Method and apparatus for controlling a three-phase electric machine

Abstract

The invention relates to a method and a device for tracking a minimum torque per ampere condition for controlling an electrical machine using direct flux vector control. The invention: - determines a torque reference; - determines a provisional flux target from the torque reference using a torque-flux dependency rule; - determines a flux target by adding a low frequency flux signal to the provisional flux target; - controls the electrical machine using direct flux vector control according to the flux target; - measures the current provided to the electrical machine at at least three time events, which are times at which the low frequency flux signal has different values; - estimates a maximum torque per ampere flux from the measured currents; - updates the dependency rule according to the estimated maximum torque per ampere flux.

Description

Technical Field

[0001] This invention generally relates to a method and apparatus for controlling a three-phase motor. This application claims priority to European Patent Application No. EP23306932.7, filed November 9, 2023, the contents of which are incorporated herein by reference. Background Technology

[0002] Due to the saturation of the magnetic components in a motor, the motor exhibits inductance that varies with current conditions. The torque produced by the motor depends largely on the value of the inductance under the current conditions considered. Therefore, several combinations of motor currents can be used to produce a given torque. Along a constant torque trajectory, a specific combination of motor currents minimizes the total root mean square (RMS) current. This ideal combination of maximum torque per ampere (MTPA) typically reduces the energy consumption of the motor driver.

[0003] For motors controlled by direct flux vector control (DFVC), the MTPA condition is determined by the optimal parameter pair that produces the expected torque level, the flux norm, and the current component perpendicular to the flux.

[0004] The MTPA condition can be determined using comprehensive knowledge of the motor inductance characteristics as a function of current.

[0005] Gaining comprehensive knowledge of machine inductance requires specific testing procedures with limited accuracy and the storage of large lookup tables (LUTs).

[0006] These lookup tables (LUTs) include inductance values ​​for current values. The drawbacks of this method relate to the lookup time before first use on the driven machine, the required memory, and accuracy, as it is difficult to obtain an effective inductance table for full load conditions without a pre-configured test load.

[0007] In addition, the inductance level may be affected by the specific temperature distribution of the motor, which varies over time based on the motor's past operating conditions. Summary of the Invention

[0008] The object of this invention is to determine the MTPA conditions without any knowledge of the machine sensing conditions.

[0009] Therefore, the present invention relates to a method for tracking the minimum torque-per-ampere condition of a motor controlled using direct flux vector control, characterized in that the method comprises the following steps: - Determine the torque reference. - Use torque-flux dependency rules to determine temporary flux targets from a torque reference. - The flux linkage target is determined by adding a low-frequency flux linkage signal to the temporary flux linkage target. - The motor is controlled using direct flux vector control based on the flux target. - Measure the current supplied to the motor at at least three time events, where each time event is a period when the low-frequency flux linkage signal has different values. - Estimate the flux linkage per ampere for maximum torque from the measured current. - Update the dependency rules based on the flux linkage per ampere of the estimated maximum torque.

[0010] The present invention also relates to an apparatus for tracking the minimum torque per ampere condition of a motor controlled using direct flux vector control, wherein the apparatus comprises: - A device for determining a torque reference. - A device for determining a temporary flux target from a torque reference using torque-flux dependence rules. - A device for determining a flux linkage target by adding a low-frequency flux linkage signal to a temporary flux linkage target. - A device for controlling a motor using direct flux vector control based on a flux target. - A device for measuring the current supplied to a motor at at least three time events, where each time event is a period when the low-frequency flux linkage signal has different values. - A device for determining the flux linkage with maximum torque per ampere from a measured current. - A device for updating the dependency rule based on the flux linkage per ampere of the estimated maximum torque.

[0011] Therefore, a device for tracking the maximum torque per ampere (MTPA) condition of a motor can identify the MTPA condition without specific knowledge of the electromagnetic characteristics of the motor to be operated. This identification can be achieved without high-frequency injection, and the low-frequency injection can be set to a frequency below the level of human hearing. Therefore, this method does not produce audible noise.

[0012] Based on specific characteristics, the torque-flux dependence rule is a linear function of the absolute value of the torque.

[0013] Therefore, this invention can adjust the flux linkage level according to varying torque conditions. Since the torque-flux linkage dependency intersects with the MTPA condition around the corresponding operating point, the motor operates close to the actual MTPA condition. Because the dependency rule is linear, the parameters involved in this invention are minimal, reducing memory consumption.

[0014] Based on specific characteristics, the measurement of the current supplied to the motor is performed at three time events under the same torque conditions: the first time event is when the low-frequency flux linkage signal is at its maximum, the second time event is when the low-frequency flux linkage signal is at its zero, and the third time event is when the low-frequency flux linkage signal is at its minimum.

[0015] Therefore, the true MTPA condition (i.e., the ideal flux linkage level that minimizes copper losses in the motor) can be determined using an interpolation method that evaluates losses at different flux linkage levels along the constant torque line.

[0016] Based on specific characteristics, temporary magnetic link targets It was identified as:

[0017] in, I 1 The current was measured at the point of the first event. It is the reference magnet link at the point of first-time event. I 2 The current was measured at the second time event. It is the reference magnet link at the second time event. I 3 The current was measured at the third time event, and It is the reference magnet link at the third time event.

[0018] Therefore, this method achieves quadratic interpolation of losses at three flux linkage levels and determines the flux linkage level that minimizes losses.

[0019] Based on specific characteristics, the measurement of the current supplied to the motor is performed at four time events: the first time event is when the low-frequency flux linkage signal is zero, the second time event is when the low-frequency flux linkage signal is at its minimum, the third time event is when the low-frequency flux linkage signal is zero, and the fourth time event is when the low-frequency flux linkage signal is at its maximum.

[0020] Therefore, the present invention can also determine the maximum torque per ampere condition under transient conditions, wherein the torque is not constant over four time events.

[0021] Determine temporary magnetic flux targets based on specific characteristics. :

[0022] Where I3 is the current measured at the third time event. It is the reference magnet link at the third time event. In reference magnet The estimated current flowing at the third time event is as follows. In reference magnet The estimated current will flow at the third time event.

[0023] Based on specific characteristics, in the reference magnetic flux and The estimated current flowing at the third time event. and Based on the current measured at the second, third, and fourth time events and reference torque It is determined based on the estimation of how losses change with torque.

[0024]

[0025] Therefore, although the first, second, third, and fourth measurements are achieved at different torque levels, they can be performed at the same torque levels. Losses are determined at different flux linkage levels under constant torque. Quadratic interpolation can be performed to determine the losses at... The MTPA condition is obtained under constant torque, even though the torque is unstable. MTPA can be obtained under transient torque conditions.

[0026] Based on specific characteristics, the torque is measured at the first and third time events. and loss To estimate how losses change with torque.

[0027]

[0028] Therefore, the variation of losses with torque is readily determined under the same flux linkage reference and is not biased by changes in losses caused by operations with different flux linkage levels. A good estimate of the variation of losses with torque yields a good determination of the isotorque curve, from which the MTPA can be extracted.

[0029] Based on specific characteristics, the torque-flux dependence rule is defined according to the following formula: .

[0030] in, It is a temporary magnet link target ,in, It is the minimum flux linkage applied to a synchronous reluctance motor. It is a torque reference, and It was identified as:

[0031] in, It is the maximum torque per ampere of the flux, measured from the measured current.

[0032] Therefore, the torque-flux dependence rule deviates from the true MTPA rule in the low torque region to ensure that the motor has sufficient magnetization to provide salient polarity and stable control. In the higher torque region, the torque-flux dependence rule intersects the true MTPA curve at the torque reference and approaches the MTPA condition near that torque reference.

[0033] Based on specific characteristics, the torque-flux dependence rule is defined according to the following formula: .

[0034] in, It is a temporary magnet link target. ,and ,in, Corresponding to the corresponding first and second time events, in the continuous torque target The estimated MTPA flux is determined.

[0035] Therefore, the torque-flux dependence rule is tangent to the true MTPA curve near the torque reference. Over a large area around the torque reference, the distance to the true MTPA curve is minimized. Attached Figure Description

[0036] The features of the present invention will become more apparent from the following description of exemplary embodiments, which is made with reference to the accompanying drawings, wherein: Figure 1 An example illustrating the architecture of a motor controller according to the present invention; Figure 2 An example of a block diagram showing the DFVC controller used in this invention; Figure 3 Example of a block diagram showing a magnetic flux observer according to the present invention; Figure 4 An example of a block diagram illustrating an MTPA module according to the present invention; Figure 5a This represents a first example of copper loss estimation at three consecutive time points according to the present invention; Figure 5b This illustrates a second example of copper loss estimation at three consecutive time points according to the present invention; Figure 6a This represents a quadratic interpolation performed according to the first example of the invention; Figure 6b This represents a quadratic interpolation performed according to the second example of the invention; Figure 7aThis represents a first example of estimation based on the dependency rule according to the present invention; Figure 7b A second example of estimation based on the dependency rule according to the present invention; Figure 8 A second example of the architecture of a motor controller according to the present invention is shown; Figure 9 An example of an algorithm for controlling a motor according to the present invention is shown. Detailed Implementation

[0037] Figure 1 An example illustrating the architecture of a motor controller according to the present invention.

[0038] The motor controller includes a torque determination module 100, an MTPA module 101, a DFVC controller 102, a PWM inverter 103 for driving the motor 105, and a measurement module 105.

[0039] The torque determination module 100 consists, for example, a subtraction module and a proportional-integral filter.

[0040] The subtraction module will use the speed reference value. Subtract velocity estimate The output of the subtraction module is provided to a proportional-integral filter to provide a torque reference. This makes the estimated speed equal to reference speed level .

[0041] The coefficients of a proportional-integral (PI) filter are based on the bandwidth of the PI filter. To set it up.

[0042]

[0043] in, It is the mechanical inertia of the system. For example, ,in, Belongs to the [1..20]Hz range. Speed ​​control frequency. The choice of the proportional-integral filter relates to the expected dynamics of the filter, i.e., the ability to drive the estimated velocity toward the reference velocity within a given time window.

[0044] The torque reference is provided to the MTPA determination module 101, and the MTPA determination module 101 obtains the torque reference. and αβ Measurement current vector in coordinate system I αβ Determine the reference level of magnetic flux and current reference level . Reference Figure 4 The MTPA determination module has been disclosed in more detail.

[0045] αβ Measurement current vector in coordinate system I αβ Provided by measurement module 105, which measures three phases. abc Motor current vector i abc And will be in three phases abc Motor current measured in i abc Converted to αβ The measured motor current vector in the coordinate system I αβ .

[0046] Magnet flux reference level and current reference level It is provided to DFVC controller 102, DFVC controller 102 will α β Voltage in coordinate system It is provided to the PWM (Pulse Width Modulation) module 103 to drive the motor 105.

[0047] Reference Figure 2 The DFVC controller 102 is disclosed in more detail.

[0048] Figure 2 An example of a block diagram illustrating the DFVC controller used in this invention.

[0049] The DFVC controller 102 consists of a DFVC module 201, a coordinate system transformation module 202, a magnetic flux observer module 200, and a differential module 203.

[0050] The magnetic flux observer 200 is based on the data provided by the measurement module 105. αβ Measurement current vector in coordinate system I αβ and αβ Voltage in coordinate system Determine flux estimation Current estimation and the estimated direction of the flux linkage vector . Reference Figure 3 The magnetic flux observer was disclosed in more detail.

[0051] flux linkage estimation and current estimation It is provided to DFVC module 201.

[0052] DFCV module 201 estimates based on flux linkage Current estimation Magnetic flux reference level and current reference level Sure fτ Voltage in coordinate system For example, DFVC module 201 is a pair of proportional-integral (PI) compensators.

[0053] Then, the coordinate system transformation module 202 uses the estimated direction of the magnetic flux vector. Will Voltage converted to αβ Voltage in coordinate system .

[0054] Estimated direction of flux linkage vector Provided with a definite velocity estimate Differential module 203.

[0055] Figure 3 An example of a block diagram illustrating a magnetic flux observer according to the present invention.

[0056] The flux linkage observer 200 includes a resistance multiplication module 300, two subtraction modules 301 and 302, an integrator 303, a multiplication module 304, a norm determination module 305, an arctangent module 306, and a collector-emitter module. αβ coordinate system to fτ Coordinate system transformation module 307.

[0057] The resistance multiplication module 300 calculates the resistance based on the motor's resistance and phase current. Determine the voltage drop R. The motor's resistance is taken from the motor's datasheet, for example.

[0058] αβ Measurement current vector in coordinate system It is provided to the resistor multiplication module, which will... αβ Measurement current vector in coordinate system Convert to αβ Voltage in coordinate system V αβ .

[0059] αβ Measurement current vector in coordinate system It is provided to the conversion module 307.

[0060] Subtraction module 301 will αβ Voltage in coordinate system and αβ Voltage in coordinate system V αβ Subtract.

[0061] Subtraction module 302 subtracts the output provided by multiplication module 304 from the output of subtraction module 301. Multiplication module 304 then... Estimated flux linkage in coordinate system Multiply by a coefficient k obs .

[0062] The output of subtraction module 303 is provided to integrator 303, which provides... αβ Flux Linkage Estimation in Coordinate System The estimate.

[0063] αβ Flux Linkage Estimation in Coordinate System Provided for providing flux linkage estimation Norm determination module 305.

[0064] αβ Flux Linkage Estimation in Coordinate System The magnetic flux linkage vector is provided to the arctangent module 306, which determines the direction of the magnetic flux linkage vector. .

[0065] The conversion module 307 uses the direction of the magnetic flux vector. Will αβ Measurement current vector in coordinate system Convert to τ Measuring current on the shaft .

[0066] Therefore, the flux observer estimates the flux norm. Perpendicular to the magnetic flux The current, and estimate flux linkage angle in coordinate system .

[0067] For example, flux linkage observer 200 integrates the reference voltage minus ohmic losses. Drift DC noise is suppressed by a high-pass filter composed of modules 302, 303, and 304. The measured current vector is rotated to obtain the estimated flux linkage angle. .

[0068] Figure 4 An example of a block diagram illustrating an MTPA module according to the present invention.

[0069] The MTPA module 101 includes a norm determination module 400, a QI interpolation module 401, a torque / flux dependence rule module 402, a low-frequency flux signal injection module 403, an addition module 404, a multiplication module 405, and a divider 406.

[0070] Norm determination module 400 determination αβ Measurement current vector in coordinate systemI αβ The norm of . αβ Measurement current vector in coordinate system I αβ The norm is provided to determine the estimated MTPA flux. λ MTPA The QI interpolation module 401 estimates the MTPA flux linkage. λ MTPA It is provided to the torque / flux dependence rule module 402.

[0071] Torque-flux dependence rule module 402 determines the temporary flux reference level. For example, the torque-flux dependence rule is a linear function of the absolute value of the torque: .

[0072] The low-frequency flux linkage signal injection module 403 provides a low-frequency flux linkage signal. For example, the low-frequency flux linkage signal is a sinusoidal signal with an amplitude equal to 2% of the rated flux linkage and a frequency of 25 Hz, or a frequency located at the edge of the human ear's frequency bandwidth.

[0073] Low-frequency flux linkage signal and temporary flux linkage reference level Summation is performed by adder module 404 to provide a magnetic flux reference level. .

[0074] Multiplier 405 will use torque reference Multiply by a constant. This constant value is the number of pole pairs of motor 104.

[0075] Divider 406 divides the output of multiplier 405 by the flux reference level. In order to obtain τ Reference current on the shaft .

[0076] The copper loss consumed by motor 104 is estimated to be proportional to the square of the measured current.

[0077] like Figure 5a As shown, copper loss is estimated at least at three consecutive time events with different flux injection levels.

[0078] Figure 5a This represents a first example of copper loss estimation at three consecutive time points according to the present invention.

[0079] The horizontal axis represents time, and the vertical axis represents the magnetic flux reference level. .

[0080] Magnet flux reference level and The value is determined by the low-frequency flux linkage signal and the temporary flux linkage reference level. The sum is generated.

[0081] like Figure 5a As shown, copper loss is estimated at three consecutive time events, labeled t1, t2, and t3, with different flux injection levels. The first time event t1 is when the low-frequency flux signal is at its maximum, the second time event t2 is when the low-frequency flux signal is at its zero, and the third time event t3 is when the low-frequency flux signal is at its minimum.

[0082] The curve marked 500 represents the torque value during the estimation period.

[0083] The curve marked 501 represents the magnetic flux reference level. The curve marked 502 represents the curve formed before the LF flux linkage signal injection. Magnet link reference generated by rule block 402 .

[0084] In the first example, estimation is performed when the torque reference observed over consecutive time events is not modified. Otherwise, the estimation step is skipped. This invention uses quadratic interpolation to identify the flux linkage value that minimizes the power delivered to motor 104 under the target torque condition, which will refer to... Figure 6a Describe it.

[0085] Figure 5b This is a second example illustrating the estimation of copper loss at four consecutive time points according to the present invention.

[0086] The horizontal axis represents time, and the vertical axis represents the magnetic flux reference level. .

[0087] Magnet flux reference level and The value is determined by the low-frequency flux linkage signal and the temporary flux linkage reference level. The sum is generated.

[0088] like Figure 5b As shown, copper loss is estimated at at least four consecutive time events labeled t11, t12, t13, and t14, which have different flux injection levels.

[0089] The curve marked 550 represents the torque value during the estimation period. For example... Figure 5b As shown, the torque references T1, T2, T3, and T4 are different at time events labeled t11, t12, t13, and t14.

[0090] The curve marked 551 represents the magnetic flux reference level. The curve marked 552 represents the curve formed before the LF flux linkage signal injection. Magnet link reference generated by rule block 402 .

[0091] Figure 6a This indicates a quadratic interpolation performed according to the first example of the present invention.

[0092] like Figure 6a As shown, the present invention uses quadratic interpolation to identify the flux linkage value that minimizes the power delivered to the motor 104 under the target torque condition.

[0093] Best Magnet Linkage Reference Based on the magnetic flux reference using the following formula and and the measured current levels at times t1, t2, and t3. Sure:

[0094] in, I 1 The current measured in the first instant of the event. It is the reference magnet link for the first-time event. I 2 The current is measured at the second time event. It is the reference magnet link for the second time event. I 3 The current is measured at the second time event. It is the reference magnet link for the third time event.

[0095] Figure 6b This indicates a quadratic interpolation performed according to the second example of the present invention.

[0096] The curve marked 650 represents constant torque T1, the curve marked 651 represents constant torque T2, the curve marked 652 represents constant torque T3, and the curve marked 653 represents constant torque T4.

[0097] point( ( ( ( The measurement points are at times t1, t2, t3, and t4, and the point ( ( It is an estimate of the power response curve at time t3 under constant torque T3.

[0098] In the implemented example, the MTPA flux linkage is estimated when the torque references differ in the first and third events. Based on the first and third loss estimates, the relationship between loss and torque under constant flux linkage is determined, since the losses are estimated under the same flux linkage conditions but different torque levels.

[0099]

[0100] Based on the second and fourth loss estimates And based on the obtained loss-torque relationship For the second and fourth flux linkage conditions, the constant torque is determined. Loss estimation .

[0101]

[0102] Then, the torque is identified using the loss estimate under the assumption of constant torque, through the following formula. The following MTPA conditions:

[0103] Figure 7a This represents a first example of an estimation based on the dependency rules according to the present invention.

[0104] Based on the first example estimate, ,in, This is the minimum flux linkage applied to the synchronous reluctance motor. The parameters are updated using only the estimated MTPA flux linkage and torque target. :

[0105] Figure 7b This is a second example of an estimation based on the dependency rule according to the present invention.

[0106] According to the second implementation example, two parameters Everything has been updated. ,and ,in, Corresponding to the continuous torque target The estimated MTPA flux is determined.

[0107] Through the update of dependency rules, the magnet link target It naturally matches the MTPA conditions.

[0108] When the torque target changes, the torque-flux dependency rule ensures the stability of the control loop, and for torque targets close to the previous torque target, the operation does not deviate far from the true MTPA condition. As the torque target changes further, some deviation from the true MTPA condition may occur. This deviation will eventually disappear because the torque-flux dependency rule is updated using the estimated MTPA condition for the new torque target.

[0109] In one variant, the dependency rule and flux linkage are piecewise linear, with different slopes at different torque intervals, to better fit the curvature of the MTPA curve relative to flux linkage over the entire torque range of the motor.

[0110] Figure 8 This is a second example of the architecture of a motor controller according to the present invention.

[0111] The motor controller 80 has, for example, components connected by a bus 801 and components such as... Figure 9 The architecture of the program-controlled processor 800 is disclosed in the paper.

[0112] Bus 801 links processor 800 to read-only memory ROM 802, random access memory RAM 803, and input / output (I / O) interface 805.

[0113] The input / output (I / O) interface 805 enables the motor controller 80 to sense a signal representing the current flowing through the motor 104.

[0114] Memory 803 includes components designed to receive and, for example Figure 9 The program's variables and instruction registers related to the publicly disclosed algorithm.

[0115] Read-only memory or possibly flash memory 802 contains, for example, Figure 9 The program instructions related to the disclosed algorithm are loaded into the random access memory 803 when the motor controller 80 is powered on. Alternatively, the program can also be executed directly from the ROM memory 802.

[0116] The calculations performed by the motor controller 80 can be implemented in software by a set of instructions or programs executed by a programmable computing machine (such as a PC (personal computer), DSP (digital signal processor), or microcontroller); or in hardware by a machine or special-purpose component (such as an FPGA (field-programmable gate array) or ASIC (application-specific integrated circuit)).

[0117] In other words, the motor controller 80 includes circuitry or means including circuitry, causing the motor controller 80 to perform actions as described above. Figure 9 The publicly available algorithm-related programs.

[0118] Figure 9 An example of an algorithm for controlling a motor according to the present invention is shown.

[0119] This algorithm is executed iteratively, for example, at twice the switching frequency. For a switching frequency of 10 kHz, the algorithm will be executed at 20 kHz.

[0120] In step S90, the motor controller determines the torque reference.

[0121] At step S91, the motor controller uses the torque-flux dependency rule to determine a temporary flux target from the torque reference.

[0122] For example, the torque-flux dependence rule is a linear function of the absolute value of the torque.

[0123] For example, the current supplied to the motor is measured at three time events: the first time event is when the low-frequency flux linkage signal is at its maximum, the second time event is when the low-frequency flux linkage signal is at its zero, and the third time event is when the low-frequency flux linkage signal is at its minimum. Temporary flux linkage target. It was identified as:

[0124] in, I 1 The current was measured at the point of the first event. It is the reference magnet link at the point of first-time event. I 2 The current was measured at the second time event. It is the reference magnet link at the second time event. I 3 The current was measured at the third time event, and It is the reference magnet link at the third time event.

[0125] For example, the current supplied to the motor is measured at four time events: the first time event is when the low-frequency flux linkage signal is zero, the second time event is when the low-frequency flux linkage signal is at its minimum, the third time event is when the low-frequency flux linkage signal is zero, and the fourth time event is when the low-frequency flux linkage signal is at its maximum. A temporary flux linkage target is then determined. :

[0126] in, I 3 The current was measured at the third time event. It is the reference magnet link at the third time event. In reference magnet The estimated current flowing at the third time event is as follows. In reference magnet The estimated current will flow at the third time event.

[0127] In step S92, the motor controller determines the flux target by adding a low-frequency flux signal to the temporary flux target.

[0128] In step S93, the motor controller uses direct flux vector control to control the motor according to the flux target.

[0129] In step S94, the motor controller measures the current supplied to the motor at at least three time events, where the time event is the time when the low-frequency flux linkage signal has different values.

[0130] At step S95, the motor controller estimates the flux linkage per ampere for the maximum torque based on the measured current.

[0131] At step S96, the motor controller updates the dependency rule based on the flux linkage per ampere of the estimated maximum torque.

Claims

1. A method for tracking the minimum torque per ampere condition of a motor controlled using direct flux vector control, characterized in that, The method includes the following steps: - Determine the torque reference. - Determine the temporary flux target from the torque reference using the torque-flux dependency rule. - The flux linkage target is determined by adding a low-frequency flux linkage signal to the temporary flux linkage target. - The motor is controlled using the direct flux vector control based on the flux target. - Measure the current supplied to the motor at at least three time events, where the low-frequency flux linkage signal has different values. - Estimate the flux linkage per ampere for maximum torque from the measured current. - Update the dependency rule based on the estimated flux linkage per ampere of the maximum torque.

2. The method according to claim 1, characterized in that, The torque-flux dependence rule is a linear function of the absolute value of the torque.

3. The method according to claim 2, characterized in that, The measurement of the current supplied to the motor is performed at three time events under the same torque conditions: the first time event is when the low-frequency flux linkage signal is at its maximum, the second time event is when the low-frequency flux linkage signal is at its zero, and the third time event is when the low-frequency flux linkage signal is at its minimum.

4. The method according to claim 3, characterized in that, Temporary magnetic link target It was identified as: in, I 1 The current measured at the first time event. It is the reference magnet link at the first time event. I 2 The current is measured at the second time event. It is the reference magnet link at the second time event. I 3 It is the current measured at the third time event, and It is the reference magnet at the third time event.

5. The method according to claim 2, characterized in that, The measurement of the current supplied to the motor is performed at four time events: the first time event is when the low-frequency flux linkage signal is zero, the second time event is when the low-frequency flux linkage signal is at its minimum, the third time event is when the low-frequency flux linkage signal is zero, and the fourth time event is when the low-frequency flux linkage signal is at its maximum.

6. The method according to claim 5, characterized in that, Temporary magnetic link target It was identified as: in, I 3 The current measured at the third time event. It is the reference magnetic link at the third time event. In reference magnet The estimated current that will flow at the third time event is as follows. In reference magnet The estimated current that will flow at the third time event.

7. The method according to claim 6, characterized in that, In reference magnetic flux and The estimated current flowing at the third time event. and Based on the current measured at the second time event, the third time event, and the fourth time event. and reference torque Determined based on estimates of how losses change with torque. 。 8. The method according to claim 7, characterized in that, Based on the torque measured at the first time event and the third time event and loss To estimate the change of the loss with torque 。 9. The method according to claim 4 or 6, characterized in that, The torque-flux dependence rule is defined according to the following formula: ， in, It is a temporary magnet link target. ,in, It is the minimum flux linkage applied to a synchronous reluctance motor. It is a torque reference, and It was identified as: in, It is the maximum torque per ampere of the flux, measured from the measured current.

10. The method according to claim 4 or 6, characterized in that, The torque-flux dependence rule is defined according to the following formula: ， in, It is a temporary magnet link target. ,and ,in, Corresponding to the corresponding first and second time events, in the continuous torque target The estimated MTPA flux is determined below.

11. A device for tracking the minimum torque per ampere condition of a motor controlled using direct flux vector control, characterized in that, The device includes: - A device for determining a torque reference. - A means for determining a temporary flux target from the torque reference using torque-flux dependency rules. - A means for determining a flux linkage target by adding a low-frequency flux linkage signal to the temporary flux linkage target. - A means for controlling a motor using the direct flux vector control according to the flux target. - A means for measuring the current supplied to the motor at at least three time events, said time events being the times when the low-frequency flux linkage signal has different values. - A device for determining the flux linkage with maximum torque per ampere from a measured current. - A means for updating the dependency rule based on the flux linkage per ampere of the estimated maximum torque.

Images