• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method for controlling a synchronous permanent magnet electric machine with salient poles for an electric or hybrid motor vehicle controlled by an electronic control unit capable of emitting current instructions, comprising steps in which: the currents at the level of the phases of the stator of the machine in a fixed three-phase reference connected to the stator, and voltage values applied in a rotary reference connected to the rotor are received, and the currents at the level of the stator phases are determined in a rotating two-phase reference linked to the rotor according to the current measurements in the three-phase reference and the rotor position. The method also comprises steps in which the magnetic flux and the rotor rotor speed are determined by an observer depending on the stator currents and the stator voltages expressed in the Park reference, and the observer is adjusted by a Kalman algorithm. in discrete extended version.
    公开号:FR3035755A1
    申请号:FR1553903
    申请日:2015-04-30
    公开日:2016-11-04
    发明作者:Mohamad Koteich；Abdelmalek Maloum
    申请人:Renault SAS；
    IPC主号:
    专利说明:

    [0001] A method for controlling a synchronous electric machine with permanent magnets for a motor vehicle.
    [0002] The technical field of the invention is the control of electrical machines and more particularly the control of synchronous electric machines with permanent magnets used for the traction of an electric or hybrid vehicle. An electric machine of the permanent magnet synchronous motor type comprises a rotor (rotating part) provided with permanent magnets and a stator (fixed part). The stator is able to generate a rotating magnetic field from a three-phase power supply (phases a, b and c). The permanent magnet rotor seeks to align with the rotating magnetic field produced by the stator. The rotor rotates at the same frequency of the stator currents, which gives the electric machine its qualifier of "synchronous". When such an electric traction machine is subjected to certain operating conditions, the permanent magnets of the rotor can be demagnetized. Among the operating conditions that can lead to demagnetization, temperature plays a vital role. For the safety of the vehicle, it is therefore important to detect the first signs that demagnetization will take place in order to prevent it. Furthermore, it is known that the sensors are influenced by their environment (noise, temperature, etc.) and may fail. Software sensors or estimators are more robust to the environment.
    [0003] From the state of the art, the following documents are known. Document US Pat. No. 8,569,986 discloses the detection of demagnetization without calculating the value of the rotor flux. This solution has a current detector, voltage detector of the 3035755 DC 2 source (direct current) which powers the motor (via power electronics). Thus, this solution has a memory and a data processing unit. WO 2004019269 discloses another solution based on the injection of high frequency currents and voltages. Degaussing is detected by examining the reaction of the salient pole machine. Obviously this solution does not apply to machines with smooth poles. The document US20110181217 discloses an algebraic relationship 10 for calculating the flux as a function of the estimated temperature using a Kalman filter. This method is complex and unreliable (estimation in open loop of the flow therefore more sensitive to noise, moreover does not work when the speed approaches zero). There is therefore a technical problem relating to the detection of the demagnetization of the permanent magnets of the rotor of an electric machine and to the control of an electric machine that is tolerant to such demagnetization. The subject of the invention is a method for controlling a synchronous electric machine with permanent magnets for an electric or hybrid automotive vehicle controlled by an electronic control unit capable of emitting current instructions, comprising steps during which the current at the stator phases of the machine in a fixed three-phase reference connected to the stator, and voltage values applied in a rotary reference connected to the rotor are received, and the currents at the stator phases are determined in a reference rotating two-phase rotor-related according to the current measurements in the three-phase reference and the rotor position. The method also includes steps in which the magnetic flux of the rotor is determined by an observer as a function of the stator currents and stator voltages expressed in the Park coordinate system, and the observer is adjusted by a discrete extended version Kalman algorithm. . The observer's setting by a discrete extended version Kalman algorithm can include the following steps: During a prediction phase, the state of the system and the covariance matrix of the system are determined. error associated with the next iteration estimated at the current iteration, as a function of the covariance matrix of the system uncertainty at the current iteration, of the covariance matrix of the error on the state to the the current iteration, from the estimated state to the current iteration and from the linearized system to. the current iteration, the gain of the observer at the current iteration is determined as a function of the covariance matrix of the error on the state at the next iteration estimated at the current iteration, of the noise covariance matrix of measurements at the current iteration and the linearized system at the current iteration, and update the state of the system at the next iteration according to the last determined measurements, of estimated magnitudes 15 corresponding, from the gain of the observer to the current iteration, and from the state to the next iteration estimated at the current iteration. The observer's dynamics can be increased by increasing the values of the uncertainty covariance matrix of the system.
    [0004] The accuracy of the observer can be increased despite the speed by increasing the values of the noise covariance matrix of measurements. The speed of the rotor relative to the stator can also be determined by the observer as a function of the stator currents and the stator voltages expressed in the Park reference. The control method has the advantage of being a passive solution, that is to say which does not require the injection of additional signals into the machine, and which has no additional detector or sensor with respect to sensors commonly used for the control of an electric machine. This method not only makes it possible to detect the magnetization, but also to have an estimate of the rotor flux, which makes it possible to act on the electrical machine before the demagnetization.
    [0005] This method also makes it possible to obtain an estimate of the speed of rotation of the rotor, which makes it possible to estimate the reliability of the corresponding speed sensor. The state vector of the observer may comprise the components of the stator currents in the Park coordinate system, the rotor speed and the rotor flux, the observer input vector may comprise the components of the stator voltages in the stator. Park marker, and the observer output vector can include the stator currents in the Park landmark.
    [0006] Other objects, features and advantages of the invention will appear on reading the following description, given solely by way of nonlimiting example and with reference to the appended drawings, in which: FIG. a synchronous electric machine with permanent magnet rotor in the Park mark, and - Figure 2 illustrates the main steps of the control method according to the invention. The control of the machine is in the Park mark 20 (d, q, 0), which is the transform of the fixed stator reference (a, b, c) by a rotation transformation, requiring the knowledge of the value of the Rotor angle 0. The Park transform matrix which transforms the three-phase quantities (a, b, c) into continuous quantities (d, q, 0) is the following: (t r r, 2r (27 cos0 cos cos 0 + x 3) (3) (27r (2r -sine -sin 0 - -sin 0 + - x 3) 3) Nk Vi N / 2 2 2 2 P (e) (Eq 1) The equivalent scheme of the electric machine in the Park marker is illustrated in Figure 1.
    [0007] Indeed, the own and mutual stator inductances depend on the position 0 of the rotor relative to the stator, also called rotor angle. Note that the rotor may have smooth poles (cylindrical) or protruding poles (non-cylindrical). The control of an electric machine provided with such a rotor is thus easier in a rotating marker such as the Park mark in which the inductances no longer depend on the position 0 of the rotor relative to the stator. Recall that in automatic and information theory, a state observer is an extension of a model represented as a state representation. When the state of a system is not measurable, we design an observer that allows to reconstruct the state from a model of the dynamic system and measurements of other quantities.
    [0008] Several state observers can be used for the control of mechanical motors without mechanical sensors, among which is the Kalman filter which is used in a wide range of technological fields. The model of the MSAP (acronym for "Permanent Magnet Synchronous Machine") in the Park marker (d, q) linked to the rotor is used for this reason. The angular position is measured by a position sensor (or can be estimated by an estimation technique). The angular position makes it possible to pass from the three-phase reference (a, b, c) to the Park mark (d, q) by applying the Park transformation (P (0)) and described above. The electric machine can be modeled in Park's two-phase reference (d, q, 0) by projecting the magnitudes of the three-phase reference (a, b, c) on a two-phase reference linked to the rotor. The transformation matrix corresponding to such a projection is P (0). It is noted that homopolar components are not taken into account. The electromagnetic equations of the system in this frame can be written in the following way: ## EQU1 ## where: vd = the voltage applied to the stator phase on the axis d 5 (corresponding to the voltage across a two-phase winding equivalent to the three-phase windings, on the axis d) vq: the voltage applied to the stator phase on the axis q id the current flowing in the stator phase d iq: the current flowing in the stator phase q 10 yd: the electromagnetic flux in the stator phase d yq: the electromagnetic flux in the stator phase q Rs: the resistance of a stator phase = p * S2 p The number of poles of the machine 15 S2: the rotational speed of the rotor The fluxes are determined by the following equations: d Ldid + 11 / f = Lq / q (Eq.3) With: L d and Lq: the inductances of phases d and q. f: the rotor flux.
    [0009] The mechanical equations of the machine are as follows: ## EQU1 ## With J: the inertia of the rotor with the load (Eq.4) ## EQU1 ## the number of pairs of rotor poles Cm and Cr: the motor and resistant torques: the viscous coefficient of friction 5 The motor torque is expressed by the following equation: 3 Cm = -2 P (Ilfdig -11 / q / d) The observer described below is based on the equations of the machine in the rotating marker of Park (d, q). It is known that in order to have measurements of currents in this reference frame, it is necessary to carry out a rotation transformation of the current measurements in the three-phase reference (a, b, c). The inputs of the system are the voltages in the Park reference (d, q), noted (vd and vq). These voltages are known since they are calculated at the control and sent to the motor. In motor modeling, the rotor flux is considered constant, since it varies very slowly in comparison with currents and velocity. This is mathematically translated as: f - 0 dt (Eq.6) The electric machine is modeled in the Park reference (d, q) in the following form: (Eq.5) - dx = dt d lq (1) f -1 (V d -Rsld qi q (9) - (Vq Rs iq - (Ldid + Vi f) (9) J (-f + 3p2 (LAid f) / qpC,) 0 (Eq.7) 25 with LA = Ld-Lq = 0 in the case of a machine with smooth poles 1d x = (1) Vif (Eq.9) 3035755 8 The measures id, iq are the currents measured then transformed in the stator mark of Park (d, q) The system of equations Eq. 7 modeling the electric machine can be reformulated in the general form of nonlinear systems: -dt dx f (x, u) y = h (x) Where x is the vector of state, u input vector (control) and y output vector (measure): (Eq. 8) It should be noted that this state vector could be rewritten without the rotation speed in the case where only the magnetic flux is determined 15 = Vd V (Eq.10) id y = (Eq.11) For the system modeled by the equations Eq.8 to Eq.11, we have 20 to formalize an observer by the following equation: x - f, u) + K (y - h (X)) dt (Eq. 12) With: the vector of the observed values corresponding to the vector x defined by the equation Eq.
    [0010] 9 25 K: Gain of the observer.
    [0011] 3035755 9 The gain K, which multiplies the error term, makes it possible to regulate the observer. This gain is calculated by the Kalman algorithm (discrete extended version).
    [0012] In order to allow the numerical resolution of the system, it is linearized in the following manner: af af Xk_1, Uk (Eq.13) H = 5 "k-1 (Eq.4) k The analytical form of the matrices Ak is thus calculated. and Hk which are, respectively, the Jacobians of the functions f and h of the equation Eq.8 with respect to the vector x These matrices are very complex so that they can not be written here They are determined by symbolic computation The process starts while the electric machine currents and voltages have been measured in the three-phase reference and converted to the Park mark by application of the matrix Eq. step 1 illustrated in FIG. 2, the values of the matrices Ak and Hk are determined as a function of the state values (currents, velocity, flux) of the previous iteration and of the measurements of the voltages by applying the equations Eq. Eq 14. At the first iteration the state values (current, velocity, flow) of the previous iteration are replaced by the corresponding initialization values. In a second step 2, a prediction phase is performed in which the state of the system is determined at the next iteration according to the data available at the current iteration. The following equations allow this prediction phase to be carried out: Xk + 1Ik Xk Tsf (xk, uk) Pk + 1Ik Pk + Ts (AkPk + PkAkT) + Qk (Eq.15) 30 With 3035755 10 TS: the sampling period k: the iteration number Pk: covariance matrix of the error on the state at the iteration k estimated at the iteration k 5: covariance matrix of the error on the state at the iteration k + 1 estimated at the iteration k Qk: matrix of covariance of uncertainties of the system the iteration k. The covariance matrix of the uncertainties of the Qk system makes account of uncertainties in the system definition, due for example to the ignorance of the system, to the approximation of the system modeling, or to the uncertainty about the values used in the system. modelization. In other words, during this step, the state 15 x of the system is determined at the iteration k + 1 estimated at the iteration k as a function, in particular, of the state of the system at the iteration k estimated at the iteration k. During a third step 3, the gain of the observer is calculated: KkPk + 11kHkT (11kPk + 11kHkT Rk) -1 (Eq.16) With Rk: noise covariance matrix of measurements at 1 iteration k. Finally, during a fourth step 4, a post update phase is performed during which the state of the system is updated at the iteration k + 1 estimated at the iteration k + 1. thanks to the information of the last measurements y and the corresponding estimated quantities h (x) as a function of the state of the system at the iteration k + 1 determined at the iteration k. The function h (x) depends directly on the modeling of the system (cf Eq.7). The following system of equations accounts for this update phase. The process steps described above are repeated so that the above-described process steps are repeated so as to enable the process steps described above to be carried out in order to reduce the temperature of the process. to have rotor flux values and speed regularly updated. The filter is adjusted by choosing the matrices Qk and Rk 5 which are often taken constant. This choice depends on the system to be observed, the parameters of the electrical machine and the environment in which the electrical machine operates (measuring noises). There is no systematic method, but the general rules are as follows.
    [0013] If one increases the values of the matrix Qk, one gives less confidence to the measurements, and the dynamics of the observer becomes faster. Increasing the values of the matrix Rk gives more confidence to the measurements, which increases the accuracy despite the rapidity. As a general rule, the matrices Qk and Rk are likely to have their values modified from one iteration k to the next. However, the present application does not require such modification. They have the form: ## EQU1 ## In general, these matrices are to be adjusted for each electric machine.
    [0014] This control method makes it possible to estimate the rotor flux and the rotor speed, and to detect the demagnetization of the rotor magnets and thus an abnormal operation (or failure) of the engine rotational speed sensor. The method and control system do not require any additional equipment over those commonly installed in a motor vehicle.
    [0015] 3035755 12 Vif 0 id. 10 x = -dt iq 1 (1) q - Rsig - (I, di d + vf j) (Eq 21) The demagnetization is detected as soon as the rotor flux Wf is below a threshold, for example 0.01 Wb . A failure of the speed sensor is detected as soon as the difference between the estimated rotational speed (θt) and the measured rotational speed horns is greater, in absolute value, than a threshold, determined according to the accuracy of the sensors and sensors. In a variant, a simplified model can be considered in which the rotational speed w is known Such a model can be formalized by the following system of equations resulting from the equation Eq. - Rsid Lq1q (9) Ld q We can eliminate the third line of the model (Eq.7), corresponding to the dynamics of w, and We consider that w is a known parameter in the dynamics of currents. same Kalman observer theory described above.
    权利要求:
    Claims (6)
    [0001]
    REVENDICATIONS1. A method for controlling a permanent magnet synchronous electric machine for an electric or hybrid motor vehicle controlled by an electronic control unit capable of emitting current setpoints, comprising steps in which: the currents at the stator phases are measured of the machine in a fixed three-phase reference connected to the stator, and one receives voltage values applied in a rotating reference linked to the rotor, one determines the currents at the level of the stator phases in a rotating two-phase reference linked to the rotor according to the measurements current in the three-phase reference and the rotor position, characterized in that it comprises steps during which the magnetic flux of the rotor is determined by an observer which is a function of the stator currents and the stator voltages expressed in the Park reference, and the observer is adjusted by a discrete extended version Kalman algorithm.
    [0002]
    The method of claim 1, wherein setting the observer by a discrete extended version Kalman algorithm includes the following steps during a prediction phase, determining the state of the system and the covariance matrix. of the error associated with the next iteration estimated at the current iteration, based on the covariance matrix of the uncertainty of the system at the current iteration, of the covariance matrix of the error on the state at the current iteration, of the state estimated at the current iteration and the linearized system at the current iteration, the gain of the observer at the current iteration is determined according to the matrix of covariance of the error on the state at the next iteration estimated at the current iteration, from the noise covariance matrix of measurements to the current iteration and from the linearized system to the current iteration, and 3035755 14 we update the state of the system at iteration s depending on the last determined measurements, the corresponding estimated magnitudes, the observer's gain at the current iteration, and the state at the next iteration estimated at the current iteration. 5
    [0003]
    The method of claim 2, wherein increasing the observer's dynamic by increasing the values of the uncertainty covariance matrix of the system.
    [0004]
    4. The method of claim 2, wherein increasing the accuracy of the observer despite the speed by increasing the values of the noise covariance matrix of measurements.
    [0005]
    5. Method according to any one of the preceding claims, wherein the speed of the rotor relative to the stator is also determined by the observer as a function of the stator currents and the stator voltages expressed in the Park reference. 15
    [0006]
    The method according to any one of the preceding claims, wherein the state vector of the observer comprises the components of the stator currents in the Park coordinate system, the rotor speed and the rotor flux, the input vector of the observer comprises the components of the stator voltages in the Park reference, and the output vector of the observer comprises the stator currents in the Park coordinate system.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP2246973B1|2018-05-30|Method for determining the position of the flux vector of a motor
    EP2790957B1|2016-03-09|Method for the contactless charging of the battery of an electric automobile
    EP1972051B1|2017-10-18|Method for determining the position of a rotor of a synchronous machine having at least one excitation winding
    FR3035755A1|2016-11-04|METHOD FOR CONTROLLING A PERMANENT MAGNET SYNCHRONOUS ELECTRIC MACHINE FOR A MOTOR VEHICLE.
    EP2414781B1|2014-12-17|Method and device for diagnosing malfunctions of sensors for determining the angular position of a rotor of a polyphase rotating electric machine
    FR3062762A1|2018-08-10|METHOD OF ESTIMATING THE ANGULAR POSITION OF A ROTOR OF AN ELECTRIC DRIVE SYSTEM
    EP2870018B1|2018-10-03|Method for controlling a power train and corresponding system
    EP3221958B1|2020-09-16|Method for controlling a synchronous electric machine with a wound rotor
    EP3391161B1|2019-10-09|Method for monitoring an electromechanical actuator system
    EP3012962A1|2016-04-27|Method for controlling a three-phase synchronous electric machine with a wound rotor
    FR3020730A1|2015-11-06|METHOD FOR ESTIMATING THE ELECTRICAL ANGLE OF AN ASYNCHRONOUS ELECTRIC MACHINE FOR A MOTOR VEHICLE
    FR3028362A1|2016-05-13|METHOD AND SYSTEM FOR CONTROLLING A SYNCHRONOUS ELECTRIC MACHINE WITH PERMANENT MAGNETS.
    EP3072019B1|2021-07-28|Critical system and associated monitoring method
    EP3314742B1|2020-08-05|Control of a synchronous electric machine with a wound rotor
    FR2986389A1|2013-08-02|CONTROL OF AN ELECTRIC MACHINE WITH PERMANENT MAGNETS
    EP3529889B1|2021-06-23|Methods and devices for estimating an angular position of a rotor
    EP2992602B1|2018-09-19|Method of verifying the operation of a motor propulsion plant fitted to an automotive vehicle and corresponding system
    FR3086473A1|2020-03-27|METHOD FOR DETERMINING THE MAGNETIC FLOW OF AN ELECTRIC MACHINE
    WO2020016001A1|2020-01-23|Method for determining the position and speed of the rotor of a wound-rotor synchronous electric machine
    WO2019008285A1|2019-01-10|Current estimation
    FR3106894A1|2021-08-06|Method for estimating the electromagnetic torque of a synchronous electric machine
    WO2016098032A1|2016-06-23|Method and device for evaluating operating parameters as well as process and system for controlling a synchronous machine
    FR2986289A1|2013-08-02|Active magnetic bearing e.g. radial-type active magnetic bearing, with self-position detection for electrical motor, has control device including position estimating unit to estimate position of ferromagnetic body forming rotor
    同族专利:
    公开号 | 公开日
    FR3035755B1|2018-04-27|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    FR2778800A1|1998-05-13|1999-11-19|Daimler Chrysler Ag|METHOD FOR ESTIMATING THE CONDITION OF ASYNCHRONOUS MOTORS WITHOUT ROTATION SPEED TRANSMITTER|WO2020058131A1|2018-09-20|2020-03-26|IFP Energies Nouvelles|Method for determining the magnetic flux of an electrical machine|
    FR3092214A1|2019-01-30|2020-07-31|Renault S.A.S|Control system of a synchronous motor with permanent magnets with determination of demagnetization|
    WO2020244954A1|2019-06-04|2020-12-10|Renault S.A.S|Method for estimating the electomagnetic torque of a synchronous electric machine|
    法律状态:
    2016-04-21| PLFP| Fee payment|Year of fee payment: 2 |
    2016-11-04| PLSC| Search report ready|Effective date: 20161104 |
    2017-04-19| PLFP| Fee payment|Year of fee payment: 3 |
    2018-04-20| PLFP| Fee payment|Year of fee payment: 4 |
    2019-04-18| PLFP| Fee payment|Year of fee payment: 5 |
    2020-04-20| PLFP| Fee payment|Year of fee payment: 6 |
    2021-04-23| PLFP| Fee payment|Year of fee payment: 7 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1553903A|FR3035755B1|2015-04-30|2015-04-30|METHOD FOR CONTROLLING A PERMANENT MAGNET SYNCHRONOUS ELECTRIC MACHINE FOR A MOTOR VEHICLE.|
    FR1553903|2015-04-30|FR1553903A| FR3035755B1|2015-04-30|2015-04-30|METHOD FOR CONTROLLING A PERMANENT MAGNET SYNCHRONOUS ELECTRIC MACHINE FOR A MOTOR VEHICLE.|
    [返回顶部]