Search

The permanent magnet synchronous motor (PMSM) is preferable for servo applications and domestic appliances, where high operating efficiency, low power loss, and ease of maintenance are required; however, the cost of the PMSM drive system is relatively high if mechanical sensors are used to measure the rotor position. In this paper, a minimal-order position observer is applied to the PMSM drive. As the rotor position is estimated with the measurements of the phase currents and line voltages, the expensive mechanical transducers are eradicated. Consequently, the cost of the whole system is reduced and the reliability increased. A number of meaningful issues have been addressed to enforce the theoretical basis of this approach, including the derivation of the discrete plant model, discussion of the modeling error and its influence on state estimation, and stability analysis of the complete system under model uncertainties. Simulation and experimental results under different operating conditions are presented...

This research proposes a new approach to control the position of a permanent magnet synchronous motor drive using the state space generalized predictive control (GPO) algorithm.

Once injected into the orbit, the satellites operate in space all the time and they are exposed under very harsh space environment such as extreme temperature variation in vacuum condition, radiation etc. As a satellite requires significant amount of investment, it is desired to understand its lifespan and performance degradation before a replacement satellite is launched. The degradation process of an electronic component is strongly related to the thermal, radiation and pressure conditions in the low earth orbit. This research focuses on the degradation and fault monitoring of the electronics components in the DC-DC converters of satellite power system. The main contributions of this research are listed below.  Developed a low sampling rate online power converter circuit parameter estimation method. The estimation accuracy is compatible with the conventional high sampling rate methods as indicated in both simulation and experimental results.  Derived the interleaved boost converter’s state space model to include the most important parasitic effects.  Developed a grouped immigration algorithm to accelerate the solution convergence rate in highly inter-relative multi-variable problems using biogeography based optimisation (BBO). The degradation in the converters are detected through the online parameter estimation. The previously reported fault detection methods for the DC-DC converter require a sampling frequency to be at least 25 times of the switching frequency. In this thesis, the new parameter estimation methods are developed based on the averaged converter circuit model. This allows the measurements to be taken only once in every few switching cycles. Therefore, this method can be implemented using low cost and low power processer. This is particular important for miniaturized satellites which have limited power budget due to the limited area of the solar array available for the satellite. The low sampling rate method has been verified through simulation and experimental study for a buck converter. The average estimation error for the circuit components are less than 5.17% in the simulation and 7.5% for the experimental results. The parameter estimation method can be extended to other converter topologies. To provide redundancy in the power system, the multi-phase interleaved converter has been developed for various applications. Such feature is useful for satellite to eliminate single point failure. To develop an online parameter monitoring system, a general state space based interleaved DC-DC converter averaged model is formulated in this thesis. Moreover, the number of measurements and unknowns in this model are scalable . As additional unknowns are added into the optimization problem for multiphase interleaved converter, the convergence rate of the conventional BBO approach dropped. To accelerate the convergence rate of BBO with moderate calculation complexity, the grouped immigration method is proposed in this research. Results have demonstrated that the convergence rate improved significantly such that the estimation error deviation for the circuit component is less than 2% and the averaged error is less than 7% when the function evaluation is restricted to 200,000. Thus it outperformed the conventional BBO method, in which the estimation error is greater than 100% due to premature termination under the same function evaluation limit. Doctor of Philosophy (EEE)

In recent years, the industry of miniature satellite based on cubesat standard has been growing rapidly. This growing industry has changed the direction of miniature satellite from educational and research purposes to application focused. As most of the implemented applications require target pointing, precise attitude determination is highly crucial to the mission success. There are several established attitude determination methods namely the TRIAD algorithm, QUEST algorithm and Kalman filter. They share a common shortcoming of requiring at least two observer-reference pairs as the input to the algorithm. This has caused an issue for nano-satellite’s attitude determination system as the magnetic field generated by the magnetic torquers corrupt the magnetometer reading while performing momentum dumping. To overcome this problem, a modified extended Kalman filter algorithm has been developed in this thesis such that it can work even with the presence of only a single observer-reference pair by trading off attitude knowledge accuracy. From the study, it was noticed that the accuracy of the attitude knowledge drops drastically with time when utilizing only a single observer-reference pair. To further improve the performance, an alternative approach has been developed such that the attitude knowledge accuracy can be retained even when magnetic torquers are switched on. The alternative approach used a hybrid optimization method based on particle swarm optimization and iterated extended Kalman filter to calibrate the magnetometer in-orbit. Beside calibrates for the coupling effects of magnetic torquers, the proposed magnetometer calibration method also calibrates for the in-orbit variation that affects the measurement accuracy of the magnetometer. Both simulation and experimental results show that the measurement accuracy of magnetometer is enhanced and the problem of magnetometer reading corruption while performing momentum dumping is resolved. MEMS based gyroscope is typically used in a nano-satellite. Its drawback is the low accuracy. While there are studies to improve the performance of MEMS based gyroscope, they either require an array of MEMS gyroscopes or a high complexity algorithm and thus not suitable for nano-satellite applications. To address this issue, a low complexity Kalman filter for MEMS based gyroscope has been introduced in this thesis. Besides having a lower computational load, it also reduces the MEMS based gyroscope noise in all three-axes without the implementation of extra gyroscopes. Master of Engineering

The lithium-ion battery has become an important energy source for many applications due to its numerous advantages such as high energy density, lack of memory effect, low self-discharging rate, and long cycle life. This also makes it a very promising candidate for pico- and nanosatellites that have space and weight constraint. For high performance and reliability, it is necessary to have accurate knowledge of the present condition of the battery and the remaining battery life. Among the many ways to monitor the battery conditions, estimation of battery’s state-of-charge (SOC) is one of the critical tasks in battery management of the satellite power subsystem. Since the satellite operates at different temperatures throughout the orbit, its operating temperature, which is the most significant factor that affects the SOC estimation, must be taken into consideration. In this thesis, four different SOC estimation methods for a satellite have been progressively proposed and developed. Under different orbital periods, the satellite operates with different subsystems. Taking advantage of different operational scenarios of satellites in different orbits, a SOC estimation method using ampere-hour counting with impulse response reset is proposed and developed. The impulse response is obtained from turning on/off of satellite subsystems and payloads that occur by the satellite operations rather than using an artificial injected pulse. The charging and discharging dependency on current have been considered and taken into account as the charge and discharge rate factors in the method. Moreover, integral resets can be conducted at all SOC levels instead of the traditional fully charged/discharged state. It eliminates the risk of increasing the number of fully charged and discharged cycles thus extending the battery life span. For this approach, it requires a lookup table to store the various impulse responses which vary at different operating temperatures. To further improve the performance, a SOC estimation method using the square root unscented Kalman filter (Sqrt-UKFST) with unit-sphere spherical transform has been developed. The Sqrt-UKFST takes advantage of Jacobian-free linearization, which is one of the major drawbacks of extended Kalman filter (EKF) based methods used in SOC estimation. It has a higher error order (second order) than EKF (first order). When it is compared with the unscented transform, fewer sigma points are needed for estimation of sample mean and covariance by using the spherical unscented transform. This results in 32% lower computational requirements. The satellite’s operating temperature varies across the orbit and it is one of the critical factors that affect the battery parameters. To include the temperature factor, the SOC estimation with a dual square root unscented Kalman filter (DUKFST) has been developed. In this approach, a dual unscented Kalman filter is used to update battery parameters at different temperature through adaptive covariance matching. The temperature effect on open-circuit voltage (OCV) has been modelled in the DUFKST. All Kalman filter (KF) based methods require the knowledge of measurement and process noise. If not determined properly, it can lead to poor performance in terms of filter divergence and instability. A SOC estimation based on particle swarm optimization (PSO) with inverse barrier constraint is proposed to overcome these drawbacks. This method neither needs to linearize the model nor requires the information on measurement and process noise. DOCTOR OF PHILOSOPHY (EEE)

In this thesis, we focus on the localization of sparse wireless sensor networks (WSN) through exploiting two types of new information, namely the deployment and negative constraint information. For this research, we have proposed and developed a new distributed localization algorithm termed as the likelihood localization algorithm (LLA). Different from other algorithms, the LLA takes advantage of the deployment information through a deployment agent (DA). After a sensor node has been deployed, the system uses both the deployment and the inter-node radio information to improve the estimates of its’ position through a maximum likelihood estimation (MLE) scheme. LLA has been implemented and evaluated using a low-cost microcontroller. Simulation and experimental results show that it outperforms the conventional approaches in terms of localization accuracy in both sparse and dense WSNs. Furthermore, an enhanced likelihood localization algorithm (ELLA) that hybridizes the MLE and an extended Kalman filter is also proposed to achieve better localization accuracy. DOCTOR OF PHILOSOPHY (EEE)

Nano-satellites and Pico-satellites – also called miniaturized satellites – are satellites weighing less than 20kg and 1.33 kg respectively. Over the last decade, there has been growing trend to build miniaturized satellites for research and technology demonstration. One of the major concerns in any satellite system is the power. The problem is more apparent in the miniaturized satellites classes as their extremely limited physical size restricts the area available for the solar panels. There have been various works in the miniaturized satellites research area to tackle this power limitation issue. One sub-area which is yet to be more extensively researched is the optimization of micro-controller’s power use. Some miniaturized satellite projects have been known for having issues in the later phase of their development for overlooking micro-controller’s power use in their early stage of the development. In the computer science field, there have been various researches aimed to reduce power consumption of Central Processing Unit (CPU). One popularly known technique which aims to reduce CPU power consumption through the optimization of the micro-processor’s power use is Dynamic Voltage Scaling/Dynamic Voltage and Frequency Scaling (DVS/DVFS). Though this technique has been widely researched in the computer science field, it has some shortcomings which put this technique off from miniaturized satellite applications. DVS/DVFS relies on special hardware components, namely, a programmable DC-DC switching voltage regulator and a high performance (≥ 400 MHz) processor with wide operating frequency ranges. While those components are available in most of the modern computers, they are hardly present in the miniaturized satellite applications, which are often controlled by low-speed (≤ 100 MHz) micro-controllers and having only a programmable clock generator to generate its system clock frequency. This research work aims to fill the gap between the research in the computer science field and the actual implementation in the miniaturized satellites. By introducing a simple and autonomous system clock frequency controller which only depend on a programmable system clock generator, this research work aims to extend the research on optimization of the high speed micro-processor’s power use to the low-speed micro-controllers, making it suitable for miniaturized satellite applications. MASTER OF ENGINEERING (EEE)

145 p. Wireless sensor network (WSN) has been actively researched in recent years because of its wide range of applications such as natural habitat monitoring, healthcare, home automation, traffic management and military applications. In many cases, the sensed data is not useful without knowing the sensor nodes' locations. Localization algorithms can be categorized as range-free and range-based. In this research, the radio frequency (RF) based approach is used. In conventional RF-based approaches, the path loss parameters in the received signal strength (RSS) - distance model are often determined based on extensive measurements conducted offline. Because the RSS measurements fluctuate with the environment, the use of fixed path loss parameters would reduce the localization accuracy. The purpose of this research is to develop a RF-based localization algorithm where the path loss parameters and the unknown locations are determined simultaneously in real time. MASTER OF ENGINEERING (EEE)

In recent years, remote imaging systems have been used for a wide range of applications in geological exploration, oceanography, meteorology, military reconnaissance, etc. Differing from the normal cameras, the remote imaging systems capture still scene with the camera in uniform motion. Limited by the system moving velocity, only a short integration time is allowed, which will result in a low signal-to-noise ratio (SNR) in dark illumination conditions. Therefore, time-delay-integration (TDI) image sensors are widely applied. In an integration-mode TDI image sensor, the active pixels are placed in more than one row (stage) in across-track direction, with the pixel stages perpendicular to the image sensor movement direction (along-track direction). When the camera system moves at a constant velocity, the pixels are exposed to the same scene stage by stage and the panoramic image in the along-track direction is thus produced. As a result, each pixel stage has contribution to the optical integration, so the effective integration time and the signal strength are enhanced. For the multi-pixel-stage integration mechanism of TDI, the charge-coupled devices (CCDs) offer an effective solution. In CCD image sensors, the photo charge is stored in the potential well of the original pixel following integration, and then transferred out row by row. Meanwhile, the charge of the same column yet different rows can be added together during the transfer operation with higher efficiency. Accordingly, CCDs can perform TDI operation without any external support. Consequently, CCDs still have the largest share of the TDI image sensor market, even though complementary metal-oxide-semiconductor (CMOS) image sensors have increased in popularity in recent years. In most office and industrial applications, TDI cameras are supported by stationary rails and powered by direct current (DC) adapters. Accordingly, the additional motion and power consumption are not serious concerns. Nevertheless, in remote imaging applications, the issues and limitations should be taken into account. Firstly, the satellites which support them are sometimes affected by unforeseen disturbances, such as adjustments of the solar panels, alterations to the momentum wheel and so on, which would introduce residual motion to the across-track direction (vibrations). In TDI CCD sensors, the traveling of the charge pockets on the focal plane should be always aligned with the along-track direction. However, owing to the vibrations, the image produced would be blurred and distorted. Furthermore, miniaturization is a prevailing trend in satellites. Consequently, the batteries and solar panels cannot provide sufficient power for the CCDs, which have higher power supply voltage and power dissipation, to sustain long-term operation. Moreover, the CCDs lack random accessibility and system-on-chip capability. Therefore, based on these factors, TDI CCD sensors have been rendered unsuitable for small satellites. In an attempt to overcome the aforementioned shortcomings, this thesis primarily focuses on the TDI CMOS image sensors. The main contributions can be summarized as four aspects: (1) A TDI CMOS image sensor was proposed with dynamic pipeline adjacent pixel signal transfer. Following the operation of conventional TDI, the photo signal will be shifted stage by stage and a given photodiode will be reset by the previous-stage photo signal. Meanwhile, this TDI sensor can also configure effective TDI stages for dynamic range and signal-to-noise ratio optimization, as well as the signal transfer direction to compensate for the vibrations. (2) A TDI architecture with single-ended column-parallel signal accumulators was proposed. The TDI operation will be conducted by the off-pixel accumulators, whereby all the photo signals of each TDI stage will be read out after exposure. Accordingly, a 256×8-pixel prototype sensor was designed and fabricated. (3) An online deblurring (ODB) algorithm was proposed to address the blurred image issue caused by vibrations, which was subsequently developed into an 8-stage TDI CMOS image sensor with 256 column-parallel signal accumulators. The sensor can compensate for image shifts on the focal plane, enabling the production of sharper images even in scenarios involving complicated vibrations. (4) A two-step analog-to-digital convertor (ADC) prediction scheme was proposed for low-power CMOS image sensors. Based on the spatial likelihood of natural scenes, the prediction scheme identifies the most significant bits (MSBs) of a selected pixel using the quantization results of its neighboring pixels in the previous row, which enables a significant reduction in A/D conversion steps on MSBs and power consumption. A 384×256-pixel prototype chip was also developed to verify the scheme. DOCTOR OF PHILOSOPHY (EEE)

Measurement of human body motion has a myriad of applications ranging from gaming, rehabilitation, animation, virtual reality, sports science and surveillance. Existing methods of motion tracking include visual, mechanical, magnetic and inertial tracking. Visual methods require line of sight and suffer from the notorious occlusion problem. For the existing mechanical or inertial tracking methods, they have cumbersome wiring which hinder the natural movements. In this thesis, a wearable wireless sensor network using inertial/ magnetic sensors is developed to overcome the limitations of these existing methods. Using tri-axial accelerometer as the sole sensor will lead to singularity when the heading axis is vertical, restricting measurement of orientation to half a vertical plane. A new factorized quaternion approach is proposed in this research to overcome this deficiency with consideration of anatomical and sensor constraints. Different from the conventional approach based on single angle-axis quaternion, the proposed approach factorizes the quaternion into two principal axis quaternions corresponding to two equivalent arm motions. This allows for the implementation of anatomical arm constraints that match the range of arm motion and reduces the ambiguity in solutions. In addition, the singularities arising from the use of tri-axial accelerometers can be detected and resolved for a transient state. A novel algorithm based on elevation and heading angles is also proposed to determine the orientation of a sensor node equipped with tri-axial accelerometer, gyroscope and magnetometer. Compared to Euler angles, the fixed elevation and heading angles are independent on the temporal order of rotations. In addition, the fixed elevation and heading angles are observable and thus more intuitively visualized. External acceleration due to the rotation of the human arm segment is estimated using the gyroscope and compensated in the accelerometer output. As the magnetometer is easily affected by magnetic disturbances caused by electrical appliances and ferrous materials, it is decoupled from the accelerometer and used solely for the determination of heading. A new singularity avoidance method is also proposed to resolve the singularity problem that may arise, thereby allowing for measurement through all possible orientations. Experiments have been performed to evaluate the proposed wearable wireless sensor network in terms of accuracy, latency and power consumption. Experimental results show that the proposed factorized quaternion approach is able to resolve transient singularities and ambiguity arising from the use of triaxial accelerometer. Performance of the novel elevation and heading algorithm is also comparable to an indirect Kalman Filter at a reduced computational cost. The implementation of a singularity avoidance strategy makes the algorithm suitable for implementation in a fixed-point processor. Magnetic disturbances affect only the accuracy of the heading angle and not the elevation angles. DOCTOR OF PHILOSOPHY (EEE)

193 p. This dissertation proposes a novel maximum power point tracking (MPPT) algorithm that works optimally for a photovoltaic system under fast changing temperature and irradiation conditions. The new approach can effectively improve the tracking speed and accuracy simultaneously, which is the contradiction in conventional methods such as P&O or INC. This is particularly important for photovoltaic systems that frequently experience rapidly changing environmental conditions. MASTER OF ENGINEERING (EEE)

Due to the shared nature of the wireless channel in a random access network, the wireless nodes in a network naturally interact with each other and thus affect the throughput performance and energy consumption of each other. The aim of this thesis is to illustrate the use of local information for the traffic estimation and the throughput optimization in random access networks. Specifically, this thesis investigates various key issues on the design of random access networks, including transmission strategy in a slotted ALOHA network, distributed traffic characterization and throughput optimization in a CSMA network and optimum backoff protocol design in a non-persistent CSMA network. The traditional use of channel state information (CSI) in a slotted ALOHA network will lead to Braess Paradox, i.e., more information resulting in worse throughput performance. This thesis first proposes a new method to exploit the CSI named adaptive CSI method to eliminate the Braess Paradox and thus to improve the throughput performance. The use of acknowledgement is explored to estimate the transmission probabilities of other nodes in the network. It thus enables the distributed implementation of the proposed method. Moreover, this research generalizes two simple yet effective guidelines to determine the best transmission strategy under different system scenarios. Secondly, two novel distributed algorithms are proposed to estimate the on-going traffic in a heterogeneous CSMA network. In contrast to centralized Markov chain models, the proposed methods are able to characterize the traffic without prior information. They are also able to detect changes in the traffic. With the use of the traffic estimation method, a throughput adjustment algorithm is also proposed to tune the transmission rate of each node with respect to a desired throughput demand. Finally, the throughput performance of a non-persistent CSMA network under imperfect sensing is thoroughly examined. It is discovered that the traditional binary exponent backoff protocol accounts for the throughput inefficiency in CSMA networks. A novel backoff protocol named probabilistic backoff CSMA (PB-CSMA) is proposed, which is able to achieve optimum throughput performance regardless of the number of nodes as well as the level of sensing errors. Using the clear channel assessment results, the distributed implementation of PB-CSMA is also introduced. Simulation and experimental studies have been conducted to evaluate the proposed algorithms in terms of accuracy, convergence speed and computational complexity. The results show that with proper use of local information, the traffic intensity can be accurately estimated, the throughput can be significantly improved and the energy consumption can be further reduced for random access networks in a distributed manner. DOCTOR OF PHILOSOPHY (EEE)

144 p. This thesis addresses the control problems in linear motion system and parallel connected inverters system. Model Predictive Control (MPC) technique has been investigated to improve the control performance of inverters and linear motor drive. Based on the characteristics of the applications, suitable control structure and techniques are proposed and developed to solve specific problems. DOCTOR OF PHILOSOPHY (EEE)

148 p. Modern fountains are becoming more sophisticated as people are pursuing better entertaining effect. More sophisticated technology, especially in the field of motion control and network systems, is demanded by the fountain industry to further expand the capability and feature of a fountain. In particular, a reliable network system is desired to support the synchronous motion control of a large number of multi-axis motion systems. MASTER OF ENGINEERING (EEE)

86 p. This M.Sc. dissertation discusses digital control strategy for single-phase Pulse Width Modulated (PWM) voltage source inverters used in Uninterruptible Power Supplies (UPS) for single-unit and parallel-unit systems. Model Predictive Control is discussed for both the single inverter and parallel inverters. In this dissertation, the development of inverter system based on state-space model is first studied. Then MPC is used to implement the control strategy for single inverter and parallel inverters. The experimental setup includes a single inverter and two inverters in parallel using a switch and a debouncing circuit. Experimental results for open-loop and closed-loop single inverter system using a linear and a dynamic load is first presented. Then experimental results for two inverters in parallel are presented. Master of Science (Computer Control and Automation)

This dissertation presents the simulation study of using an ultra-wideband (UWB) radar for locating objects within or behind the wall. The UWB radars are useful in a number of applications and through-wall imaging is one of them. It can be used in search and rescue missions, behind the wall target detection, surveillance, object sensing through smoke and dust columns, etc. Master of Science (Computer Control and Automation)

Photovoltaic systems are useful in many applications around the world in many situations. The photovoltaic systems are comprised of photovoltaic cells, devices that convert light energy directly into electricity. The photovoltaic array or solar array absorbs energy from the sun and produces the output current. The solar array output current depends on the changing atmospheric conditions, the insolation level and temperature. In this dissertation, rapidly and slowly changing atmospheric conditions are used to simulate the array output current. Master of Science (Computer Control and Automation)

Radio frequency identification or RFID is a technique used to identify and track assets. The assets of interest depend on the industry in which it is used. It may be a rail car in the case of the railroad industry, a circuit board in a manufacturing plant or a passenger's baggage in the airline industry. In each case, knowing where the asset is of value due to its cost, customer service improvement possibilities, or impact on the process in which it is involved. RFID is a relatively new technology, but is finding many new applications in various industries. The technology has been recognized by a number of organizations through the establishment of standards. RFID technology holds promise for managing products through out their lifecycle. A typical RFID system comprises three elements: radio frequency transponders, a radio frequency reader unit, and a receiving and transmitting antenna. This project aims to create a RFID simulator. An easy to use graphical user interface is created using LabVIEW 7.0. Issues addressed include the simulation of virtual RFID tags, the RFID reader, protocols for communication between the tags and the reader. The process by which the tag modulates the carrier signal, and the algorithm for simultaneously receiving and decoding information from more than one tag are also studied and implemented. This LabVIEW-based simulator is ideal for studying and experimenting the passive RFID at 125kHz frequency. Master of Science (Computer Control and Automation)

The communications infrastructure that has become so much a part of daily life is expanding at an exponential rate. The advancement of network and storage technology has resulted increased usage and transportation of multimedia information which requires a high speed, high performance, flexible designs. Design of a customized base band signal processing and its implementation at silicon level is an achievable target these days using reconfigurable platforms. Master of Science (Computer Control and Automation)

Utilizing the field programmable gate arrays (FPGA) for a base band design to provide a unique solution for data packetizing (data link layer issues with the OSIreference model) and secured transmission of multimedia (audio) data regardless of the channel used for transmission. Master of Science (Computer Control and Automation)

Photovoltaic systems convert light energy into electricity. Photovoltaic cells or solar cells rely on the photovoltaic effect to absorb the energy of the sun and cause current to flow between two oppositely charge layers. A solar cell has a maximum power operating point (MPOP) which varies with changing atmospheric conditions, radiation and temperature. The current produced by a solar cell depends on the voltage, i.e. the current is the function of the voltage and therefore the generated power can be varied. Master of Science (Computer Control and Automation)

The objectives of this research are to investigate the various force sensors technologies, implement the printed circuit board insertion setup and investigate the issue of the force sensor’s noise and investigate the application of predictive filtering and the wavelet transform processing techniques. Master of Engineering

New methods of optimizing the performance of a brushless DC linear drive for high precision have been developed in this research. Model-based predictive control approach is first developed to control the position and the speed to track the reference profiles. The system performs well for movement that is above the millimeter range. Due to the frictional effect, the system fails in the micrometer range. Therefore, friction model has been included in the system model for optimization and compensation. Unlike most friction compensation approaches, that use the static friction model, this thesis uses the dynamic friction model that can describe both static and friction dynamics such as stick-slip, pre-sliding displacement etc. As the model is nonlinear, a new technique for identifying the LuGre friction parameters using the genetic algorithm is developed. This offline technique is superior than the conventional method as the global solution is obtainable. The mathematical derivation of the MPC controller gain vectors becomes impossible when the nonlinear friction model is included in the system. A new offline technique for optimizing the controller using the genetic algorithm has been developed. The experimental results have shown a significant improvement of the performance as compared to the conventional MPC approach especially in the micrometer range. Master of Engineering

In this thesis, a state space predictive control scheme to provide high performance and robust positioning of a brushless DC motor drive is proposed. In this scheme, a rational function weighting factor is introduced into the cost function of the predictive controller. Our studies indicate that this weighting factor can be tuned to improve the robustness of the servo drive in terms of larger inertia mismatch bound than the generalized predictive control scheme. Based on the proposed control scheme, the relationships between the controller parameters and the properties of the closed loop system, including settling time of step response and the inertia mismatch bound, are investigated. Some tuning rules of the controller are summarized. Master of Engineering

This thesis presents the development of two new multiobjective GA optimization techniques and its application to the modeling and controller tuning of the hard disk drives. For the drive modeling, two different approaches using black box and grey box modeling are presented. As multiobjective structural and parametric search is required for the modeling, we use the Hierarchical Genetic Algorithms (HGA) in conjunction with two newly developed multiobjective GA techniques. Master of Engineering

The permanent magnet synchronous motor (PMSM) is widely used in high precision servo control systems for its high efficiency and low power loss. Without the field windings, the power density is increased, while the size and weight of the machine are reduced. The elimination of mechanical contacts such as brushes and slips releases the limitation on the speed range and current capacity. This also eases the maintenance of the PMSM. In this thesis, we study the control and estimation techniques of the PMSM without mechanical sensors. Doctor of Philosophy (EEE)

Permanent Magnet Synchronous motors (PMSMs) have been used in a wide range of applications due to their high torque to inertia ratio, higher power density and efficiency. To drive the motor, the instantaneous rotor position is needed to perform electronic commutation. This is normally achieved by using position encoder or resolver mounted onto the motor shaft. Due to the disadvantages and limitations of the mechanical position sensors, there has been an increasing interest in developing techniques for eliminating the position sensors to improve reliability and to reduce the cost of the drive system. Master of Engineering

This thesis proposes a new control approach to control the position of a brushless DC motor drive using the state space generalized predictive control (GPC) algorithm. The basic idea of the GPC is to calculate a sequence of current and future control signals that minimises a cost function based on predictions of the plant output, setpoint and the available information. The aim of the thesis is to determine a tuning process for the application of GPC to the motor drive. Hence, the effect of the motor and the controller parameters on the performance and the robustness of the drive is examined. The study shows that the rise time and settling time of the servo system have an approximate linear relationship with the prediction horizon. Therefore, it is used to tune the controller of the drive. Also, the study shows that the control weighting factor can be used to smoothen the controller output. In motion control, the mismatch of the actual and nominal motor inertia readily affects the performance and robustness of the drive. In most applications, the exact load inertia is unknown. To design the controller, the study shows that with proper selection of prediction horizon and the nominal inertia used in the design, we can achieve a reasonable performance and robustness. Master of Engineering