WO2015028242A2 - Contrôle prédictif par modèles d'un système électrique - Google Patents

Contrôle prédictif par modèles d'un système électrique Download PDF

Info

Publication number
WO2015028242A2
WO2015028242A2 PCT/EP2014/066420 EP2014066420W WO2015028242A2 WO 2015028242 A2 WO2015028242 A2 WO 2015028242A2 EP 2014066420 W EP2014066420 W EP 2014066420W WO 2015028242 A2 WO2015028242 A2 WO 2015028242A2
Authority
WO
WIPO (PCT)
Prior art keywords
electrical
future
electrical system
inverter
rectifier
Prior art date
Application number
PCT/EP2014/066420
Other languages
English (en)
Other versions
WO2015028242A3 (fr
Inventor
Peter AL-HOKAYEM
Thomas BESSELMANN
Stefan ALMER
Tobias Geyer
Nikolaos Oikonomou
Joachim FERREAU
Original Assignee
Abb Technology Ag
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Abb Technology Ag filed Critical Abb Technology Ag
Publication of WO2015028242A2 publication Critical patent/WO2015028242A2/fr
Publication of WO2015028242A3 publication Critical patent/WO2015028242A3/fr

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P23/00—Arrangements or methods for the control of AC motors characterised by a control method other than vector control
    • H02P23/12—Observer control, e.g. using Luenberger observers or Kalman filters
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/048—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators using a predictor
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
    • H02M5/00—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases
    • H02M5/40—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases with intermediate conversion into dc
    • H02M5/42—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases with intermediate conversion into dc by static converters
    • H02M5/44—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases with intermediate conversion into dc by static converters using discharge tubes or semiconductor devices to convert the intermediate dc into ac
    • H02M5/453—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases with intermediate conversion into dc by static converters using discharge tubes or semiconductor devices to convert the intermediate dc into ac using devices of a triode or transistor type requiring continuous application of a control signal
    • H02M5/458—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases with intermediate conversion into dc by static converters using discharge tubes or semiconductor devices to convert the intermediate dc into ac using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only
    • H02M5/4585—Conversion of ac power input into ac power output, e.g. for change of voltage, for change of frequency, for change of number of phases with intermediate conversion into dc by static converters using discharge tubes or semiconductor devices to convert the intermediate dc into ac using devices of a triode or transistor type requiring continuous application of a control signal using semiconductor devices only having a rectifier with controlled elements
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P27/00—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage
    • H02P27/04—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage
    • H02P27/06—Arrangements or methods for the control of AC motors characterised by the kind of supply voltage using variable-frequency supply voltage, e.g. inverter or converter supply voltage using dc to ac converters or inverters
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
    • H02M1/00—Details of apparatus for conversion
    • H02M1/0067—Converter structures employing plural converter units, other than for parallel operation of the units on a single load

Definitions

  • the invention relates to the field of control of electrical equipment.
  • the invention relates to a method, a computer program, a computer-readable medium and a controller for controlling an electrical system as well as to an electrical system controlled by such a controller.
  • An electrical system as described herein may be adapted for transforming a first AC current into a second AC current to be supplied to a rotating electrical machine such as a motor or a generator, for example such as an asynchronous machine, a synchronous machine or a doubly-fed machine.
  • the electrical system comprises a converter with an active rectifier for transforming the first AC current / voltage from an electrical grid into a DC current and an inverter for transforming the DC current into the second AC current.
  • the converter is typically employed to transform an AC current of fixed frequency into an AC current of varying frequency, or vice versa.
  • the power flow is reversed and the varying frequency AC current of the machine is rectified to a DC current and subsequently inverted into a fixed frequency AC current of the grid.
  • the AC power of fixed frequency is provided by an electric grid, while the AC power of varying frequency is used to drive the electric AC machine.
  • a torque reference provided by a speed controller is typically transferred into a current control reference to be controlled by a current controller.
  • the current controller determines firing angles for the rectifier which are translated into switching instances in a first modulator.
  • firing angles of the inverter are determined in a separate control loop by means of a power factor controller, whose task it is to ensure reliable operation of the inverter and close to unity power factor in the stator windings, depending on the state of the drive.
  • a second modulator similar to the first modulator for the rectifier translates the inverter firing angles into switching instances of the semiconductors.
  • the rectifier usually plays a crucial role in controlling the torque of the AC machine, and is thus integrated in many control approaches.
  • a separation is typically taking place and separate SISO (single-input-single-output) control loops are used for the rectifier and the inverter.
  • SISO single-input-single-output
  • switching instances of the inverter are determined by an upper controller (e.g. torque and or flux controller) and a modulator, which generates the switching instants from signals provided by the upper controller.
  • the rectifier on the other hand is used to provide a stable DC link voltage, and is controlled by a second upper controller (e.g. voltage and current controller) and a second modulator to determine the switching instances for the rectifier.
  • a second upper controller e.g. voltage and current controller
  • standard control approaches for voltage-source inverters usually split the control problem into two control problems for the rectifier and the inverter, assuming that the DC link acts as a strong storage element that facilitates this separation. However, when the DC link capacitors are small, this separation may not hold.
  • an excitation control loop may be provided in the case of a synchronous machine.
  • the excitation control loop is an additional, essentially decoupled control loop.
  • Model predictive control is used to control inverters. For example, model predictive direct torque control as described in EP 2 348 631 A1 is used for controlling the electric machine and focuses on the inverter side and the Neutral Point control problem while a substantially stiff full DC link voltage is assumed.
  • An aspect of the invention relates to a method for controlling an electrical system.
  • the electrical system comprises an electrical converter for interconnecting an electrical source (such as a grid or a generator) with an electrical load (such as at least one rotating electrical machine).
  • the electrical converter (eventually together with its control system) may be seen as an electrical drive for driving the electrical rotating machine, which may be an electrical generator or an electrical motor.
  • the electrical converter comprises at least one active rectifier and at least one inverter, i.e. may be adapted for transforming a first AC current from the electrical grid into a DC current and the DC current into a second AC current of different, variable frequency to be supplied to the electrical machine and/or vice versa.
  • Both the rectifier and the inverter may comprise semiconductor switches, which are used to control the current flowing through the converter.
  • the semiconductor switches may be controlled by a control system providing switching instants (i.e. signals whether a specific semiconductor switch should be opened or closed).
  • the converter, rectifier, inverter, electrical machine, etc. may be high power devices, i.e. devices that are adapted for processing current of more than 100 A and/or voltages of more than 1000 V.
  • the method comprises the steps of: receiving a set point (torque and/or flux) reference for the load and a latest estimated state for the converter, wherein the estimated state comprises estimated is computed based on measurement values of the electrical system; predicting a sequence of future states and future inputs over a horizon of at least two time instants with a mathematical model of the electrical system, wherein the future inputs comprises future rectifier reference values for the at least one rectifier and future inverter reference values for the at least one inverter and wherein the sequence of future states is predicted with a mathematical model of the electrical system by minimizing a cost function that may reflects the performance, the mathematical model being based on a number of differential and/or difference equations relating to a state of the electrical system with a future state of the electrical system; and selecting future rectifier reference values to be input to a modulator of at least one rectifier and future inverter reference values to be input to the a modulator of the at least one inverter by selecting at least one future time instant
  • the common controller determines reference values for both the rectifier and the inverter, which by respective modulators may be translated into switching instants for the rectifier and the inverter.
  • the rectifier and/or inverter reference values may be firing angle, modulation indices, reference fluxes and/or reference voltages.
  • the common controller may compute continuous control variables or reference values (e.g. modulation indices or firing angles of rectifier and inverter), which is forwarded to modulators which again determine the exact switching times of the semiconductors in the rectifier and inverter.
  • the mathematical model may encode (i) the average behaviour of rectifier and inverter, (ii) the machine dynamics, (iii) the DC link dynamics and optionally (iv) the grid dynamics, (v) the excitation of the electrical machine and (vi) drive shaft dynamics.
  • the future states are determined such that a cost function is minimized, for example a quadratic function that is quadratic in the difference between the future states and reference states.
  • the model is used to predict the reaction of the electrical system to changes in the control input, and selects the control input which minimizes a constrained finite time optimal control problem.
  • the future states comprise future reference values for the inverter and the rectifier and the reference values for the next future time instant (or the next two, three or more time instants) are supplied to the modulators, which determines the switching instants for the next or future time instant.
  • the impact of control actions on the rectifier, the DC link, and the inverter may be predicted simultaneously, and thereby allowing for better coordination of the control actions and thus an improved control performance of the electric system. Disturbances coming from the mechanical side (electrical machine) or from the grid side may be handled more effectively.
  • the torque or current control layer may be replaced by a common controller controlling the air gap torque by means of a model predictive control approach.
  • the control method may be applied to different topologies of converters.
  • the topology may be modelled in one common model, which also may take the physical interactions of the different subsystems such as rectifiers, inverters, DC links, etc. into account.
  • the model of the electrical system comprises a model of the at least one active rectifier and a model of the at least one inverter.
  • the interaction of the inverter and the rectifier may be encoded into the model by corresponding equations.
  • the model of the electrical system comprises a model of the electrical grid. Also the behaviour of the electrical grid and/or at least a filter and/or a transformer at the connection point to the grid or the electrical load may be encoded into the model.
  • the model of the electrical system comprises a model of the at least one rotating electrical machine.
  • the behaviour of the rotor may be encoded into the model and, for example, the degree of excitation of the electrical machine may be controlled.
  • the electrical system comprises at least two rotating electrical machines and the model of the electrical system comprises a model of the at least two electrical machines.
  • the backreaction and/or the mutual influence of electrical machines rotating with different speeds may be encoded into the model.
  • the electrical converter comprises a voltage-source converter (which comprises a DC link with capacitors as energy storage) and the future rectifier reference values comprise modulation indices for the at least one active rectifier and the future inverter reference values comprise modulation indices for the at least one inverter.
  • the common controller may generate modulation indices for both the rectifier and the inverter based on a model including the rectifier, the inverter and optional the DC link. Modulation indices may indicate that the rectifier and/or inverter should output a specific voltage at a specific time instant.
  • the electrical converter comprises a current-source converter (which comprise a DC link with inductors as energy storage) and the future rectifier reference values comprise firing angles for the at least one active rectifier and the future inverter reference values comprise firing angles for the at least one inverter.
  • the common controller may generate firing angles for both the rectifier and the inverter based on a model including the rectifier, the inverter and optional the DC link. Firing angles may indicate, when (at least one phase) should be switched to specific voltage with respect to the period of the frequency of the respective voltage and/or current.
  • the method further comprises the steps of: determining future switching states for the at least one active rectifier from the selected future future rectifier reference values and/or determining future switching states for the at least one inverter from the selected future inverter reference values.
  • the control system may comprise modulators for each of the rectifiers and/or inverters that translate the respective reference values into switching instants for the respective rectifier and/or inverter.
  • the model is a linearized model of the electrical system and the linearized model comprises linear equations derived from non- linear differential equations modelling the electrical system.
  • the model may comprise a number of non-linear first order differential and possibly algebraic equations that describe the physical behaviour of the electrical system. Those equations may be cast into linear equations relating states and reference values (inputs to the system) at a first time instant into states and reference values at a future time instant. I.e., the linear equations may be discrete-time equations.
  • these equations may be solved by a solver of the common controller based on the latest estimated states, which tries to minimize the cost function while generating a solution to the linear equations.
  • the model comprises constraints on future states and/or future inputs of the electrical system.
  • constraints for example, may comprise upper and lower limits on physical quantities like voltages, currents, torques, etc.
  • the cost function may be a quadratic function of the future states and/or future inputs.
  • the cost function may be encoded with a matrix multiplied from each side with vectors containing a difference between the future states and reference states.
  • linearized equations together with constraints and a quadratic cost function may result in a quadratic programming problem that may be solved with standard methods.
  • the measurement values are measured currents and/or voltages of the electrical system and/or estimated values of the estimated state comprising at least one of a current, voltage, flux, power and torque of the electrical system.
  • the estimated state i.e. the output of the electrical system
  • the control system may comprise a special subcontroller called estimator for determining the estimated state.
  • control system and/or the common controller may comprise a processor and a non-volatile memory, in which the computer program for executing the method is stored.
  • a computer-readable medium may be a floppy disk, a hard disk, an USB (Universal Serial Bus) storage device, a RAM (Random Access Memory), a ROM (Read Only Memory) and an EPROM (Erasable Programmable Read Only Memory).
  • a computer-readable medium may also be a data communication network, e.g. the Internet, which allows downloading a program code.
  • a further aspect of the invention relates to an electrical system with an electrical converter for interconnecting an electrical source with at least one rotating electrical machine, wherein the electrical converter comprising at least one active rectifier and at least one inverter; and with a controller for controlling the electrical converter as described above and in the following.
  • the method and/or the common controller may have the following benefits.
  • the control actions of rectifier and inverter may be coordinated. This coordination may result in a better transient behaviour, and better ride-through capabilities, thereby reducing the number of trips due to voltage dips and thus increasing the reliability of the system.
  • the method supports the definition of constraints on system variables, e.g. to limit the currents or voltages at the DC link.
  • the method may take the dynamics and natural frequencies of the electric drive and the electrical machine into account to dampen such oscillations and even pro-actively prevents these frequencies from being excited by means of the control inputs.
  • the predictive control approach may allow to predict the behaviour of the dynamic system before it occurs. If an event such as a constraint violation is predicted, countermeasures may be taken before this event occurs, resulting in an improved control performance.
  • the flexibility of the method allows the use of a nonlinear model of the electric drive, and to take process dynamics or dynamics of the grid into account.
  • the operation of an electrical system with an electrical machine may include a number of requirements: high dynamic performance during transients, grid-friendly operation (compliance with grid codes), machine- and process-friendly operation, robustness with respect to grid imbalances, weak grid, voltage dips, load changes and/or tripless operation of the full converter 12 under various load conditions. All these requirements may be met with the control method as described above and in the following.
  • the method is capable of controlling the air gap torque applied by the electrical machine to the machinery (such as a pump or turbine), the degree of magnetization of the electrical machine, the DC-link voltage or DC-link current and/or the interface quantities (flux, current, voltage, reactive power) with the electrical grid.
  • the machinery such as a pump or turbine
  • the degree of magnetization of the electrical machine the DC-link voltage or DC-link current and/or the interface quantities (flux, current, voltage, reactive power) with the electrical grid.
  • Fig. 1 schematically shows an electrical system according to an embodiment of the invention.
  • Fig. 2 schematically shows a part of an electrical system according to a further embodiment of the invention.
  • Fig. 3 schematically shows a part of an electrical system according to a further embodiment of the invention.
  • Fig. 4 schematically shows a part of an electrical system according to a further embodiment of the invention.
  • Fig. 5 shows a flow diagram for a method for controlling an electrical system according to an embodiment of the invention.
  • Fig. 1 shows an electrical system 10 with an electrical converter/drive 12 that interconnects an electrical grid 14 with an electrical machine 16.
  • the converter 12 comprises three basic elements: An active rectifier 18, also called ARU (Active Rectifier Unit) or AFE (Active Front End), a DC link 22, also called CBU (Capacitor Bank Unit), and an inverter 20, also called INU (Inverter Unit).
  • ARU Active Rectifier Unit
  • AFE Active Front End
  • DC link 22 also called CBU (Capacitor Bank Unit
  • INU Inverter Unit
  • the rectifier 18 is connected to the three-phase AC grid 14, e.g. by means of a transformer or a Direct-To-Line configuration in which a large filter and common mode choke are inserted.
  • the rectifier 18 is electrically connected to the DC link 22, which again is electrically connected to the inverter 20.
  • the inverter 20, and thus the electric converter 12 is connected to the electrical machine 16, for example a motor or generator.
  • the depicted topology is only one possible variant.
  • the connections between the described elements may vary. For instance, instead of single three-phase connections, dual three-phase or multiple three-phase connections are common.
  • the DC link 22 may be connected as a two-port network, or in other configurations.
  • connection between the converter 12 and the grid 14 may also encompass a transformer and/or different filters.
  • the connection between the converter 12 and the electrical machine 16 may also encompass filters. Both connections may comprise long cables, inducing additional dynamics to the electrical system 10.
  • the topology of the three basic elements 18, 20, 22 may also differ greatly, be it in the type of elements (e.g. capacitive or inductive DC link 22 IGBTs, IGCTs, or thyristors as switches in the rectifier 18 and inverter 20), the topology of the elements (e.g. half bridges versus full bridges), or the number of elements (multiple DC links, multiple rectifiers or inverters). All the described elements (cables, filters, transformers, etc.) as well as combination thereof may be modelled into the mathematical model as described below.
  • the electrical system 10 and in particular the converter 12 is controlled by a control system 24 that receives measured values 26 from converter 12 and measured values 28 from electrical machine 16.
  • the measured values 26 may comprise a DC link voltage, switching states of the rectifier 18 and/or inverter 20, and in general voltages and currents of the different components of the converter 12.
  • the measured values 28 may comprise voltages and/or currents in the stator and/or rotor of the electrical machine 16 and mechanical quantities such as the speed of the electrical machine 16.
  • a reference speed 30 is input to the control system 10, which reference speed 30 may be provided from a superordinated control system.
  • the control system 24 outputs switching states 32 for rectifier 18 and switching states 34 for inverter 20, which are predicted with a model predicted control approach from the inputs 26, 28 and 30 as described in the above and in the following.
  • the control system 24 has the task to control the switching elements of both the rectifier 18 and the inverter 20. in order to operate the electrical machine 16 reliably and efficiently. Emphasis should be put on the coordination of the actions of rectifier 18 and inverter 20.
  • the state of the DC link 22 may depend on the power electronic components on both sides, and/or a change of the switching pattern on one side may require immediate response in the switching pattern of the other side. In fact, most DC links 22 are undersized to reduce manufacturing costs, and as such the stored DC energy may be depleted within the range of a few milliseconds, if the control actions of rectifier 18 and inverter 20 are not coordinated.
  • the converter 12 may be faced with external disturbances from the grid side 14 or the machine side 16. A dip in the grid voltage or a sudden load change is best handled with immediate coordinated control action of both rectifier 18 and inverter 20.
  • control system 24 a possible setup of the control system 24 is described.
  • the components of the control system may be at least partially implemented in hardware or in software.
  • the components of the control system 24 are software modules running in one or more processors provided by the control system 24.
  • the control system 24 comprises a state estimator 36, which estimates an latest estimated state 38 and a latest speed 40 from the measured values 26, 28.
  • the latest speed 40 is supplied to a speed controller 42, which generates a torque reference 44 from the speed reference 30 and the latest speed 40.
  • the (air gap) torque reference 44 may also be provided externally.
  • Both the latest state 38 and the torque reference 44 are input to a common controller 46, which generates rectifier reference values 48 input to a rectifier modulator 50 and inverter reference values 52 input to an inverter modulator 54.
  • the modulator 50 generates the switching instant 32 for the rectifier 18 and the modulator 54 generates the switching instants for the inverter 20.
  • the common controller 46 may be seen as an integrated control layer for both the rectifier 18 and the inverter 20. In the case of a current source inverter, the common controller 46 may determine the firing angles 48, 52 of both rectifier 18 and inverter 20. In the case of a voltage-source inverter, the common controller 46 may determine the modulation indices 48, 52 of both rectifier 18 and inverter 20. The control actions of rectifier 18 and inverter 20 are thus determined by the same common controller 46, allowing a systematic coordination of control actions.
  • the lower-level control and modulation controller 50, 54 may be separate, and need not be part of the common controller 46.
  • control method performed by the common controller 46 may be combined with several other converter/drive control concepts, for instance DTC (direct torque control), MPDTC (model predictive direct torque control), MP3C (model predicted pulse pattern control) and/or PWM (pulse width modulation).
  • DTC direct torque control
  • MPDTC model predictive direct torque control
  • MP3C model predicted pulse pattern control
  • PWM pulse width modulation
  • the control method of the common controller is based on model predicted control, which comprises a mathematical model 56 of the electrical system 10, which will be described in more detail with reference to Fig. 4.
  • the common controller 46 receives as input the measured or estimated latest states 38 of the electrical system 10 and the torque reference value 44 and uses the model 56 to compute the reference values 48, 52 such as firing angles, modulation indices, or current, voltage, power or flux references for the modulators 50, 54.
  • the following two figures show electrical systems 10 with other topologies that may be controlled with the control system 24. Due to the common, integrated controller 46 with a single model 56 of the electrical system 10 that also encodes the mutual influences of the different components, the control method may be beneficial for controlling converters 12 with a more complicated structure.
  • Fig. 2 shows a control system 10 with two subconverters 58 that are connected in parallel to the grid 14 and each supply an electrical machine 16.
  • Each subconverter 58 comprises a rectifier 18, DC link 22 and inverter 20 connected in series such as the converter 12 of Fig. 1.
  • the control system 24 generates the references values 32, 34 for both rectifiers 18 and both inverters 20.
  • Fig. 3 shows a control system 10 with a converter 12 that has one DC link 22, two parallel rectifiers 18 and two parallel inverters 20 connected to the common DC link 18. Also this topology may be modelled with a mathematical model 56 and used for applying model predictive control.
  • Fig. 4 shows a so called doubly fed machine comprising an rotating electrical machine 16 that is connected via a direct connection 60 and via a converter 12 with the grid 14. Also this topology may be modelled with a mathematical model 56 and used for applying model predictive control.
  • Fig. 5 shows a method for controlling the electrical system 10 with the control system 24.
  • the controller 46 may be based on a number of software routines or may be at least implemented in hardware.
  • the software routines and/or the hardware may be prepared by means of an initialization phase, afterwards the control method may be performed in an online phase.
  • Fig. 4 only shows the online phase. ln the initialization phase, the controller 46 and in particular the model 56 is prepared for application in a specific electrical system 10 (for example such as shown in Fig. 1 to 3). This preparation comprises two steps:
  • a dynamic model of the electric converter 12, the electrical machine 16 and/or the grid is provided.
  • the objectives in the form of a cost function
  • constraints of the electrical system 10 are provided.
  • the model 56 is typically stated in ordinary difference equations of the form
  • x(k+1) f(x(k),u(k)) (Eq. 1)
  • y(k) g(x(k),u(k)) (Eq. 2)
  • k represents discrete time instants
  • x(k) represents the state of the electric system 10 at time k
  • u(k) its inputs 48, 52, 34 at time k such as firing angles or modulation indices for the modulators 50, 54:
  • y(k) represents the measurable outputs 26, 28 of the electric system 10 at time instant k such as voltages, currents or the rotational speed of the electrical machine 16.
  • f is typically a nonlinear function describing the dynamic behaviour of the electrical system
  • g is typically a nonlinear function describing how the outputs depend on the states and inputs of the electrical system 10.
  • Fig. 4 shows a method that may be repeated every time instant k or in regular intervals every time instant nk, where n is a number.
  • software routines may be executed periodically online on a real-time computing platform, for example every few hundred microseconds.
  • the method may be performed by a collection of software routines on a real-time computing platform of the controller 24 or the common controller 46.
  • the collection of software routines may include a nonlinear or linear model of the electrical system 10 (for example modelling a transformer, the rectifiers 18, the DC link 22, the inverter 20, a load, the electrical machine 16, a filter, a cable, the grid 14, etc.).
  • the collection of software routines may further include a quadratic programming (QP) preparation algorithm and a quadratic programming (QP) solver, which may be used for solving the model such that a quadratic cost function is minimized as will be explained below.
  • QP quadratic programming
  • QP quadratic programming
  • the nonlinear optimization problem may be solved directly by means of an NLP solver, or the derivation of the QP from the NLP may be achieved by linearizing the nonlinear model 56 at each sampling instance.
  • NLP solver instead of solving the full NLP, one may alternatively also solve a simplified problem which still contains nonlinearities such as bilinear terms.
  • the estimator 36 receives the measurements 26, 28 and estimates the estimated state 38 of the electrical system 10. Furthermore, the speed controller 42 receives the speed reference 30 and determines the torque reference 44. In step 72, the common controller receives the torque reference 44 and the latest estimated state 38 and predicts a sequence of future states and future inputs over a horizon of at least two time instants with the model 56.
  • the estimated state 38 may comprise measurement values 26, 28 from the converter 12 and the electrical machine 16 and estimates ⁇ ( ⁇ ) of the states x(k) as provided by the estimator 36.
  • step 72 the constrained finite-time optimal control problem is linearized to arrive at a sub-problem in form of a quadratic problem (QP), which is easier to solve than the original problem.
  • the nonlinear model (Eq. 1 -2) is integrated along the prediction horizon p starting at the latest estimates 38 of the initial state ⁇ ( ⁇ . Also first- order derivatives of the predicted states with respect to the initial state ⁇ ( ⁇ as well as the control inputs u(k) (i.e. the reference values 48, 52) are determined.
  • the quadratic objective function of Eq. 5 penalizes the deviation of the predicted future states and inputs z from their references r, where Q is a positive semidefinite quadratic weight matrix used to tune the MPC controller.
  • the linearized model 56 equations (Eq. 3-4) are incorporated into the QP problem by means of the equality constraints of Eq. 7.
  • G eq and b eq represent the linearized state- update equations (Eq. 3) over the whole prediction horizon in a compact form.
  • Physical as well as desired limitations on the states and control inputs can be incorporated by means of the inequality constraints given by G, perpetu and £), ⁇ rad.
  • the cost function (Eq. 5) may comprise a linear objective term of the form g'(z-r), where the gradient g is an additional tuning parameter to provide more flexibility in penalizing set-point deviations.
  • the gradient g is an additional tuning parameter to provide more flexibility in penalizing set-point deviations.
  • the QP problem formulation (Eq. 5 to 7) exhibits a special sparsity structure as it comprises optimization variables for both states ⁇ ( and control inputs u(k).
  • One way to exploit this structure in terms of computational efficiency is to employ a sparse QP solver.
  • Another approach is to use the linearized state-update equations (Eq. 3) to remove all but the initial state ⁇ ( from the vector z and thus from the QP problem formulation. This approach referred to as “state-elimination” or “condensing” may lead to a smaller-scale, but dense QP problem. This second approach is typically more efficient on short prediction horizons.
  • the convex QP (Eq.
  • step 74 future rectifier reference values or control inputs 48 to be input to the at least one rectifier 18 and future inverter reference values or control inputs 52 to be input to the at least one inverter 20 are determined by selecting at least one future time instant from the optimized control inputs u opt (k). It is common practice in MPC to implement only this first piece u opt (k) of the optimized control input profile in a moving horizon fashion.
  • step 76 the optimized control inputs are distributed as reference values to the modulators 50 and 55.
  • the modulator 50 determines future switching states 32 for the at least one active rectifier 18 from the selected rectifier reference values 48 and the modulator 54 determines future switching states 34 for the at least one inverter 20 from the inverter reference values 52.
  • the whole procedure is repeated starting with step 70.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Inverter Devices (AREA)
  • Feedback Control In General (AREA)

Abstract

L'invention concerne un système électrique (10) qui comprend un convertisseur électrique (12) destiné à interconnecter une source électrique (par exemple, un réseau) (14) à au moins une charge électrique rotative (16), le convertisseur électrique (12) comprenant au moins un redresseur actif (18) et au moins un onduleur (20). Un procédé permettant de commander le système électrique (10) comprend les étapes consistant à : recevoir une référence de point de réglage (44) pour la ou les charges électriques (16) et un dernier état estimé (38) pour le convertisseur (12), l'état estimé (38) comprenant des valeurs estimées qui sont estimées à partir de valeurs de mesure (26, 28) du système électrique (10) ; prédire une séquence d'états futurs et d'entrées futures sur un horizon d'au moins deux instants temporels avec un modèle mathématique (56) du système électrique (10), les entrées futures comprenant des valeurs de référence de redresseur futures pour le ou les redresseurs (18) et des valeurs de référence d'onduleur futures pour le ou les onduleurs (20), et la séquence d'états futurs étant prédite avec le modèle mathématique (56) du système électrique (10) en réduisant à un minimum une fonction de coût, le modèle mathématique (56) étant basé sur un certain nombre d'équations se rapportant à l'état actuel du système électrique (10) avec un état futur du système électrique (10) ; et sélectionner des valeurs de référence de redresseur futures (48) pour les entrer dans le ou les redresseurs (18) et des valeurs de référence d'onduleur futures (52) pour les entrer dans le ou les onduleurs (20) par sélection d'au moins un instant temporel futur dans l'horizon.
PCT/EP2014/066420 2013-08-28 2014-07-30 Contrôle prédictif par modèles d'un système électrique WO2015028242A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP13182060 2013-08-28
EP13182060.7 2013-08-28

Publications (2)

Publication Number Publication Date
WO2015028242A2 true WO2015028242A2 (fr) 2015-03-05
WO2015028242A3 WO2015028242A3 (fr) 2015-10-08

Family

ID=49385096

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2014/066420 WO2015028242A2 (fr) 2013-08-28 2014-07-30 Contrôle prédictif par modèles d'un système électrique

Country Status (1)

Country Link
WO (1) WO2015028242A2 (fr)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016177535A1 (fr) 2015-05-05 2016-11-10 Abb Schweiz Ag Procédé de commande hybride pour convertisseur électrique
WO2016207165A1 (fr) 2015-06-23 2016-12-29 Abb Schweiz Ag Procédé de commande d'un système de compresseur pendant des baisses de tension
WO2018162334A1 (fr) * 2017-03-07 2018-09-13 Robert Bosch Gmbh Procédé et dispositif de réglage d'un entraînement électrique et entraînement électrique
CN108599547A (zh) * 2018-04-28 2018-09-28 西安理工大学 三相电压型功率因数校正变换器鲁棒模型预测控制方法
CN109038673A (zh) * 2018-08-27 2018-12-18 上海理工大学 光伏发电系统的模型预测优化控制方法
CN111948946A (zh) * 2020-08-24 2020-11-17 淮阴工学院 一种基于hji理论的鲁棒评价逆变控制系统及其设计方法
CN112803861A (zh) * 2021-03-19 2021-05-14 哈尔滨理工大学 一种永磁同步电机三矢量模型预测控制的无零矢量算法
CN110895721B (zh) * 2018-09-12 2021-11-16 珠海格力电器股份有限公司 电器功能的预测方法及装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6448735B1 (en) * 2001-04-26 2002-09-10 Abb Automation Inc. Controller for a wound rotor slip ring induction machine
IT1399117B1 (it) * 2010-04-01 2013-04-05 Nuovo Pignone Spa Sistema e metodo di smorzamento del modo torsionale basato su anello ad aggancio di fase
EP2621074A1 (fr) * 2012-01-24 2013-07-31 ABB Research Ltd. Contrôle prédictive d'un convertisseur électrique avec fonction de coût pour les processeur multicoeurs

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016177535A1 (fr) 2015-05-05 2016-11-10 Abb Schweiz Ag Procédé de commande hybride pour convertisseur électrique
US10020753B2 (en) 2015-05-05 2018-07-10 Abb Schweiz Ag Hybrid control method for an electrical converter
WO2016207165A1 (fr) 2015-06-23 2016-12-29 Abb Schweiz Ag Procédé de commande d'un système de compresseur pendant des baisses de tension
WO2018162334A1 (fr) * 2017-03-07 2018-09-13 Robert Bosch Gmbh Procédé et dispositif de réglage d'un entraînement électrique et entraînement électrique
CN108599547A (zh) * 2018-04-28 2018-09-28 西安理工大学 三相电压型功率因数校正变换器鲁棒模型预测控制方法
CN109038673A (zh) * 2018-08-27 2018-12-18 上海理工大学 光伏发电系统的模型预测优化控制方法
CN109038673B (zh) * 2018-08-27 2022-05-27 上海理工大学 光伏发电系统的模型预测优化控制方法
CN110895721B (zh) * 2018-09-12 2021-11-16 珠海格力电器股份有限公司 电器功能的预测方法及装置
CN111948946A (zh) * 2020-08-24 2020-11-17 淮阴工学院 一种基于hji理论的鲁棒评价逆变控制系统及其设计方法
CN111948946B (zh) * 2020-08-24 2022-05-17 淮阴工学院 一种基于hji理论的鲁棒评价逆变控制系统及其设计方法
CN112803861A (zh) * 2021-03-19 2021-05-14 哈尔滨理工大学 一种永磁同步电机三矢量模型预测控制的无零矢量算法

Also Published As

Publication number Publication date
WO2015028242A3 (fr) 2015-10-08

Similar Documents

Publication Publication Date Title
WO2015028242A2 (fr) Contrôle prédictif par modèles d'un système électrique
Vatani et al. Indirect finite control set model predictive control of modular multilevel converters
US9590530B2 (en) Controlling a modular converter in two stages
EP2963759B1 (fr) Dispositif convertisseur de courant pour connexion au réseau
WO2016012347A1 (fr) Commande prédictive de modèle d'un convertisseur modulaire à niveaux multiples
CN107408880B (zh) 电气转换器系统及其控制方法、计算机可读介质和控制器
Yan et al. Optimized and coordinated model predictive control scheme for DFIGs with DC-based converter system
Hassine et al. Predictive control strategies for wind turbine system based on permanent magnet synchronous generator
Besselmann et al. Model predictive control of load-commutated inverter-fed synchronous machines
EP2884653B1 (fr) Améliorations apportées ou se rapportant aux commandes de convertisseurs
Shadmand et al. Auto-tuning the cost function weight factors in a model predictive controller for a matrix converter VAR compensator
US11404968B2 (en) Current source converter with dynamical firing angle determination
US10020753B2 (en) Hybrid control method for an electrical converter
CN111656673A (zh) 针对两个或更多个控制目标，首先求解成本函数，然后求解第二成本函数的顺序预测控制方法
Vamvoudakis et al. Online optimal switching of single phase DC/AC inverters using partial information
Minchala-Avila et al. A model predictive control approach for integrating a master generation unit in a microgrid
Radmanesh et al. Linear Modelling of Six Pulse Rectifier and Designee of Model Predictive Controller with Stability Analysis
Almér et al. Nonlinear model predictive torque control of a load commutated inverter and synchronous machine
Chaabane et al. Indirect self tuning adaptive control of double stars induction machine by sliding mode
Vu et al. A generalized parameter tuning method of proportional-resonant controllers for dynamic voltage restorers
Gryning et al. Stability boundaries for offshore wind park distributed voltage control
Mossa Effective predictive flux control for a five phase induction motor drive with inverter output filter
EP3292623B1 (fr) Procédé de commande hybride pour un convertisseur électrique
Ryvkin et al. Sensorless oil drowned pump drive
Choi et al. A novel MPC-SVM strategy for direct torque flux control of an induction motor drive system using a matrix converter

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14750333

Country of ref document: EP

Kind code of ref document: A2