Electromagnetic levitation has been commonly researched for the use in ground transit systems. It is ideal for high-speed applications that require low friction. The principle is simple, use electromagnetic force to balance the force imposed by gravity. However, for attractive levitation the system is unstable and nonlinear. Two dominant approaches to this problem have been to use a state feedback control system or a simple linear PID compensated control architecture. State feedback is a well-known control technique, but is complicated to implement and can rely on linearization of the system dynamics. The simple PID control structure is very easy to implement, but can have severe performance degradation in the presence of noise. This system can usually be identified by its large acoustic noise. This is primarily due to the differential term in the controller. This thesis proposes a solution that uses two concepts: Current Command Generation (CCG) and a closed velocity loop. CCG linearizes the control structure by utilizing the known magnetic properties of the system to convert a desired force to a current for any given air gap. This removes squared command terms from the control structure. This allows for a reliable and predictable implementation of linear feedback control systems. The PID implementation of an attractive levitation system uses two control loops. The inner loop is a current controller, which receives current commands from the outer position loop. The proposed control architecture uses three loops. The innermost loop is the current controller, which receives current commands for the CCG. The middle loop is a velocity controller, which receives commands from the position (outer most) loop and produces force command output used as inputs to the CCG. The three loops consist of two Proportional Integral (PI) controllers for the current and velocity controllers and a Proportional (P) controller. There is no derivative term, making the proposed solution's performance far less dependent on noise. This architecture removes the necessity of nonlinear elements in the control architectures and improves noise rejection through the use of the velocity loop. The acoustic noise performance of this system is enhanced by both of these methodologies and is shown in the experimental setup. / Master of Science

Electromagnetically driven dosing pumps are suitable for metering any kind of liquid in motor vehicles in a precise manner. Due to the working principle and the pump design, an undesired noise occurs when the armature reaches the mechanical end stops. The noise can be reduced by an adequate self-learning control of the supply energy using a position estimation and velocity control. Based on preliminary investigations /1/, a method for noise reduction is realised by using a user-friendly, tiny and cost-efficient hardware, which enables a use in series manufacturing. The method requires only a voltage and current measurement as input signals. The core of the hardware is an 8-bit microcontroller with 8 kilobytes flash memory including necessary peripherals. A smart software development enables an implementation of the entire noise reduction method onto the tiny flash memory.

[pt] Discute-se a modelagem de segunda ordem do acionamento por motor de indução de rotor bobinado usando cascata subsíncrona e o controle de velocidade do mesmo. As estratégias de controle habituais são criticadas e propõe-se um controlador PI com realimentação proporcional de corrente. Propõe-se também um algoritmo de projeto considerando as possibilidades de ajustes finais. / [en] The second order modeling and speed control of the Wound Rotor Induction Motor Drive using Subsynchronous Converter Cascade are discussed. The usual control projects is criticized, and a PI controller with current proportional feedback, and also a algorthm for the design with possibilities of final adjustments are presented.

[pt] O controle de velocidade do motor de indução do rotor bobinado, através de uma cascata de conversores estáticos no rotor, conhecido como Sistema Kramer Estático, é considerado um dos melhores acionamentos para motores acima de 500 Hp, devido às suas características de baixo custo inicial, menor complexidade do sistema de conversores e regeneração de energia. Nestas últimas décadas, as pesquisas nesta área se concentraram basicamente na melhoria da estrutura dos conversores estáticos, o projeto de controle recebeu pouca atenção e sua estrutura de controle continuava inspirada no controle do motor de corrente contínua. Em um recente trabalho (1987) foi proposto um controlador PI com realimantação proporcional de corrente no elo c.c. . Contudo, este projeto, como os outros, foram baseados em uma modelagem simples. Nesta dissertação de mestrado, revê-se e critica-se modelagens e projetos de controle existentes, desenvolve- se um modelo de 6ª ordem e propõe-se um projeto de controle que atente às especificações usuais de controle. A estrutura proposta é composta de um regulador de estado mais um observador parcial de estado. Esta estrutura é fisicamente realizável e o seu algoritmo de projeto é simples, claro e direto. Comprova-se que o uso de modelos mais completos e estruturas de controle mais eficazes não implicam em controladores complexos e irrealizáveis. / [en] The speed control of wound rotor induction motor by a static converter cascade (Kramer Static System) is considered one of the best large motor drive (above 500 Hp.) due to its characteristics of low initial cost, less complexity of the converters and energy regeneration. In the last decades, the researches emphasize basicilly the improvement of the converter structure. The control design was almost forgotten and its structure was Kept based on the direct current motor control. In 1987, it was proposed a PI controller with proportional feedback of direct current. But, this last design, like the others, was based on a simple model. In this dissertation, the modellings and control designs are reviewed and criticized. Besides this, a sixth order model is developed and a new control design is proposed in order to solve the usual control requirements. This proposed structure is composed of a state regulator plus a partial state observer. It´s also of easy achievement and its design algolirhm is simple, easy and direct. It´s confirmed that the use of morepowerful control structure doesn´t implacate in complex and unfeasible controller.

Electromagnetically driven dosing pumps are suitable for metering any kind of liquid in motor vehicles in a precise manner. Due to the working principle and the pump design, an undesired noise occurs when the armature reaches the mechanical end stops. The noise can be reduced by an adequate self-learning control of the supply energy using a position estimation and velocity control. Based on preliminary investigations /1/, a method for noise reduction is realised by using a user-friendly, tiny and cost-efficient hardware, which enables a use in series manufacturing. The method requires only a voltage and current measurement as input signals. The core of the hardware is an 8-bit microcontroller with 8 kilobytes flash memory including necessary peripherals. A smart software development enables an implementation of the entire noise reduction method onto the tiny flash memory.

No description available.

Active control of sound and vibration fields has become an strong area of research over the past few decades. In regards to the active control of acoustic radiation from vibration fields, known as active structural acoustic control (ASAC), there have been many different methods employed to understand structural and acoustic relationships and to control vibrations to limit the acoustic radiation. With active sound field control, sensors, usually microphones, need to be dispersed in the sound field, or an array of microphones must be placed directly in the sound field which, in many cases, uses up too much space for practical applications. To remedy this, objective functions have been transferred to the structure, sensing vibrations rather than pressures. A small, integrated array of structural sensors can be placed on the structure, reducing the system's overall footprint. Acoustic energy density has become a well established objective function, which produces a more global effect using only a local measurement. Another benefit of acoustic energy density lies in the breadth of sensor placement. While acoustic energy density has proven successful in active noise control (ANC), the quantity deals with pressures, not surface vibrations. The problem with ASAC is that an objective function with the robustness of acoustic energy density does not yet exist. This thesis focuses on a structural error sensing technique that mimics the properties of acoustic energy density control in the sound field. The presented structural quantity has been termed Vcomp, as it is a composite of multiple terms associated with velocity. Both analytical and experimental results with the control of this quantity are given for a rectangular plate. The control of Vcomp is compared to other objective function including squared velocity, volume velocity and acoustic energy density. In the analytical cases, the benefits include: control at higher structural modes, control largely independent of sensor location, and need for only a single point measurement of squared Vcomp with a compact sensor. The control at higher frequencies can be explained by the control of multiple acoustic radiation modes. Experimental results offer some validity to the analytical benefits but alternate sensing techniques need to be investigates to more fully validate these benefits.

There are many forms of speed control for an autonomous ground vehicle currently in development. Most use a simple PID controller to achieve a speed specified by a higher-level motion planning algorithm. Simple controllers may not provide a desired acceleration profile for a ground vehicle. Also, without extensive tuning the PID controller may cause excessive speed overshoot and oscillation. This paper examines an approach that was designed to allow a greater degree of control while reducing the computing load on the motion planning software. The SpAM+PI (Speed-based Acceleration Map + Proportional Integral controller) algorithm outlined in this paper uses three inputs: current velocity, desired velocity and desired maximum acceleration, to determine throttle and brake commands that will allow the vehicle to achieve its correct speed. Because this algorithm resides on an external controller it does not add to the computational load of the motion planning computer. Also, with only two inputs that are needed only when there is a change in desired speed or maximum desired acceleration, network traffic between the computers can be greatly reduced. The algorithm uses splines to smoothly plan a speed profile from the vehicle's current speed to its desired speed. It then uses a lookup table to determine the correct pedal position (throttle or brake) using the current vehicle speed and a desired instantaneous acceleration that was determined in the splining step of the algorithm. Once the pedal position is determined a PI controller is used to minimize error in the system. The SpAM+PI approach is a novel approach to the speed control of an autonomous vehicle. This academic experiment is tested using Odin, Team Victor Tango's entry into the 2007 DARPA Urban Challenge which won 3rd place and a $500,000 prize. The evaluation of the algorithm exposed both strengths and weaknesses that guide the next step in the development of a speed control algorithm. / Master of Science

Longitudinal velocity control, or adaptive cruise control (ACC), is a common advanced driving feature aimed at assisting the driver and reducing fatigue. It maintains the velocity of a vehicle and ensures a safe distance from the preceding vehicle. Many models for ACC are available, such as Proportional, Integral, and Derivative (PID) and Model Predictive Control (MPC). However, conventional models have some limitations as they are designed for simplified driving scenarios. Artificial intelligence (AI) and machine learning (ML) have made robust navigation and decision-making possible in complex environments. Recent approaches, such as reinforcement learning (RL), have demonstrated remarkable performance in terms of faster processing and effective navigation through unknown environments. This dissertation explores an RL approach, deep deterministic policy gradient (DDPG), for longitudinal velocity control. The baseline DDPG model has been modified in two different ways. In the first method, an attention mechanism has been applied to the neural network (NN) of the DDPG model. Integrating the attention mechanism into the DDPG model helps in decreasing focus on less important features and enhances overall model effectiveness. In the second method, the inputs of the actor and critic networks of DDPG are replaced with outputs of the self-supervised network. The self-supervised learning process allows the model to accurately predict future states from current states and actions. A custom reward function has been designed for the RL algorithm considering overall safety, efficiency, and comfort. The proposed models have been trained with human car-following data, and evaluated on multiple datasets, including publicly available data, simulated data, and sensor data collected from real-world environments. The analyses demonstrate that the new architectures can maintain strong robustness across various datasets and outperform the current state-of-the-art models.

An assessment study of a novel approach is presented that combines discrete state-space Dynamic Programming and Pontryagin’s Maximum Principle for online optimal control of hybrid electric vehicles (HEV). In addition to electric energy storage and gear, kinetic energy and travel time are considered states in this paper. After presenting the corresponding model using a parallel HEV as an example, a benchmark method with Dynamic Programming is introduced which is used to show the solution quality of the novel approach. It is illustrated that the proposed method yields a close-to-optimal solution by solving the optimal control problem over one hundred thousand times faster than the benchmark method. Finally, a potential online usage is assessed by comparing solution quality and calculation time with regard to the quantization of the state space.