Method for estimating the speed and position of a rotor of a wound rotor synchronous machine

Abstract

The invention relates to a method for estimating the speed and position of a rotor (50) of a wound rotor synchronous machine (50) supplied by a three-phase power network, the method comprising: a step of injecting a high-frequency voltage signal into the three-phase power network; a step of demodulating (101) the current transformed by the second transformation step (101), including high-pass filtering or band-pass filtering and used to determine an estimated error signal (ε); estimating (102) a phase shift (φ) resulting from the rotor acceleration and the high-pass filtering or band-pass filtering of the demodulation step (101). comp ) to refine the estimated error signal (∈) determined during the demodulation step (101); a step of separating the high frequency component of the measured current from the low frequency (103); the method further comprises a second part (12) for gradually estimating the position, speed and acceleration of the rotor using uncorrelated gain parameters based on the sign of the estimated error obtained.

Description

Technical Field

[0001] The invention relates to the field of wound rotor synchronous motors.

[0002] More particularly, the present invention relates to a method for determining the position and speed of a rotor of a wound rotor synchronous electric machine. Background Art

[0003] In order to control a wound rotor synchronous machine (abbreviated as WRSM), it is usually necessary to know the position and speed of the rotor.

[0004] One solution known from the prior art consists in mounting one or more mechanical position and speed sensors on the mechanical shafts of the machine.

[0005] However, these mechanical sensors are expensive, bulky, sensitive to the environment (temperature, noise, mechanical vibration, electromagnetic compatibility, etc.), and can reduce system reliability.

[0006] Therefore, in order to avoid the use of mechanical sensors, control methods without mechanical sensors have been developed to ensure the same or even better control quality than methods using mechanical sensors for control.

[0007] Typically, these sensorless control methods use mechanical position / speed estimation methods based solely on the measurement of current in closed-loop mode, also known as software sensors.

[0008] Also known are methods for estimating the position / speed of a rotor by injecting a high-frequency signal, as described in document US2004070360 A1, which method has the effect of allowing detection to be less dependent on machine parameters.

[0009] However, these methods still rely on the parameters of the motor, and more specifically for WRSM, on the stator inductance seen by the rotor. In addition, these techniques rely on knowing the characteristics of the injected signal, such as amplitude and frequency.

[0010] Therefore, there is a need for a position / speed estimation method that is more reliable and less dependent on the parameters of a wound-rotor synchronous machine. Summary of the invention

[0011] To this end, a method for estimating the speed and position of a rotor of a wound rotor synchronous motor powered by a three-phase inverter is proposed, the method comprising:

[0012] - a step of measuring the three-phase current at the input terminals of the wound-rotor synchronous machine;

[0013] - A step of transforming the measured three-phase current into a two-phase reference system;

[0014] - The first part includes:

[0015] - a step of injecting a high frequency voltage signal at the input of the machine;

[0016] Characterized in that the first part further includes determining a rotor position error value, including:

[0017] - a second step of transforming the measured converted current into a two-phase reference frame by rotating it by π / 4 radians;

[0018] - a step of demodulating the current transformed by this second transformation step, comprising high-pass filtering or band-pass filtering and allowing determination of an estimated error signal;

[0019] - a step of estimating the phase shift resulting from the rotor acceleration and from the high-pass filtering or band-pass filtering of the demodulation step, in order to refine the estimated error signal determined in the demodulation step;

[0020] - a step of separating the high-frequency components from the low-frequency components of the measured current; said separation step is independent of the low-pass filtering and allows determining the sign of the rotor position estimation error;

[0021] The method further comprises estimating the position, the velocity and a second part of the rotor acceleration step by step using mutually unrelated gain parameters according to the signs of the obtained estimation errors.

[0022] Thus, a relatively simple and robust estimate of the position, speed and acceleration of the wound rotor can be obtained solely from the sign of the estimated error obtained (the sign being defined from the error signal calculated by injecting a high frequency voltage). This makes it possible in particular to obtain rotor position estimates, speed estimates and acceleration estimates which are independent of each other and more particularly calibrated by independent gains from each other.

[0023] Advantageously and in a non-limiting manner, the demodulation step comprises a high-pass filtering of said current. The demodulation is therefore relatively simple and robust and does not generate delays with respect to the estimate obtained of the rotor position.

[0024] Advantageously and in a non-limiting manner, the phase shift estimation step comprises a low frequency filtering. Thus, the estimation of the phase shift is relatively simple and efficient.

[0025] Advantageously and in a non-limiting manner, the phase shift estimation step comprises a phase locked loop. Thus, the estimation of the phase shift is controlled relatively robustly.

[0026] Advantageously and in a non-limiting manner, the step of separating the high frequency components from the low frequency components of the measured current comprises calculating a rotor position estimation error signal defined by the following equation:

[0027]

[0028] Among them, I cn is the magnitude of the negative component of the stator current, ω c is the angular frequency of the injected high-frequency signal, φ comp is the estimated phase shift, and is the rotor position error.

[0029] The sign of the rotor position error can therefore be simply determined from the estimated error signal, thereby then enabling the second part of the method to be implemented to obtain a simple and robust estimate of the speed, position and acceleration of the rotor.

[0030] Advantageously and in a non-limiting manner, the second part comprises the implementation of at least one low-pass filter. This low-pass filter makes it possible to limit the jitter phenomena of the rotor position error sign function.

[0031] In particular, the low pass filter is a 4th order filter. Therefore, such a filter does not produce undesirable effects (such as phase shifts) on the estimates of the speed, position and acceleration of the rotor.

[0032] The invention also relates to a device for estimating the speed and position of a rotor, comprising means for implementing the method as described above.

[0033] The invention also relates to an electrical assembly comprising a wound rotor synchronous machine and an estimation device as described above.

[0034] The invention also relates to a motor vehicle comprising an electrical assembly as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Other specific features and advantages of the invention will become apparent from reading the following description of a specific embodiment of the invention given in an indicative and non-limiting manner, with reference to the accompanying drawings, in which:

[0036] - Figure 1 is a schematic diagram of a control component of a motor according to an embodiment of the present invention;

[0037] - Figure 2 is a schematic diagram of an estimation method according to an embodiment of the present invention;

[0038] - Figure 3 is based on Figure 2 A representation of the step of estimating the phase shift of the stator current of the method of an embodiment;

[0039] - Figure 4 is based on Figure 2A representation of a high frequency / low frequency separation step independent of a low pass filter of a method of an embodiment;

[0040] - Figure 5 is based on Figure 2 A view of a second estimation portion of a method of an embodiment of the present invention; and

[0041] - Figure 6 is a representation of the geometric transformation of the current relative to the rotor reference frame,

[0042] -as well as Figure 7 is a representation of the continuous algorithm according to equation (12). DETAILED DESCRIPTION

[0043] refer to Figure 1 , a control assembly 1 of an electric machine, here for example an electric vehicle 1 , comprises a torque setpoint device 2 (for example an accelerator pedal 2 ) for requesting torque from the electric machine.

[0044] The torque setpoint generated by the torque setpoint device 2 is then processed by a current regulator 3 and then by an inverter 4 in order to supply a suitable control current to the electric machine 5 , here a wound rotor synchronous machine 5 .

[0045] In order to allow efficient control of the electric machine, the position of the rotor of the electric machine (in other words the angular position of the rotor relative to the stator), its speed and advantageously its acceleration must be known. To this end, an estimation method 6 is implemented.

[0046] because Figures 2 to 6 These figures relate to the same embodiment of the estimation method according to the invention and will therefore be discussed simultaneously.

[0047] A method for estimating the speed and position of a rotor of a six-wound rotor synchronous machine comprises the steps of measuring 10 three-phase currents and two method parts: a first part 100 comprising signal processing and demodulation and a second part 200 comprising estimating position and speed from the results of the first part.

[0048] First, the method implements the measurement of the three-phase current i at the input of the 10-winding rotor synchronous motor. a 、i b 、i c However, this step does not have to be performed before the first part 100 of the method, and this step can also be performed during the first part 100 of the method, for example, when the measured three-phase current value i is required. a 、i b 、i c Executed before.

[0049] Then, the measured three-phase current i a 、ib 、i c Transform into the two-phase reference system αβ.

[0050] To this end, a transformation is applied in the rotor reference frame 50, such as Figure 6 Therefore, according to the measured three-phase current i a 、i b 、i c , the system of measured two-phase currents is derived by applying the following equations

[0051]

[0052] This equation (1) describes the three-phase current i according to a static three-phase to two-phase transformation 13 (here the Concordia transformation) to the reference frame αβ a 、i b 、i c The measurement results.

[0053] In order to model the high-frequency behavior of the synchronous machine, a model based on the following two equations is then applied:

[0054] - Voltage-Flux Model:

[0055] -Current-Flux Model:

[0056] in, and are the average inductance and differential inductance of the machine, L d and L q is the inductance on the axes d and q of the rotating two-phase reference frame dq (which is the Park reference frame), and denote the three-phase voltage and current of the machine seen at the stator, respectively, and is the stator flux of the machine),

[0057] In order to estimate the position, speed and acceleration of the AC motor, the so-called pulsation technique is implemented, in which the estimated two-phase reference system The offset of the current measurement result in the estimated reference frame represents the estimated Parker reference frame is Therefore, the reference system of the axes dm and qm is the same as that of the injection axis and The reference frame is shifted

[0058] On the axis A high frequency voltage is injected and the current is measured on the axis dm and the axis qm.

[0059] refer to Figure 6 , in particular the angular phase shift is represented in the rotor reference frame 50 .

[0060] The pulsation technique allows the injection of a high frequency (HF) voltage into the estimated two-phase reference system. middle:

[0061]

[0062] in:

[0063] V c is the amplitude of the injected HF voltage; and

[0064] ω c is the angular frequency of the injected HF voltage.

[0065] refer to Figure 6 , the offset relative to the injection reference frame is obtained as follows The current in the reference frame

[0066]

[0067] in, as well as are the amplitudes of the positive components I cp , the amplitude of the negative component I cn , and the fundamental component of the stator current θ is the position of the rotor, and is the estimated position of the rotor.

[0068] Then, a step 101 of demodulating the resulting signal after injecting the high frequency voltage is implemented.

[0069] For this purpose, a high-pass filter (abbreviated to HPF) or, alternatively, a band-pass filter (single frequency filtering, abbreviated to SFF) is used to phase-shift the The current in the reference frame Filter to eliminate its fundamental component.

[0070] The resulting high frequency current is obtained according to the following equation

[0071]

[0072] It is expanded by trigonometric functions:

[0073]

[0074] In this development, the difference Used to extract the position estimation error signal Position estimation error signal The extraction 101 corresponds to the demodulation 101 of the signal.

[0075]

[0076] The estimated error signal ∈ is formulated according to equation (7), and the angular error between the position of the rotor and the estimated position of the rotor is is a function of the estimated error signal ∈.

[0077] Therefore, by analyzing the estimated error signal ∈, as described below, the position error can be derived from the sign of the estimated error signal ∈ The sign of the position error The sign of makes it possible to determine estimates of the position, velocity and acceleration of the rotor in the second part of the method.

[0078] Once the estimated error signal ε has been obtained according to equation (7), in other words, once demodulation has been performed, a phase shift estimation step 102 is implemented, such as Figure 3 shown.

[0079] The speed change during the acceleration phase of the machine will be at the signal carrier level (cos(ω c t+φ acc )) produces a phase shift φ acc .

[0080] The use of a high-pass filter HPF or, alternatively, a band-pass filter (single frequency filtering, abbreviated SFF) in the demodulation step 101 also generates a signal at the carrier level (cos(ω c t+φ HBF )) produces a phase shift φ HBF .

[0081] The signal of the carrier thus experiences these delays, and its expression (as previously formulated in equation (7)) becomes:

[0082] ∈=A cos(ω c t+φ comp ) (8)

[0083] in, And φ comp =φ HBF +φ acc .

[0084] To extract the position estimate error in term A, we multiply ∈ by the term Therefore, it is necessary to estimate the phase shift φ comp .

[0085] By multiplying the estimated error in equation (8) by the term Get the following formula:

[0086]

[0087] as well as

[0088]

[0089] By applying a low pass filter (abbreviated as LPF), the following equation is obtained:

[0090]

[0091] And, by applying a continuous algorithm (here a phase-locked loop, abbreviated as PLL) to (11), the phase shift φ can be calculated comp Estimates:

[0092]

[0093] The phase shift φ comp The estimate of then makes it possible to obtain:

[0094]

[0095] And therefore

[0096]

[0097] Specifically, the purpose of estimating the phase shift is to reconstruct the high-frequency carrier signal To obtain the square of this component, the square of the carrier (high frequency) [cos(ω c t+ω comp )] 2 or φ comp It is an unknown quantity.

[0098] The above calculations allow to estimate the phase shift φ comp , which is equal to

[0099] Therefore, the phase shift estimation error is sent to the continuous sum error optimization (PLL) step (ref. Figure 7 ) Converges to φ comp .

[0100] Then, a step 103 of separating high frequency components from low frequencies is performed so that the use of a low pass filter (LPF) can be avoided. In the following description, this is referred to as the LPF estimation step 103.

[0101] The estimated error ∈ of the position of the included machine has been determined previously, so that the estimated error can be expressed according to the following equation:

[0102]

[0103] By making this estimated error equal to the phase shift Multiply the terms of to obtain the following formula:

[0104]

[0105] In the context of salient-pole wound-rotor machines (referred to as salient-pole rotor machines), since L q >L d ,so And-I cn >0.

[0106] thus:

[0107]

[0108] in:

[0109]

[0110]

[0111] Here, the term symbol means "the symbol of the contained expression".

[0112] The expression according to equation (15) describes the sign of the position estimation error without using the conventional technique based on a low-pass filter (LPF).

[0113] This estimation error according to equation (15) is then injected as information step by step and convergently in a finite time into a set of consecutive steps 200 according to the invention.

[0114] This set of steps 200 corresponds to a second part 200 of the method according to the invention, the purpose of which is to estimate the position, speed and acceleration of the alternating current motor.

[0115] In the second part 200, in order to make the process of setting up the technique for estimating position, velocity and acceleration straightforward, an estimator (also called an observation) is implemented (the estimator is robust and works step by step (e.g. Figure 4 As shown)), the position state, velocity state and acceleration state are converged one by one independently of each other. This makes it possible to set these states to converge in a finite time, with each state considered separately.

[0116] When the estimated position is is considered equal to the actual position θ, in other words, when the position error Approximately regarded as zero time:

[0117] if

[0118] but

[0119]

[0120] Now, the only measurements of the estimator are:

[0121] From the first part 100 of the method

[0122] The proposed robust step-by-step observations for estimating position, velocity, and acceleration are defined by the following equations:

[0123]

[0124] in:

[0125]

[0126] in,

[0127]

[0128] as well as

[0129]

[0130] as well as

[0131]

[0132] in,

[0133]

[0134] Therein, TZ denotes a Z transform which makes it possible to transform a time function σ(t) into a discrete function σ(z).

[0135] The function f(z) is introduced to detect the jitter phenomenon, because only the sign of the estimated error is available as information of the observed quantity; the position of the rotor is not available for measurement.

[0136] To obtain the filtered velocity and acceleration The value of , implements the 4th order low pass filter (LPF) used in the second part 200 of the method, reference Figure 5 These low-pass filters are introduced to reduce jitter in the sign function and do not affect the position, velocity and acceleration estimates, since these estimates are advantageously uncorrelated.

[0137] The virtual mechanical system for the designed position, velocity and acceleration of the observed quantities (21), (22) and (23) is as follows:

[0138]

[0139] Equations (33), (34) and (35) define the estimated errors of position, velocity and acceleration between equations (30)-(31)-(32) and the observations (21)-(22)-(23):

[0140]

[0141] The estimated error dynamic range is derived from the following equation:

[0142]

[0143] Among them, K θ >Max(|e ω |), K w >Max(|e α |) and K α >0 gain defines positive values ​​to limit noise.

[0144] Therefore, it is clear that the method according to the present invention ensures that the dynamic range of the estimation errors of position, velocity and acceleration (36)-(37)-(38) converges to zero within a finite time.

Claims

1. A method for estimating the speed and position of a rotor of a wound rotor synchronous machine (5) powered by a three-phase inverter, the method comprising: - measuring the three-phase current (i) at the input end of the wound rotor synchronous motor (5) a 、i b 、i c ) steps; - The measured three-phase current (i a 、i b 、i c ) is converted into the first conversion step of the current in the two-phase reference system αβ; - The first part (100) comprises: - a step of injecting a high frequency voltage signal at the input terminal of the wound rotor synchronous machine; Characterized in that the first part (100) further comprises determining a rotor position error value, which comprises: - The injection reference frame is rotated by π / 4 radians The second transformation step of transforming the current under the two-phase reference system dmqm into the current under the two-phase reference system dmqm; - a demodulation step of demodulating the current transformed by this second transformation step, comprising a high-pass filtering or a band-pass filtering and allowing the determination of an estimated error signal (ε); - Estimate the phase shift (φ) resulting from the rotor acceleration and the high-pass or band-pass filtering of this demodulation step comp ), to refine the phase shift estimation step of the estimated error signal (∈) determined in the demodulation step; - a separation step of separating the high-frequency components from the low-frequency components of the measured current; said separation step is independent of the low-pass filtering and allows the sign of the error in the estimation of the rotor position to be determined; The method further comprises estimating the position, the speed and a second part of the rotor acceleration step by step using mutually unrelated gain parameters according to the signs of the obtained estimation errors (200).

2. The method according to claim 1, characterized in that The phase shift estimation step includes low frequency filtering.

3. The method according to claim 2, characterized in that The phase shift estimation step includes a phase locked loop.

4. The method according to claim 1 or 2, characterized in that: The separation step includes calculating the rotor position estimation error signal defined by the following equation: Among them, I cn is the magnitude of the negative component of the stator current, ω c is the angular frequency of the injected high-frequency signal, φ comp is the estimated phase shift, θ is the rotor position, is the estimated position of the rotor, and is the rotor position error.

5. The method according to claim 1 or 2, characterized in that: The second part (200) includes implementing at least one low pass filter.

6. The method according to claim 5, characterized in that The low pass filter is a 4th order filter.

7. A device for estimating the speed and position of a rotor, the device comprising means for implementing the method as claimed in any one of claims 1 to 6.

8. An electrical assembly comprising a wound rotor synchronous machine and the apparatus of claim 7.

9. A motor vehicle comprising an electrical assembly as claimed in claim 8.

Images