze environmental conditions and mitigate risky behaviors, thereby preventing collisions during the decision-making process. One of the key innovations of this research is the design and implementation of a multi-criteria reward function that... 

Deformation and Settlement of the ground surface and stability of large underground space is the main concern in the excavation under urban area. In order to control settlement and increase stability, it is necessary to use pre-Supporting system prior to excavation. Concrete Arc Pre-supporting System (CAPS) is one of the most useful pre-supporting methods which is used in exchange station of line 3 and 4 of Mashhad subway. Line 3 and 4 are parallel in Shohada square and the groundwater level is above the whole station structure. To analyze deformation of the ground due to excavation stages, the station was modeled in 2D and 3D by Abaqus. In this study, to estimate ground Deformation, we need... 

The airport terminal is difficult to plan and design due to the different components and the intricate complexity of its parts. At the airport terminal, we face a queue system which simulation is always a good way to analyze. In addition, computer simulation using modern softwares with the capability to model different types of systems, including the airport terminal, while being less realistic than analytical models, is less costly than operational Exercise.In this study, a supportive approach for better and specific management of Iranian airport terminal buildings with discrete-event simulation is proposed, considering the need for managing the airport terminals building in the country and... 

Limited water resources in Iran have made dams an important part of the country’s infrastructure. As a result of the large number of in-progress and expected dam projects in the country, consideration of sustainability factors during planning, construction and operation phases of dam projects can create significant economical, social and environmental impacts in the long term. To come up with an inclusive sustainable development, new sustainability assessment approaches seek for sustainable development at the entire project’s life-cycle; they are looking for sustainable development at all design, construction, operation, maintenance and even demolition phases. In this research we have... 

Nowadays, most of businesses are becoming computer-based. This trend causes the organizations to become more vulnerable to internal and external threats. Internal threats are done through insiders. Insiders unlike outsiders have capabilities which aid them to easily carry out their illicit activities. Unfortunately, most of research in computer security has focused on the outsider threats, whereas the most damaging is the result of insiders’ illicit activities.This dissertation addresses three primary necessities of the research in the area of insider threat: a consistent definition of insider threat, a classification of types of insider threats, and a classification of vulnerabilities which... 

This project focuses on the Tennessee East-man challange problem(TEP) and the impact of setpoint changes and controller tunning on plant performance and profitability. TE has four major units. it is an unstable system due to the reaction within the reactor and a recycle streamcausing snow ball effect. As a result the control of this system is not easy. The study has been conducted along two main directions. First, selecting the optimal values for the setpoints and applying them in the simulation. secondly, try to find better parameter for controller in simulation. Simulation is based on working of Ricker. In TEP like the other multivariable systems, it is ameanable to use a centralized... 

Decentralized control is a typical method for the design of industrial MIMO control systems. However, in most multivariable systems, the existence of large interactions limit the achievable performance of such controllers. In the Nyquist Array design method proposed by Rosenbrock, the system interactions can be reduced using a precompensator, such that a decentralized controller may be used with acceptable performance. A well designed pre-compensator will effectively decouple the system and the multivariable systems may be considered as a set of SISO systems. It is therefore obvious that the key element in the Nyquist Array design method is the design of the pre-compensator. There are two... 

Integrity of closed-loop systems with integral action control is characterized by using a series of successively more difficult to obtain conditions: Integral Stabilizable (IS), Integral Controllable (IC), Integral Controllable with Integrity (ICI), and Decentralized Integral Controllability (DIC). The IS condition requires that for a Linear Time-Invariant (LTI) finitedimensional (FD) plant, G(s), one finds a stabilizing decentralized controller with integral action. For G(s) to be IC, in addition to being IS, one should be able to reduce the gain of
all control loops by the same factor from a finite value to (but not including) zero without introducing instabilities. ICI requires that... 

Intraction in multi input multi output (MIMO) system makes control of these systems difficult.There are two ways to control such systems; Centralized and Decentralized control. One way in which a centralized controller is designed is using a precompensator. With application of the pre-compensator, the MIMO sytem is decoupled into several SISO systems and one can control these seperatly. Almost all previews methods for the design of pre-compensator assume that the model of the plant containsno uncertainties.In this thesis a new method is proposed which takes into account model uncertaintes in the design of the pre-compensator.The form of the uncertainties considred an interval-parametric  

Multivariable control systems as used in the industry usually contain integral action in order to improve tracking of the output with respect to step reference changes. These systems are usually subject to input saturation due to actuator or sensor nonlinearities. The aim of this work is find the conditions for the integrity concept of such multivariable system, e.g. IS, IC, ICI and DIC. First we develop the conditions on the controller gains necessary stabilize the systems and then show that these conditions can changes when one examines integrity conditions  

The servomechanisms have wide industrial application such as robotics, Computer Numerical Control (CNC) machines, automated manufacturing and solar tracking. Several methods have already been proposed for servo motor control that the most important of them are Proportional Integral Derivative (PID), H2/H--, intelligent based controllers, and model predictive control. Problem specifications and user requirements, will determine the appropriate approach in each case. The aim of this thesis is construction of a real-life camera stabilizer to be used in the motion picture industry. Effects of base movement on the performance should be decreased by choosing appropriate sensors and control... 

In continous chemical process, disturbances usually propagate along the process. This project describes a data-driven method for identifying the direction of propagation of disturbances using historical process data. The concept is the application of Transfer Entropy, a method based on the conditional probability density functions that measures directionality of variation. For demonstration purposes the transfer entropy method is used to isolate the root-cause of the disturbance in the Tennassee Eastman benchmark process  

In this thesis, the optimal control structure selection and design of sparse multi-variable control systems is addressed. A fundamental challenge which frequently emerges in engineering, social, and economic sciences, is the optimal selection of a subset of elements, in order to maximally fulfil a design objective. In practice, it is required to solve this underlying selection problem in conjunction with a non-linear or non-convex optimization which is designed to ensure desired performance. The requirement to solve these two problems simultaneously is what makes it inherently difficult; one which has thus far eluded efforts to develop a systematic means of determining its solution. In spite... 

Design of fault tolerant controllers with a reliable and robust control structure that can accommodate sensor and actuator failure is a challenging problem. In the process control literature, the term “integrity” is used to characterize these properties for decentralized control systems. It refers to the ability to retain system stability when one or some of the control loops get in or out of service or when loop gains are detuned.Conditions pertaining to the integrity of the closed-loop system are described by a series of successively more difficult to attain criteria whose evaluation is only based on the process steady state gains. These are Integral stabilizibility (IS), Integral... 

For increasing longevity of a structure, it is not enough to notice only to designing aspects. In operation phase of a construction, we can increase the level of service and longevity. Asset-based Organizations, in addition to design their new assets, should maintain them. Infrastructures, such as road, bridge, dam, and water network, are important because of social and economic aspects. So they need special attention. There are efficient models for infrastructure maintenance that consider infrastructure element information, such as condition rate, weight, dimensions, deterioration rate, and cost of maintenance as inputs and give planning for maintenance in future. This planning, involves a... 

Annually there are many small-scale industry projects implemented in Iran, most of which are financed by local financial institutes. The financing agreements between project owners and financial institutes are usually formed and finalized during the initiation phase and are based on feasibility studies done at the early stages; before project design and construction begin. However, in many cases actual costs of projects exceed the costs estimated in feasibility studies which have been approved by financial institutes. Limited financial resources in the country, from one side, and the lengthy process of increasing the project financing limit, from another side, cause delays in the project... 

Optimization-based wind farms control focusing on fatigue loads minimization represents a promising approach to establish a balanced tradeoff between power production in wind farms and lifetime extension of the wind turbines. This dissertation aims to develop a novel strategy to determine individual wind turbine power set-points in such a way that fatigue loads imposed on the units are minimized. The research considers main oscillation modes of the turbine structure and the determining forces to evaluate the amount of fatigues.Based on detailed modelling of wind farms which includes influences of wake effects on the mean wind speed and turbulence intensity, a novel approach is used to... 

Nowadays, maintenance management is very important in manufacturing industries and reducing the time and cost of maintenance can have a significant impact on their performance. Traditional maintenance management systems have weaknesses such as poor user interface, lack of visualization and high complexity, limited access to required documents. On the other hand, in the process of implementation of maintenance management in manufacturing industries, weaknesses in teamwork and poor information flow between different groups, especially production and maintenance groups, are evident. For this purpose, a framework has been presented in this study to improve the process of maintenance management... 

During the last decades, the exponential population growth as well as the adverse impacts of climate change have brought about serious challenges in supplying freshwater demands. In this regard, buildings account for a considerable portion of freshwater withdrawal. Accordingly, urgent measures should be taken to tackle this crisis. Rainwater harvesting (RWH), Greywater recycling (GWR), and Hybrid rainwater greywater (HRG) systems are considered efficient substitute resources that are able to reduce the consumption of water supplied by centralized networks in buildings. Nonetheless, there is a vital need for the financial and technical assessment of these systems. Additionally, multiple...