Browsing by Author "Westwick, David T."
Now showing 1 - 14 of 14
Results Per Page
Sort Options
Item Open Access Adaptive Backstepping Hybrid Force Position Control Free-Space and On Contact Operations(2020-05-27) Doctolero, Samuel; MacNab, Chris J. B.; Westwick, David T.; Goldsmith, Peter B.A set of adaptive hybrid force-position controllers are designed with the intent of operating in free space and on surface-contact without relying on a switching scheme. The first set of controllers create the framework for a backstepping adaptive hybrid force-position controller, and the second set take the framework further by designing it to be fully adaptive. On both sets, controllers were originally designed with a fully rigid robot manipulator but it is taken a step further by incorporating flexible joints in the system. In other words, a total of four hybrid force-position controllers are created. Controllers were designed using a Lyapunov stable backstepping approach, to ensure a Uniformly Ultimately Bounded (UUB) system. The proposed controllers utilize Cerebellar Model Articulation Controllers (CMACs) to adapt to unknown functions or dynamics, and do not require knowledge of the surface model, nor location. The controllers for the rigid case are verified via experimentation on a Quanser 2DOF serial manipulator, with an attached OptoForce 3 directional force sensor at the end-effector, and a foam as the surface to interact with. The controllers for the flexible jointed case are verified via simulations with a similar model as the Quanser manipulator and a nonlinear stiffness surface to interact with. The experiments and simulations test the effectiveness of the controllers by comparing them against traditional hybrid force-position schemes (stiffness or resolved acceleration). In terms of its robustness, controllers undergo additional payload to change the system dynamics and to asses how the robot behaves. Finally, adaptability is tested by letting the robot follow the same trajectory over multiple cycles and record its errors and neural network weights over time.Item Open Access Adaptive CMAC Neural Network Backstepping Control on a 2 DoF planar Manipulator in J-inverse Control Method(2020-09-25) Bonakdar, Iman; Pieper, Jeffery Kurt; MacNab, Chris J. B.; Goldsmith, Peter B.; Westwick, David T.Hybrid force/position control has been investigated in many studies in the presence of constraint compliance. Yet, the effect of compliance in the force sensor was left untouched. This study incorporates the force sensor model in the manipulator dynamic model and examines the impact of force sensor compliance on the system’s behavior. It has been shown in the literature that contact compliance and joint flexibility are the sources of instability in hybrid force/position J-inverse control. Nevertheless, using the hybrid force/position J-inverse control in the absence of these conditions causes instability in the system even far from the manipulator workspace singularities. This would lead us to suspect the compliance in the force sensor as a culprit. In this regard, we introduce an adaptive backstepping control that utilizes the Cerebellar Model Articulation Controller (CMAC) neural networks to handle model nonlinearities and parameter uncertainties. The backstepping method deals with force sensor compliance and constraint compliance by introducing virtual control signals. The model nonlinearities and unknown parameters are learned online by the controller with the neural networks’ help, and therefore, delivering an adaptive control design. By the employment of the Lyapunov theorem, we were able to derive the proposed controller and prove the stability of the system. The suggested controller has provided a better performance with respectively lower control effort in the J-inverse control law compared to the J-transpose control law.Item Open Access Decoupling Methods for the Identification of Polynomial Nonlinear Autoregressive Exogenous Input Models(2020-08) Karami, Kiana; Westwick, David T.; Behjat, Laleh; Nielsen, Jorgen; Norwicki, Edwin Peter; Bai, ErweiDeveloping a mathematical model of the system to be controlled is a significant part of a control design process, the more accurate the model, the more precise the control design can be. Since most of the systems around us behave nonlinearly, linear models are not always adequate. Therefore, nonlinear models should be considered. Many different model structures have been used in the literature such as the Volterra series, block-structured models, state-space representations, or nonlinear input-output models. Each of these structures has the ability to represent a large class of nonlinear systems but with its own drawbacks. Many are black-box modeling approaches and do not provide any intuition regarding the system. Many also suffer from the curse of dimensionality, becoming overly complex as the severity of the nonlinearity increases. It would be more practical if a nonlinear system can be approximated by a simpler model that is more accessible and understandable. During this research, the focus was on one of the widely used nonlinear input-output models known as the polynomial Nonlinear Autoregressive eXogenous input (NARX) model, as it represents a large class of nonlinear systems. A decoupling approach is proposed for polynomial NARX models. This technique replaces the multivariate polynomial that characterizes the NARX model with a decoupled model comprising a mixing matrix followed by a bank of univariate polynomials and a summation. While the proposed decoupling algorithm reduces the number of parameters significantly, performing the decoupling involves solving a non-convex optimization, which must be solved iteratively. Different initialization techniques are proposed for this optimization. In addition, identification algorithms are developed in both prediction error and simulation error minimization frameworks. The results of the decoupling approach are verified on two nonlinear identification benchmark problems and show promising outcomes since the number of parameters decreases significantly while the model accuracy remains high. Also, the decoupled model is capable of providing some insight into the identified model, as it is no longer a black-box.Item Open Access Development of Physics-Based Models of Lithium-ion Battery Energy Storage for Power System Techno-Economic Studies(2023-09-21) Vykhodtsev, Anton; Rosehart, William D.; Zareipour, Hamidreza; Thangadurai, Venkataraman; Westwick, David T.; Nielsen, Jorgen S.; Liang, HaoThe pathway to achieving a sustainable, low-carbon power system includes the widespread integration of energy storage to tackle intermittency of renewable energy sources and provide stability to the grid through various grid services. Among the wide range of stationary energy storage technologies available, the lithium-ion battery dominates the growth in installations throughout the world. Although lithium-ion battery energy storage systems are complex grid assets with nonlinear characteristics and lifespans that depend on operating conditions, the majority of economic assessments are conducted using a simple energy reservoir model that does not consider the physical processes occurring inside the lithium-ion battery storage. This thesis focuses on the development of physics-based models for lithium-ion battery energy storage in power system techno-economic studies. The aim of this work is to assist developers and investors in making better-informed decisions. In this work, modelling approaches used to represent lithium-ion battery energy storage in power system operation and planning studies are reviewed. The role of advanced models in enhancing the accuracy of economic evaluations and producing feasible schedules for battery storage providing transmission-level services is discussed. More importantly, this work proposes three physics-based mixed-integer models for battery energy storage for use in power system operation research studies. The first model is based on the single particle model and replicates the nonlinear operational characteristics of the battery. This model can be used for short-term operation studies. The second proposed model combines the widely-used energy reservoir model with the physical description of solid electrolyte interphase formation as a degradation mechanism. This model has been tested for long-term studies in both energy and power grid applications. Finally, the third proposed model is a data-driven model that accurately reproduces the degradation processes and nonlinear performance of the lithium-ion cell. The model facilitates long-term assessment of battery energy storage and effectively tracks both capacity and power fade over time. The results obtained from all the models are validated using the digital twin, which is based on the single particle model.Item Open Access Dynamic Eye-in-Hand Visual Servoing with Neural-Adaptive Backstepping(2019-03-22) Roy, Preston Logan George; Macnab, Chris J. B.; Nielsen, John; Goldsmith, Peter B.; Westwick, David T.This thesis investigates eye-in-hand visual servoing, where a camera on the robot arm provides information for the motor-control feedback loop. Current methods use a dual-loop strategy, where the outer-loop uses the visual servo error to compute desired joint velocities, while an inner-loop accomplishes the tracking. Since it is difficult to establish global stability with this strategy, this thesis instead investigates backstepping control. This provides a guarantee of uniformly ultimately bounded signals and explicitly accounts for the coupling between outer and inner loops. First a method with knowledge of the feature Jacobian is developed, then it is extended to an adaptive method that uses supervisory estimates of the feature Jacobian to maintain stable adaptation. The methods are further extended to the visual servo control of n-link robots, multiple features using a switched controller, and visual tracking. Non-linearities in the system are approximated using the computationally-efficient Cerebellar Model Articulation Controller neural network.Item Open Access Efficient Digital Predistortion for Next-Generation Wireless Systems Using Optimization and Signal Processing Techniques(2018-07-16) Abdelhafiz, Abubaker Hassan Babiker; Ghannouchi, Fadhel M.; Behjat, Laleh; Westwick, David T.; Helaoui, Mohamed; Sesay, Abu B.; Eriksson, ThomasAs consumers expect better and faster service and data-rates from wireless providers, the limits of existing wireless networks become more apparent. In response, the wireless communication industry steadily marches towards the next generation of network technology to meet this demand. The upcoming fifth generation (5G) of wireless networks promises connection speeds and data rates that are one hundred times faster than the existing networks, and vastly improved signal and connection quality. The path towards 5G however is not an easy one; as there are many significant challenges facing the design of radio frequency (RF) systems standing between the concepts of 5G, and its real-world implementation; such as the use of Massive Multiple-Input Multiple-Output systems, multi-band transmission and ultra-wideband signaling in the 5G standard. The move from the present fourth generation (4G) networks to 5G represents a fundamental paradigm shift in design methodology and theoretical concepts employed, which means that the current linearization and digital predistortion (DPD) techniques are not directly scalable to 5G. As such, a new generation of DPD methods suitable for the wireless networks of tomorrow and beyond is needed. The end goal of this thesis is to pave the road towards 5G by developing scalable and efficient DPD techniques applicable to 5G. This is achieved through approaching the three key aspects of 5G DPD and developing novel solutions for each: DPD model complexity reduction, MIMO DPD and multi-band DPD. Using mathematical knowledge, optimization techniques and signal processing methods, techniques for DPD complexity reduction, a scalable MIMO DPD that takes strong leakage between the different branches of MIMO transmitter into account, and a cost-effective technique for the mitigation of intermodulation distortion in multi-band systems are developed, and extensive measurements are performed to validate the theory for each of the various contributions. As a result, this work expands the horizon of knowledge in the three key areas of DPD research, and bringing the implementation of 5G RF one step closer to reality.Item Open Access Guidelines for Developing Low Frequency Model Equivalents in Emerging Electrical Grids(2018-12-06) Al-Eryani, Sameh Mohammed Anas; Zareipour, Hamidreza; Westwick, David T.; Nowicki, Edwin PeterDynamic model equivalents of power systems is an ongoing topic that continues to be relevant despite computing power advancements. Today's electrical grids are going through significant changes. The integration and installation of technologies that use power electronic devices pose a continuous need for stability studies in many fronts. Literature focuses on dynamic model equivalent techniques, their development and validation. However, the literature lacks comprehensive guidelines to produce equivalent models consistently. This thesis presents a general procedure and guidelines to develop low frequency dynamic model equivalents. The proposed procedure and guidelines are aimed at developing consistent and reliable low frequency model equivalents. The guidelines will be demonstrated on a test system to validate the recommendations and show the impact of not following a consistent methodology in developing the equivalent. Finally, the procedure and guidelines will be applied on Alberta Interconnected Electric System to demonstrate the application on a real system model.Item Open Access Methodology of Robot-Assisted Tool Manipulation for Virtual Reality Based Dissection(2019-03-29) Trejo Torres, Fernando Javier; Hu, Yaoping; Sesay, Abu B.; Westwick, David T.; Chan, Sonny; Liu, Peter J.Robot-assisted (RA) surgery employs a master-slave system, in which a surgeon's hand manoeuvres the stylus of a hand controller (master) mapped at the operation site to indirectly manipulate a surgical tool attached to the end-effector of a robot (slave). Hence, RA surgery has two drawbacks. Firstly, the transfer of tool-tissue interaction forces to a surgeon is either absent or inaccurate. Secondly, RA surgery incorporates motion coupling (MC) and motion coupling plus orientation match (MC+OM) as indirect modes of tool manipulation, which disregard a pose (position and orientation) match (PM) between the mapped stylus and the tool. This may cause inadvertent tissue trauma during tasks like dissection, which spends ~35.0% of surgery time. Due to the potential of virtual reality (VR) based surgical training, this thesis presents a methodology to address the drawbacks on a VR simulator of soft-tissue dissection. The methodology comprises the formulations and evaluations of an analytic model that estimates dissection forces; and a PM algorithm. The simulator interfaced with the haptic device PHANToM Premium 1.5/6DOF (as a hand controller) to deliver the model forces, and incorporated the kinematics of the device and neuroArm (a neurosurgery robot) for the PM algorithm. The evaluation of the model for estimating dissection forces collected at the tool speeds of 0.10, 1.27, and 2.54 cm/s indicated a force estimation > 80.0%, a computation time < 1.0 ms (the device's update period), and a bandwidth < 30.0 Hz (the device's bandwidth). Moreover, the model lessened cognitive workload for dissections executed at 0.10 cm/s. The evaluation of the PM algorithm revealed a position match < 30.0 µm (the position resolution of the device and neuroArm), an orientation match < 10.0° (to minimize the surgeon's disorientation), and a computation time < 500.0 µs (a half of the device's update period). Additionally, the algorithm became useful to maintain an accurate tool speed and reduce tissue trauma for dissections performed at 0.10 cm/s. The outcomes imply the suitability of the methodology for VR-based RA dissection and their potential to suggest guidelines for VR-based RA dissection training.Item Open Access Multi-objective optimization using evolutionary algorithms: Application to the control of flow past a circular cylinder(2018-11-22) Bingham, Conrad Cole; Martinuzzi, Robert John; Morton, Chris R.; Hu, Yaoping; Ziadé, Paul; Westwick, David T.; Epstein, Marcelo D.Modifications to the vortex shedding dynamics from a circular cylinder of diameter D are investigated experimentally in a free surface water channel. The vortex shedding is modified via the placement of a control cylinder of diameter \textit{D}/8 in the vicinity of the main cylinder. A methodology is presented to link changes in the wake dynamics and loading on the main cylinder. The analysis combines Fourier Modal Decomposition, Proper Orthogonal Decomposition, and phase averaging. Based on differences in the wake dynamics, the influence of the control cylinder can be classified according to its placement: (i) in the free stream outside the main cylinder shear layer; (ii) within the main cylinder shear layer; and (iii) in the recirculation region. While fluctuating lift is significantly reduced in all cases, the mean and fluctuating drag are affected differently. A generalized model-free method to optimize parameters for open-loop and closed-loop control in fluid mechanics applications is then presented. A multi-objective evolutionary algorithm (MOEA) is employed to minimize the oscillating lift caused by vortex shedding from the main cylinder. The control cylinder is prescribed a position as well as a periodic motion in two dimensions. The MOEA efficiently handles the larger optimization parameter space. The first objective of the algorithm is to minimize the fluctuating force coefficient $C_{L_{RMS}}$, while the second objective is to minimize of the actuation power required to drive the control cylinder. The final solution suppresses $C_{L_{RMS}}$ by over 90\% using near-zero actuation power. Further, the MOEA automatically provides a sensitivity study as to the influence of the different parameters and also in which spatial area the greatest influence is expressed.Item Open Access Nonlinear MPC tracking control and set point control for wastewater treatment processes(2019-01-21) Sadeghassadi, Mahsa; Macnab, Chris J. B.; Westwick, David T.; De la Hoz Siegler, Hector; Nielsen, John; Trifkovic, Milana; Ray, Ajay KumarThis thesis concerns the design of feedback controls of a biological wastewater treatment plant (BWT), specifically the benchmark simulation model number 1, and methods for determining optimal set points. In BWT, biological organisms remove unwanted substances including nitrogen, ammonia, and organic material. The feedback controls can manipulate aeration and flow rates in order to control the dissolved oxygen concentration and nitrite/nitrate concentrations. The most basic function of the feedback controls is to ensure that effluent quality meets a pre-determined environmental standard in an energy-efficient manner. Identifying optimal set points can be as important, or more important, in reducing the contaminants/cost as the feedback/feedforward strategy used to track the set point. Thus, choosing an appropriate nitrate/nitrite and oxygen set point, and then maintaining the set point, defines the important objectives of the current work. Several novel methods are developed and compared with a PI control. Initially, Lyapunov-based adaptive controllers with fuzzy set point regulators are designed for both loops. Compared with the existing methods, the proposed methods demonstrate great potential for improving system performance. Moreover, switching techniques on an external carbon source input are proposed to prevent the risk of too much or too little food and/or too little dissolved oxygen. Then, design of the dissolved oxygen (DO) variable set point is presented in parallel to the DO set point tracking control, based on Artificial Neural Network (ANN) models used for set point design and for prediction within the DO Neural Networks Model Predictive Control (NNMPC) algorithm. The solution of an offline multi-objective optimization problem during the first two days of dry weather conditions is used as the initial set point, and then changes in the moving direction provided by an ANN model. Compared with the existing methods, the proposed method shows ability of reducing the effluent quality and the operational cost simultaneously. Next, a single-optimization problem along with an ANN model designs the nitrate/nitrite set point in order to reduce violations in the ammonium and nitrogen limits. The results prove the near-complete removal of violations by using the proposed method. The method in the last chapter includes a way to adjust set point to respond to varying conditions and a model predictive control scheme, which utilizes a cerebellar model arithmetic computer (CMAC). This controller is an adaptive one, since our model used in the MPC updates on-line and in real-time and can thus change due to unknown and changing dynamics. This technique avoids the need for any a-priori estimation step. The CMAC controller learn the desired control signal in a Lyapunov-stable scheme, which provides a guarantee of uniformly ultimately bounded signals.Item Open Access On Impact of Load Modelling on Motor-Start Simulation of Large Power Systems(2019-01-02) Chih, Edmond Kamhoi; Zareipour, Hamidreza; Westwick, David T.; Pahlevani, Majid; Nowicki, Peter P.Dynamic load models are important in the planning and the operation of power systems, as they replicate the responses of electrical devices in power system simulations. One of these load models that is used is the CLOD model. The parameters used for this model are typical assumptions based on typical system conditions and studies. However, with the development and integration of new technologies at load sites, these assumptions that have prevailed to date may no longer be suitable. In addition, existing literature does very little in discussing how applying this model impacts the outcome of motor start simulations. Thus, the objective of this research is to investigate how using the CLOD model with different parameters impacts the outcome of the dynamic motor-start simulations on the Alberta Interconnected Electrical System.Item Open Access On Verification of Load Models for Frequency Response Using PMU Data(2020-04-18) Fung, Sin Yee; Zareipour, Hamidreza; Westwick, David T.; Nowicki, Edwin Peter; Sesay, Abu B.During frequency excursions, not only generators respond to the frequency deviation. Most electric loads are sensitive to large variations in frequency that change their power consumption as a result. Therefore, load modeling is a crucial element in power system simulation, and a better understanding of load response in frequency excursion events is necessary. The main objective of this thesis is to investigate the performance of available load models used in the industry to represent the frequency response of system load elements during frequency excursion events in power system simulations. A dynamic simulation approach is proposed to investigate and evaluate the frequency response of an industrial load simulated with different load models recommended in the industry based on real-life frequency excursions. PMU data collected in Alberta’s grid is used for the scope of studies. Load models explored include CLOD, ZIP and exponential load models. Furthermore, measurement-based load modeling is defined to optimize the load model parameters to improve the accuracy of load response for frequency excursion studies. The performances of these optimized models are compared. The results provide a better understanding of load response and guidelines for choosing load models to be used in frequency excursion studies, to make better judgements in power system analysis and planning in terms of power system security and required balancing resources.Item Open Access Real-time Collision Detection Algorithm for Humanoid Robots(2020-01-29) Moghaddasi, Shadi; Ramírez-Serrano, Alejandro; Westwick, David T.; Li, LepingHumanoid robots (humanoids) are highly capable of assisting humans and working with them in cluttered and confined environments. However, they are not completely ready to work in close proximity with humans while not risking the safety of themselves and the objects and people around them. Current methods have not been fully successful in preparing humanoid for safe Human Robot Interaction (HRI) because they rely on expensive and fragile equipment, and erroneous techniques. This thesis presents a novel real-time methodology that enables the safe close proximity HRI for all types of humanoids (controlling systems, etc.). The proposed approach employs signals from robots’ motor joints and data from the computers running the robot to develop a collision detection algorithm. Using this algorithm, humanoids will be able to speedily identify impacted joints during a collision. Experimental results for the humanoid robot Taiko are presented to demonstrate the applicability of the proposed approach.Item Open Access Robust Neural-Adaptive and Fuzzy-Adaptive Control of Flexible-Joint Robots with a Cerebellar Model Arithmetic Computer(2019-01-24) Razmi, Mohammadsaleh; Macnab, Chris J. B.; Pieper, Jeffery Kurt; Westwick, David T.The control of non-linear, non-minimum phase systems with a high degree of flexibility require a robust, stable controller. This thesis further develops and tests two recently proposed robust modifications of a weight update rule for direct adaptive control using the Cerebellar Model Articulation Controller (CMAC). Previous designs for robust adaptive CMAC control can stop weight growth before bursting can happen but in a very conservative way, which usually sacrifices so much performance that the method is no longer desirable. The main purpose of this thesis is to determine how to guarantee uniformly ultimately bound signals in adaptive non-linear control systems that have high performance. To ensure the validity of these methods, a Two-Link Flexible-Joints Robot (TLFJR) is utilized experimentally. TLFJR is a highly non-linear system which is prone to instability. Hence our novel robust controllers are important in the control of this system. Using our novel methods, TLFJR with added sinusoidal disturbance is controlled with a guarantee of both stability and performance. These techniques consists of two CMACs, namely robust and performance, working in parallel in a backsteppable control input form. Backstepping is used to handle all non-linearities of system with the availability of just one input. The robust controller is a CMAC with traditional weight update rules; here we use an e-modification with a conservative value of v. The performance CMAC is designed based on a novel stable weight update modification methods: the weight smoothing method(chapter 3) and a near-optimal method (chapter 4). When these methods are applied to the TLFJR experimentally, the results (chapter 5) show that the controller can stabilize the TLFJR with a guarantee of performance, as well.