Browsing by Author "Hinkkanen, Marko, Prof., Aalto University, Department of Electrical Engineering and Automation, Finland"

Filter results by typing the first few letters
Now showing 1 - 7 of 7
  • Bearingless Motors: Modeling and Control
    (2021) Sokolov, Maksim
     School of Electrical Engineering | Doctoral dissertation (article-based)
    Bearingless motors integrate the functions of an active magnetic bearing (AMB) and an electric motor in the same magnetic circuit, which is utilized both for driving the motor and for maintaining active levitation. This research focuses on two types of bearingless machines: rotating synchronous reluctance machines (SyRM) and linear flux-switching permanent-magnet (FSPM) machines. The complex nature of these devices results in non-trivial control challenges, which require accurate modeling of the magnetic behavior and of the force production of these machines. Particular attention in modeling is given to the effects of magnetic saturation and air gap variation. Both effects can result in degraded control performance or even instability if not accounted for. For bearingless SyRMs, an explicit-function-based magnetic model including cross-saturation is proposed. The model is able to predict radial forces even in a saturated machine. Furthermore, based on the textbook model of bearingless SyRMs, an improved model is developed that includes a more precise inverse air-gap approximation. The improved model has better accuracy in the predicted variation of forces and inductances due to eccentricity. For bearingless FSPM linear machines, a dynamic model based on an equivalent magnetic circuit is proposed, taking into account the effects of the saturation, the air gap variation, and the attraction force due to the permanent-magnet (PM) leakage flux. An analysis and characterization of the studied machines is conducted using the finite-element method (FEM). The proposed dynamic models are utilized as a basis for the development of model-based control systems. Classical state feedback control with direct pole placement is applied to bearingless machines. However, traditional current controllers cannot guarantee consistent performance in the presence of saturation and cross-coupling effects. These effects are automatically taken into account by the proposed state-space flux-linkage controller. A digital implementation of the controller is provided and robustness against the system parameter inaccuracies is analyzed. For levitation control, an observer-based state-space levitation controller is designed. Analytical tuning rules for each proposed controller are presented. Furthermore, feedback linearization is used for accurate calculation of current references based on the requested forces and torque. The applicability of the developed modeling and control methods is demonstrated with experimental results from three prototype bearingless machines including levitation, rotation, and propulsion tests.
  • Control Methods for Permanent-Magnet Synchronous Reluctance Motor Drives
    (2019) Awan, Hafiz Asad Ali
     School of Electrical Engineering | Doctoral dissertation (article-based)
    This thesis deals with control methods for synchronous motors with a magnetically salient rotor, such as the synchronous reluctance motors, interior permanent-magnet synchronous motors, and permanent-magnet synchronous reluctance motors. An exact hold-equivalent discrete-time motor model is developed. The motor model is then used in the design and analysis of current controllers and flux observers. A state-feedback current controller with an integral action and reference feedforward is designed directly in the discrete-time domain. The time delays are inherently taken into account in the design. The proposed current control design improves the dynamic performance and robustness especially at high stator frequencies. Furthermore, the saturation characteristics can be properly included in the controller. For sensorless control, a speed-adaptive full-order flux observer is designed and analyzed directly in the discrete-time domain. The proposed sensorless control system enables the operation at very low sampling to fundamental frequency ratios (below ten). For energy-efficient optimal control, a computation method for the control look-up tables is developed. When combined with an identification method for the magnetic model, the proposed method enables the plug-and-play startup of an unknown motor. The developed method is capable of producing optimal references along the maximum torque-per-ampere locus, at the current limit, at the maximum torque-per-volt limit, and in the field-weakening region. Apart from the current controllers, a feedback-linearization stator-flux-oriented control method and its systematic design procedure is developed. Stator-flux-oriented control enables the use of much simpler reference calculation methods. The simplicity of stator-flux-oriented control is tempting for many applications, while better control performance can be achieved with the proposed discrete-time control designs. The developed control methods can be applied in hybrid or electric vehicles, heavy-duty working machines, and industrial applications. The designed controllers and flux observers are experimentally evaluated using a 6.7-kW synchronous reluctance motor drive and a 2.2-kW interior permanent-magnet synchronous motor drive.
  • Dynamic Modeling and Standstill Identification for Induction Motor Drives
    (2022) Mölsä, Eemeli
     School of Electrical Engineering | Doctoral dissertation (article-based)
    This dissertation deals with dynamic modeling and standstill self-commissioning of a three-phase induction motor drive. Induction motor is the most popular rotating electrical machine, whose main benefits are simplicity and robustness. A variable-speed induction motor drive is not only suitable for for almost all fields of industry, but also for technologies including electric vehicles. An accurate and efficient control improves the overall efficiency of the drive and its application. High-performance control methods are often based on the dynamic model of the machine. As the electrical machines highly saturate during normal operation, the dynamic model must include the magnetic saturation. Objective of this dissertation is to develop a robust and reliable standstill identification method, which can be implemented into the standard frequency converter. A majority of the induction machines are low-power machines (< 100 kW). These machines are often equipped with closed rotor slots and deep rotor bars. The thin bridges closing the rotor slots saturate highly as a function of the rotor current. The impedance of the rotor bars also varies much as a function of the rotor current frequency due to the deep-bar effect. The standstill identification requires to use high-frequency excitations moving through the rotor. If this kind of excitations are used, the effects of slot-bridge saturation and deep rotor bars must be compensated for. To be able to develop a robust and reliable identification method, an extended dynamic model, which takes into account the above mentioned effects, is first developed. The model extensions can be plugged into a standard machine model and parametrized easily. The proposed model can also be applied to time-domain simulations, real-time control, and identification. The proposed identification method is based on an advanced model of a squirrel-cage induction motor. The model includes the deep-bar effect and the magnetic saturation characteristics. The excitation signals are fed to the stator using a standard inverter without compensating for its nonlinearities. The saturable stator inductance is first identified by means of a robust flux-integration test, during which unknown voltage disturbances are canceled with suitably selected current pulses. Then, the deep-bar characteristics are identified by means of a DC-biased sinusoidal excitation using different frequencies. Finally, the cross-saturation characteristics of the rotor leakage inductance are identified by altering the DC bias of the excitation signal. The identified characteristics are transformed to the parameters of the advanced motor model accounting for the interrelations of the above-mentioned phenomena. Since the physical phenomena affecting the standstill identification process are properly included in the identified model, less approximations are needed and more accurate parameter estimates are obtained. The designed model and the identification method are experimentally evaluated using 2.2-kW, 5.6-kW and 45-kW induction machines.
  • Estimation Methods for Grid-Voltage Sensorless Control of Converters Equipped with an LCL Filter
    (2016) Kukkola, Jarno
     School of Electrical Engineering | Doctoral dissertation (article-based)
    Grid-connected three-phase power electronic converters, equipped with an inductor-capacitor-inductor (LCL) filter, are widely used in energy production and consumption. The LCL filter effectively attenuates the switching harmonics of the converter; however, a drawback of the filter is its resonating behavior. The resonance can be damped by means of converter control, and the damping becomes easier and more effective, if the states of a dynamic full-order LCL-filter model are known (measured or estimated). The grid voltage is typically measured for converter control, but replacing voltage sensors with estimation may reduce system costs at low power ratings. Alternatively, a voltage estimation in parallel with the measurement increases system reliability. This thesis proposes estimation methods for grid-voltage sensorless control of converters equipped with an LCL filter. Only the converter AC currents and the DC-link voltage are measured for the control system. An adaptive observer is proposed for a combined state and grid-voltage estimation based on the full-order LCL-filter model. For unbalanced grid conditions, the observer is augmented with a disturbance model for the negative-sequence grid-voltage component. The nonlinear estimation-error dynamics of the observer are linearized and theoretically analyzed. The proposed observer is experimentally tested as a part of a grid-voltage sensorless control system, where the estimated states are applied in state-space current control. Based on the linearized dynamics, an analytic design procedure is presented for the observer in the continuous and discrete-time domains. The design procedure retains a link between the observer gains and dynamic performance, thus resulting in symbolic expressions for the gains as a function of the performance specifications and LCL-filter model parameters. The proposed observer can estimate the grid-voltage magnitude, frequency, and angle. In unbalanced grid conditions, the augmented observer can also estimate the negative-sequence component of the grid voltage. Estimated quantities can be used in the converter control system. The analytic design procedure enables the proposed estimation methods to be applied with different converters and LCL filters and further enables automatic tuning of the methods, for example, at the converter startup. The proposed methods can be applied, for example, in active-front-end rectifiers of motor drives or solar inverters.
  • Model-Based Position Estimation for Synchronous Reluctance Motor Drives
    (2014) Tuovinen, Toni
     School of Electrical Engineering | Doctoral dissertation (article-based)
    This thesis deals with model-based position-estimation methods for synchronous reluctance motor drives. The position estimation methods should be robust against modeling uncertainties, and reliable position information should be obtained at all speeds, including standstill. Since the considered position-estimation methods are model based, methods to obtain the model parameters under normal operation of the drive are proposed. Three observer structures are studied: a first-order observer, a reduced-order observer and a speed-adaptive full-order observer. The stability of the observers is studied via small-signal linearization under erroneous model parameters. Based on the analysis, robust gain selections are proposed with maximal tolerance for model parameter uncertainties. For improved performance, the observers are augmented with parameter adaptation laws. The first-order observer and the reduced-order observer are augmented with resistance adaptation laws at low speeds, using information from the back-electromotive force. The speed-adaptive full-order observer is augmented with inductance adaptation law at high speed, using information from the back-electromotive force. At low speeds, the speed-adaptive full-order observer is augmented with resistance adaptation law, using information from high-frequency signal injection. The stability of the augmented observers are studied via small-signal linearization under erroneous model parameters. Based on the analysis, stabilizing gain selections are proposed. The observers are experimentally evaluated using a 6.7-kW synchronous reluctance motor drive. The first-order observer seems unsuitable for these. The behaviour of the reduced-order observer is closely related to the behaviour of the speed-adaptive full-order observer, but the speed-adaptive full-order observer can be made less sensitive to parameter variations and measurement noise due to the additional degrees of freedom. When the speed-adaptive full-order observer is augmented with parameter-adaptation laws, small position-estimation error is obtained at all speeds, including standstill.
  • Modeling and Control of Bearingless Linear Motors
    (2024) Hosseinzadeh, Reza
     School of Electrical Engineering | Doctoral dissertation (article-based)
    This dissertation proposes dynamic modeling and model-based control design for bearingless linear motor systems. A bearingless motor produces both the traction and levitation forces using the same iron core. In bearingless linear systems, the magnetic levitation of the moving part (mover) renders the linear mechanical bearings redundant. However, controlling bearingless systems is more challenging than those with mechanical bearings. Conventional control methods, such as Proportional-Integral-Derivative (PID) controllers, can be employed, but their tuning method is often heuristic. Therefore, this dissertation considers model-based control methods. The prerequisite to this control design methodology is a sufficiently accurate dynamic model of the system. Thus, this dissertation considers two dynamic models for bearingless systems. A dynamic model is presented for a bearingless Linear Flux-Switching Permanent-Magnet (LFSPM) motor. This dynamic model is developed based on magnetic equivalent circuits, including the effects, such as magnetic saturation and airgap variation. For a double-sided bearingless LFSPM motor system, a state-feedback control method is presented. An approximate feedback linearization method is proposed to account for any nonlinear airgap dependency and the cross-coupling between the attraction and thrust forces. For energy efficient control, a resistive-loss minimization method is proposed. The minimization algorithm calculates the minimum reference currents for the given force references and the position of the mover. The results from the optimization method are used in the form of lookup tables and artificial neural networks. A comparison is provided between the two implementation methods. A dynamic model is developed for six-degree-of-freedom bearingless systems. The example system is a quadruple-sided bearingless linear motor system comprising eight three-phase motor units. To model the unbalanced magnetic torque, each motor unit is spatially divided into several submotors. The submotors of a motor unit have the same current, while the flux linkages and forces are different in a tilted position. The developed models can be utilized in time-domain simulations for system analysis and real-time control system development. The methods presented in this dissertation are evaluated using double-sided and quadruple-sided bearingless flux-switching permanent-magnet linear motors.
  • Weak-Grid Tolerant Control Methods for Voltage-Source Converters
    (2021) Rahman, F. M. Mahafugur
     School of Electrical Engineering | Doctoral dissertation (article-based)
    This thesis deals with weak-grid tolerant control methods for voltage-source converters. The converters suffer from stability problems when connected to weak grids. A grid is categorized as weak if it has a large impedance behind the point of common coupling. In this thesis, two different control methods, state-space current control and power-synchronization control, are studied focusing on weak grids. In the state-space current control, the converter is considered to be equipped with an LCL filter and connected to an inductive grid. The current controller is directly designed in the discrete-time domain. Either the converter or grid current is measured and other states needed by the current controller are estimated using an observer. The closed-form expressions of the discrete-time model are used to design and analyze the current controllers and state observers. The control design is based on the direct pole-placement method with the radial projection for the resonance damping of the LCL filter. The current controller provides good dynamic performance and robust operation against grid strength variations. Furthermore, the controller enables converter operation in a wide range of sampling frequencies. For unbalanced grid operations, the state-space controller is extended to double-frequency current control, during which both positive and negative sequences of the current can be simultaneously controlled. In the power-synchronization control, the converter is considered to be equipped with an L filter and the controller is designed in the continuous-time domain. The power-synchronization control allows robust stability irrespective of the grid strength and of the operating point. Robustness is obtained by analytic gain selections, whereby adequate stability margins are obtained. An enhancement of the power-synchronization control is realized by adding a reference feedforward of the active power, by which a fast dynamic performance can be achieved without step-response ringing and overshoot. Both the state-space current control and power-synchronization control are experimentally evaluated using a 12.5-kVA 50-Hz grid-connected converter. The control methods can be applied in industrial applications, for example, in solar PV and wind turbine systems, as well as the active front-ends of motor drives.