Sliding-mode observer for permanent magnet synchronous motor

Abstract

The invention relates to a sliding mode observer for a permanent magnet synchronous motor. The sliding-mode observer calculates stator current estimation values of the motor in the alpha axial direction and the beta axial direction by using a stator current equation under an alpha-beta static coordinate system, and the extended counter electromotive force estimation values in the alpha axial direction and the beta axial direction are related to the stator current estimation values in the corresponding axial directions, the variable speed reaching law is applied to calculation of the alpha-axial and beta-axial extended counter electromotive force estimated values, system buffeting can be reduced, the calculation amount can be reduced compared with an existing sliding mode observer, and when the alpha-axial and beta-axial extended counter electromotive force estimated values are calculated, the system buffeting can be reduced. According to whether the absolute value of the current error in the alpha axial direction and the absolute value of the current error in the beta axial direction exceed the corresponding switching point or not, segmented calculation is carried out, the control gain of the alpha axial extension counter electromotive force estimation value and the control gain of the beta axial extension counter electromotive force estimation value after the switching point are increased, and therefore the observation precision can be improved, and the motor performance can be improved.

Description

Technical Field

[0001] The present invention relates to the field of motor control, and in particular to a sliding mode observer for a permanent magnet synchronous motor. Background Art

[0002] With the rapid development of power electronics, microelectronics, and permanent magnet material technologies, as well as the deepening of control theory research, permanent magnet synchronous motors (PMSMs) have been widely used in various applications, including new energy vehicles, thanks to their advantages such as small size, light weight, low moment of inertia, and excellent speed regulation. However, the demand for high-quality manufacturing also places higher demands on the control performance of PMSMs.

[0003] Acquiring rotor position and speed information is a critical step in the closed-loop vector control system of a permanent magnet synchronous motor (PMSM). The accuracy of rotor position and speed impacts the performance of the entire PMSM control system. Currently, methods for acquiring position and speed signals can be categorized as either sensored or sensorless. Common sensored methods include encoders and Hall sensors. However, these methods not only increase the complexity of the control system but also reduce system reliability. Consequently, a growing number of sensorless control methods have been proposed, including flux observers, high-frequency signal injection, and sliding mode observers. Among these, the sliding mode observer (SMO) has been widely used in practical engineering applications in recent years due to its simple structure and strong robustness.

[0004] However, existing sliding mode observers usually use a constant velocity reaching law to obtain the estimated value of the extended back electromotive force, which requires a large amount of calculation and the observation accuracy needs to be improved. Summary of the Invention

[0005] The present invention provides a sliding mode observer for a permanent magnet synchronous motor, which can reduce the amount of calculation and achieve higher observation accuracy compared with existing sliding mode observers.

[0006] The sliding mode observer for a permanent magnet synchronous motor provided by the present invention calculates the estimated stator current values ​​of the motor in the α-axis and β-axis directions using the stator current equation in the α-β stationary coordinate system, wherein the estimated stator current value in the α-axis direction has an α-axis current error compared to the actual current in the α-axis direction, and the estimated stator current value in the β-axis direction has a β-axis current error compared to the actual current in the β-axis direction, and the estimated extended back electromotive force values ​​in the α-axis and β-axis directions are related to the estimated stator current values ​​in the corresponding axes; and

[0007] When the absolute value of the α-axis current error does not exceed the first switching point, the α-axis extended back electromotive force estimation value satisfies the following first relationship: α =(KRs )s α +η|s α |sgn(s α );

[0008] When the absolute value of the α-axis current error exceeds the first switching point, increasing the gain coefficient before the sign function sgn(·) in the first relationship, wherein the first switching point is a positive number and is less than the maximum absolute value of the α-axis current error;

[0009] When the absolute value of the β-axis current error does not exceed the second switching point, the β-axis extended back electromotive force estimation value satisfies the following second relationship: β =(KR s )s β +η|s β |sgn(s β );

[0010] When the absolute value of the β-axial current error exceeds the second switching point, increasing the gain coefficient before the sign function sgn(·) in the second relationship, wherein the second switching point is a positive number and is less than the maximum absolute value of the β-axial current error;

[0011] Among them, eα and eβ are the estimated values ​​of the α-axis extended back electromotive force and the β-axis extended back electromotive force, respectively, K and η are positive and adjustable gain parameters, s α and s β are the sliding mode functions of the α-axis and β-axis, R s is the motor resistance.

[0012] Optionally, when the absolute value of the α-axis current error exceeds the first switching point, the α-axis extended back electromotive force estimation value satisfies the following third relationship:

[0013] e α =(KR s )s α +(η'|s α |+(η-η')|Δ1|)sgn(s α ), η'>η, η' is a positive and adjustable gain parameter, and Δ1 is the first switching point.

[0014] Optionally, when the absolute value of the α-axis current error exceeds the first switching point, the α-axis extended back electromotive force estimation value satisfies the following fifth relationship:

[0015] n>1, Δ1 is the first switching point.

[0016] Optionally, when the absolute value of the β-axial current error exceeds the second switching point, the β-axial extended back electromotive force estimation value satisfies the following fourth relationship:

[0017] e β =(KR s )s β +(η'|s β |+(η-η')|Δ2|)sgn(s β ), η'>η, η' is a positive number and

[0018] The adjustable gain parameter, Δ2, is the second switching point.

[0019] Optionally, when the absolute value of the β-axial current error exceeds the second switching point, the β-axial extended back electromotive force estimation value satisfies the following sixth relationship:

[0020] n>1, Δ2 is the second switching point.

[0021] Optionally, the first switching point is equal to the second switching point.

[0022] Optionally, the sliding mode observer calculates a rotor position estimate using the α-axial extended back electromotive force estimate and the β-axial extended back electromotive force estimate, and the rotor position estimate satisfies in, is the estimated value of the rotor position.

[0023] Optionally, the sliding mode observer calculates a speed estimate using the rotor position estimate, and the speed estimate satisfies in, is the estimated value of the rotational speed, s is the differential operator, and Ts is the filtering time constant.

[0024] Optionally, the jitter frequency of the rotational speed estimation value is 4 times the fundamental frequency of the current.

[0025] Optionally, the stator current equation is:

[0026]

[0027] in, and are the estimated values ​​of the stator current in the α-axis and β-axis, L s is the motor inductance, V α and V β are the stator voltages in the α-axis and β-axis directions, e α and e β They are the α-axial extended back electromotive force estimated value and the β-axial extended back electromotive force estimated value respectively.

[0028] The sliding mode observer for a permanent magnet synchronous motor provided by the present invention calculates the estimated stator current values ​​of the motor in the α-axis and β-axis directions using the stator current equation in the α-β stationary coordinate system. The estimated extended back electromotive force values ​​in the α-axis and β-axis directions are related to the estimated stator current values ​​in the corresponding axes. The invention also has the following features and advantages:

[0029] On the one hand, when the absolute value of the α-axis current error does not exceed the first switching point, the α-axis extended back electromotive force estimate satisfies the first relationship. When the absolute value of the β-axis current error does not exceed the second switching point, the β-axis extended back electromotive force estimate satisfies the second relationship. The rotor position estimate and the speed estimate can be further calculated based on the extended back electromotive force estimate. The first relationship and the second relationship apply a variable speed convergence law, which helps to reduce system chattering. Therefore, before obtaining the position estimate, there is no need to weaken the chattering, and then there is no phase lag due to the chattering weakening. Furthermore, there is no need to perform position compensation due to the phase lag. Compared with the existing sliding mode observer, the amount of calculation can be reduced, making the structure of the sliding mode observer more concise.

[0030] On the other hand, when the absolute value of the α-axial current error exceeds the first switching point, the gain coefficient before the sign function sgn(·) in the first relationship is increased, and the first switching point is less than the maximum absolute value of the α-axial current error. When the absolute value of the β-axial current error exceeds the second switching point, the gain coefficient before the sign function sgn(·) in the second relationship is increased, and the second switching point is less than the maximum absolute value of the β-axial current error. In this way, the control gain of the α-axial extended back electromotive force estimation value after the first switching point and the control gain of the β-axial extended back electromotive force estimation value after the second switching point can be increased, thereby improving the observation accuracy and helping to improve the motor performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 4 is a block diagram of a sensorless speed regulation system for a permanent magnet synchronous motor according to an embodiment of the present invention.

[0032] Figure 2 2 is a schematic structural diagram of a sliding mode observer according to an embodiment of the present invention.

[0033] Figure 3 1 is a waveform diagram of the α-axial current error and the β-axial current error in one embodiment of the present invention.

[0034] Figure 4 This figure shows the position estimate and position reference values ​​when calculating the extended back-EMF estimate with and without switching points.

[0035] Figure 5FIG. 1 is a waveform diagram showing how the gain coefficient before the sign function changes with time when the extended back-EMF estimate is calculated with and without a switching point. ...

[0036] Figure 6 1 is a waveform diagram of three-phase current in one embodiment of the present invention.

[0037] Figure 7 1 is a waveform diagram of a rotational speed estimation value obtained with a switching point when calculating an extended back electromotive force estimation value in one embodiment of the present invention. DETAILED DESCRIPTION

[0038] The sliding mode observer of the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description. It should be noted that the drawings are greatly simplified and not to exact scale, and are intended solely to facilitate and clarify the purpose of illustrating the embodiments of the present invention.

[0039] Figure 1 The sensorless speed control system for permanent magnet synchronous motor is shown. Figure 1 , M represents a permanent magnet synchronous motor (PMSM), which uses a three-phase inverter bridge to achieve motor control signal conversion and power conversion. In the sensorless speed regulation system, the three-phase stator current i is measured when the motor is running. abc , and perform Clark transformation to two-phase current i α and i β ,i α and i β mutually orthogonal, i α and i β The current i in the rotating coordinate system is obtained by Park transformation d and i q , the current i d and i q The actual value and reference value of each qref and i dref ) is compared with the error obtained as the input of the corresponding automatic current regulator (ACR), and the voltage vector V is obtained through the automatic current regulator (ACR). q and V d , the voltage vector V q and V d Through the inverse Park transformation into two-phase voltage V α and V β , V α and V β The three-phase voltages V are calculated based on the synthesized voltage vector using the space vector pulse width modulator (SVPWM). a 、V b and V cAnd the conduction time of the switch tubes of each bridge arm of the three-phase inverter bridge, and then the motor is controlled by the three-phase inverter bridge.

[0040] In the above sensorless speed regulation system, the two-phase current i is sampled by the sliding mode observer. α and i β And the two-phase voltage V α and V β , to estimate the rotor position (i.e. rotor angle) and motor speed, and obtain the position estimate and speed estimates The position estimate is used for Park transform and inverse Park transform. Used for speed feedback control, specifically to make the speed estimate With reference speed The error is calculated and the speed error is input into the ASR controller to generate the q-axis reference current i required for current loop control. qref .

[0041] An embodiment of the present invention relates to a sliding mode observer that can be used in the sensorless speed regulation system of the above-mentioned permanent magnet synchronous motor. More specifically, the sliding mode observer is used in the speed regulation system of a surface-mounted permanent magnet synchronous motor. The sliding mode observer uses the stator current equation in the α-β stationary coordinate system to calculate the stator current estimation value of the motor in the α-axis and β-axis directions, wherein the α-axis stator current estimation value has an α-axis current error compared to the α-axis actual current, and the β-axis stator current estimation value has a β-axis current error compared to the β-axis actual current. The extended back electromotive force estimation values ​​of the α-axis and β-axis directions are related to the stator current estimation values ​​of the corresponding axes. The extended back electromotive force estimation values ​​can be used to further calculate the rotor position estimation value and the speed estimation value, thereby realizing rotor position observation and speed observation. In this embodiment of the present invention, a variable-speed reaching law is used to calculate the estimated values ​​of the extended back-EMF in the α-axis and β-axis. This reduces the amount of computation compared to existing sliding mode observers. Furthermore, when calculating the estimated values ​​of the extended back-EMF in the α-axis and β-axis, the gain coefficient preceding the sign function sgn(·) in the calculation formula varies depending on whether the absolute value of the axial current error in the corresponding axis is before or after the switching point, thereby improving observation accuracy. A detailed description is as follows.

[0042] The current state equation of the surface-mounted permanent magnet synchronous motor in the α-β stationary coordinate system is shown in formula (1-1):

[0043]

[0044] In formula (1-1), i α is the α-axial stator current, i β is the β-axis stator current, t is time, V αis the α-axis stator voltage, V β is the β-axis stator voltage, E α is the α-axial expansion back electromotive force, E β is the β-axial expansion back electromotive force, R s is the motor resistance, L s The internal sense of the motor.

[0045] α axial expansion back electromotive force E α and β axial expansion back electromotive force E β It can be expressed by formula (1-2):

[0046]

[0047] In formula (1-2), is the permanent magnet flux, θ e is the rotor position, ω e is the electrical angular velocity.

[0048] Figure 2 FIG. 4 shows the structure of the sliding mode observer according to an embodiment of the present invention. Figure 2 In the embodiment of the present invention, the stator current equation in the α-β stationary coordinate system is shown in Formula (1-3):

[0049]

[0050] In formula (1-3), is the estimated value of the α-axis stator current, is the estimated value of the β-axis stator current, e α is the estimated value of the α-axial extended back electromotive force, e β Estimated value of β-axis extended back EMF.

[0051] In this embodiment, the sliding mode function is designed as shown in formula (1-4):

[0052]

[0053] In formula (1-4), is the difference between the estimated value of the α-axis stator current and the actual α-axis current, recorded as the α-axis current error, is the difference between the estimated value of the β-axis stator current and the actual β-axis current, which is recorded as the β-axis current error.

[0054] In this embodiment, according to the α axial current error If the absolute value of exceeds the first switching point Δ1, the estimated back electromotive force value e will be extended to the α axis. α The calculation is divided into two stages, and the difference between the two stages is that the gain coefficient before the sign function sgn(·) in the calculation formula is α axial current error. When the absolute value of is greater than the first switching point Δ1, it increases; according to the β axial current error If the absolute value of exceeds the second switching point Δ2, the estimated value of the back electromotive force e is extended to the β axis. β The calculation is divided into two stages, and the difference between the two stages is that the gain coefficient before the sign function sgn(·) in the calculation formula is adjusted to the β-axis current error. increases when the absolute value of is greater than the second switching point Δ2.

[0055] The first switching point Δ1 and the second switching point Δ2 are determined by the axial current errors in the α-axis and the β-axis. When setting or adjusting the first switching point Δ1 and the second switching point Δ2, the α-axis current error can be obtained from the historical data of the sliding mode observer. The absolute maximum value and β axial current error The absolute maximum value of the first switching point Δ1 is positive and less than the α axial current error The absolute maximum value of the second switching point Δ2 is positive and less than the β axial current error The historical data may be derived from the α-axis current error calculated using the first switching point Δ1 and the second switching point Δ2. Absolute value data and β axial current error The absolute value data can also be obtained when the switching point is not used (that is, when calculating the extended back EMF estimate, the gain coefficient before the sign function sgn(·) is no longer based on the α-axis current error or β axial current error The α-axial current error calculated by Absolute value data and β axial current error The absolute value data of α axial current error is shown in formula (1-5). The absolute value and β axial current error The calculation formula for the estimated values ​​of the α-axial and β-axial extended back electromotive force when the absolute value of is less than the corresponding switching point.

[0056] Formula (1-5) is as follows:

[0057]

[0058] In formula (1-5), K and η are positive and adjustable gain parameters, where α is the estimated value of the back electromotive force e α Using the first relational expression at the top of equation (1-5), is the α axial current error The absolute value of Δ1 is the first switching point, and the β-axial extended back electromotive force estimated value e βUsing the second relational expression at the bottom of equation (1-5), is the β axial current error The absolute value of , Δ2 is the second switching point.

[0059] Subtracting equation (1-3) from equation (1-1) and combining it with equation (1-4) yields the stator current error equation, as shown in equation (1-6):

[0060]

[0061] In order to detect the convergence performance of the sliding mode observer, the Lyapunov function is defined as shown in formula (1-7):

[0062]

[0063] Taking the derivative of both sides of the equal sign of formula (1-7), we get formula (1-8):

[0064]

[0065] Substituting equations (1-6) and (1-5) into equation (1-8), we can obtain equation (1-9):

[0066]

[0067] According to formula (1-9), when and hour, The system is convergent. Therefore, the sliding mode observer of this embodiment can ensure and Converges to a bounded range. If the value of (K+η) is increased, the bounded range will be reduced. When the bounded range is small enough, e α ≈E α , e β ≈E β .

[0068] like Figure 1 and Figure 2 As shown, when the sliding mode observer is working, the two-phase current i α and i β And the two-phase voltage V α and V β, the relationship between the estimated value of the extended back electromotive force in the α-axis and β-axis and the sliding mode function (the relationship includes the above-mentioned formula (1-5)) and the equation of the sliding mode function (i.e., formula (1-4)) are substituted into formula (1-3), and an equation including only the estimated value of the stator current in the α-axis as an unknown quantity and an equation including only the estimated value of the stator current in the β-axis as an unknown quantity can be obtained, thereby calculating the estimated value of the stator current in the α-axis and β-axis, and further calculating the above-mentioned sliding mode function, and then further solving the estimated value of the rotor position. and speed estimates

[0069] Reference Figure 2 , the estimated rotor position The inverse tangent function can be used for calculation, as shown in formula (1-10):

[0070]

[0071] Further estimates can be made based on the position Calculate an estimate of the rotor's speed The speed estimate can be adjusted as needed In one embodiment, the speed estimate is not filtered. Filtering is performed, the position estimate can be directly Derivative to get the speed estimate The present invention is not limited to this. Considering that as the bounded range is reduced, the vibration is strengthened and the vibration of the rotation speed is increased, therefore, in this embodiment, when calculating the rotation speed estimation value A low-pass filter is introduced when the input voltage is 0.01 V, as shown in formula (1-11):

[0072]

[0073] In Equation (1-11), s is the differential operator, and Ts is the filter time constant. Furthermore, to reduce chattering, a saturation function can be used instead of a sign function when calculating the estimated α-axis extended back-EMF and the estimated β-axis extended back-EMF.

[0074] In this embodiment, the estimated value of the extended back electromotive force on the α-axis is e α and the estimated extended back EMF value of the β axis e β The calculation of the speed reaching law (refer to formula (1-5)) is applied to reduce the system chattering, so as to obtain the position estimate Previously, there was no need to perform vibration reduction, and the speed estimate The filtering is used to obtain the position estimate Therefore, it will not cause phase lag and does not need position compensation, which helps to reduce the amount of calculation.

[0075] Figure 3 The waveforms of the α-axis current error and the β-axis current error are shown. Figure 3 The α-axis current error and the β-axis current error have waveforms similar to sine and cosine. In this embodiment, by introducing a first switching point Δ1 and a second switching point Δ2, the sliding mode observer increases the gain coefficient before the sign function sgn(·) in the first and second relationships in equations (1-5) above after the absolute value of the α-axis current error exceeds the first switching point Δ1 and the absolute value of the β-axis current error exceeds the second switching point Δ2, thereby improving the accuracy of position observation.

[0076] In one embodiment, after the absolute value of the α-axis current error exceeds the first switching point Δ1, the gain coefficient before the sign function sgn(·) in the above first relationship increases linearly and remains continuous. At this time, as an example, the α-axis expanded back electromotive force estimation value e α Calculate using the following third relationship: e α =(KR s )s α +(η'|sα|+(η-η')|Δ1|)sgn(s α ), η'>η, η' is a positive and adjustable gain parameter. Combining the above first and third relationships, the control input of the α-axis extended back electromotive force estimation value eα in this embodiment is obtained, as shown in formula (1-12):

[0077]

[0078] In one embodiment, after the absolute value of the β-axis current error exceeds the second switching point Δ2, the gain coefficient before the sign function sgn(·) in the above second relationship increases linearly. At this time, as an example, the β-axis expanded back electromotive force estimation value e β Use the following fourth relationship to calculate: e β =(KR s )s β +(η'|s β |+(η-η')|Δ2|)sgn(s β ), η'>η, η' is a positive and adjustable gain parameter. Combining the second and fourth relationship equations, the estimated value of the β-axial extended back electromotive force in this embodiment is obtained: β The control input is shown in formula (1-13):

[0079]

[0080] In one embodiment, after the absolute value of the α-axis current error exceeds the first switching point Δ1, the gain coefficient before the sign function sgn(·) in the above first relationship increases nonlinearly and remains continuous. At this time, as an example, the α-axis expanded back electromotive force estimation value e α The calculation is done using the following fifth relationship: n>1. Combining the first and fifth equations, we can obtain the estimated value of the α-axial extended back electromotive force e in this embodiment: α The control input is shown in formula (1-14):

[0081]

[0082] In one embodiment, after the absolute value of the β-axis current error exceeds the first switching point Δ2, the gain coefficient before the sign function sgn(·) in the above second relationship increases nonlinearly and remains continuous. At this time, as an example, the β-axis expanded back electromotive force estimation value e β The following sixth relationship is used for calculation: n>1. Combining the second and sixth equations, we can obtain the estimated value of the β-axial extended back electromotive force e in this embodiment: β The control input is shown in formula (1-15):

[0083]

[0084] The above-mentioned first switching point Δ1 can be selected as a value greater than 0 and less than the maximum absolute value of the α-axial current error as needed, and the second switching point Δ2 can be selected as a value greater than 0 and less than the maximum absolute value of the β-axial current error as needed. Since the amplitudes of the α-axial current error and the β-axial current error are close, the first switching point Δ1 and the second switching point Δ2 can be set to be equal, for example.

[0085] Figure 4 The position estimation value and the position reference value are shown when the extended back electromotive force estimation value is calculated with and without the switching point; wherein, when the switching point is not used, that is, the first switching point Δ1 is not used to increase the gain coefficient before the sign function sgn(·) after the switching point, the extended back electromotive force estimation value is calculated using the above-mentioned first and second relationship formulas, and the gain coefficient of the sign function sgn(·) in the first relationship formula is maintained at η|s α |, the gain coefficient of the sign function sgn(·) in the second relation is kept η|s β |, the position estimate can be obtained by calculation like Figure 4 As shown in "Position estimate 1" in the figure, when there is a switching point, taking the calculation using equations (1-14) and (1-15) as an example, when the absolute value of the α-axis current error exceeds the first switching point Δ1, the gain coefficient of the sign function sgn(·) is When the absolute value of the β-axis current error exceeds the second switching point Δ2, the gain coefficient of the sign function sgn(·) is For example, if n is 3 / 2, the position estimate can be obtained by calculation. like Figure 4 As shown in "Position estimate 2" in the figure. Figure 4 It can be seen that "position estimated value 2" is closer to the position reference value than "position estimated value 1", which means that when the α-axial extended back electromotive force estimation value and the β-axial extended back electromotive force estimation value are calculated with a switching point (i.e., the first switching point Δ1 and the second switching point Δ2 mentioned above), after the absolute value of the corresponding axial current error is greater than the corresponding switching point, the gain coefficient is increased, so that the control gain is increased, which can improve the accuracy of the position estimation value.

[0086] Figure 5 The waveforms of the gain coefficient before the sign function changing with time when the extended back electromotive force estimate is calculated with and without the switching point are shown. Taking formula (1-12) as an example, if the first switching point Δ1 is not used to increase the gain coefficient before the sign function sgn(·) after the switching point, the α-axis extended back electromotive force estimate is calculated using the above-mentioned first relationship. The gain coefficient of the sign function sgn(·) is still η|s when the absolute value of the α-axis current error exceeds the first switching point Δ1. α |, if the first switching point Δ1 is used so that the gain coefficient before the sign function sgn(·) is (η'|s α |+(η-η')|Δ1|). Figure 5 As shown, since the gain coefficient increases when the absolute value of the α-axis current error exceeds the first switching point Δ1, the jitter of the speed estimation value will also increase. Therefore, when the speed estimation value is During filtering, when the absolute value of the α-axis current error exceeds a first switching point Δ1 and when the absolute value of the β-axis current error exceeds a second switching point Δ2, the filtering time constant may be reduced.

[0087] Figure 6 and Figure 7 The waveforms of the three-phase current and the speed estimation obtained by calculating the extended back electromotive force estimation value with the switching point are shown respectively. The waveform of Figure 6 As shown in Figure 2, when the time is from 0.23s to 0.24s, the current fundamental wave completes one cycle. Correspondingly, as Figure 7 As shown, when the time is from 0.23s to 0.24s, the estimated speed The four fluctuations are caused by the use of switching points (i.e., the first switching point Δ1 and the second switching point Δ2) when calculating the extended back electromotive force estimate. When the extended back electromotive force estimate is calculated with the switching points, the corresponding speed estimate is The chattering frequency is four times the fundamental frequency of the current.

[0088] The sliding mode observer described in the above embodiment can be implemented by a hardware circuit and / or by a program set on a storage medium. The hardware circuit may include one or more computers, hardware and devices that implement the above functions. When implemented by a computer program set on a storage medium, the computer program may be a program for executing each step in the computer, a program for forming the computer function into each tool, or a program for enabling the computer to implement each of the above functions. The storage medium may include a hard disk, a random access memory (RAM), an external storage medium, a storage device via a communication line, a register in a central processing unit (CPU), etc. Whether in software or hardware, its individual parts can be implemented by people familiar with the electronics and software fields, so its details will not be repeated in this specification.

[0089] In the sliding mode observer described in the above embodiment, the stator current equation in the α-β stationary coordinate system is used to calculate the estimated stator current values ​​of the motor in the α-axis and β-axis directions. The extended back electromotive force estimated values ​​in the α-axis and β-axis directions are related to the stator current estimated values ​​in the corresponding axes. On the one hand, when the absolute value of the α-axis current error does not exceed the first switching point Δ1, the corresponding extended back electromotive force estimated value satisfies the above first relationship. When the absolute value of the β-axis current error does not exceed the second switching point Δ2, the corresponding extended back electromotive force estimated value satisfies the above second relationship. The rotor position estimated value and the speed estimated value can be further calculated based on the extended back electromotive force estimated value. The first relationship and the second relationship apply the variable speed approach law, which helps to reduce system vibration, so that no jittering is required before obtaining the position estimate. The vibration is weakened, and then no phase lag is generated due to the weakening of the chattering vibration, and further, there is no need to perform position compensation due to the phase lag. Therefore, compared with the existing sliding mode observer, the amount of calculation can be reduced, making the structure of the sliding mode observer simpler. On the other hand, when the absolute value of the α-axial current error exceeds the first switching point Δ1, the gain coefficient before the sign function sgn(· in the first relationship is increased. When the absolute value of the β-axial current error exceeds the second switching point Δ2, the gain coefficient before the sign function sgn(· in the second relationship is increased. In this way, the control gain of the α-axial extended back electromotive force estimate after the first switching point Δ1 and the control gain of the β-axial extended back electromotive force estimate after the second switching point Δ2 can be increased, thereby improving the observation accuracy and contributing to improving the motor performance.

[0090] The above description is only a description of the preferred embodiment of the present invention, and does not limit the scope of the rights of the present invention. Any person skilled in the art can make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the scope of protection of the technical solution of the present invention.

Claims

1. A sliding mode observer for a permanent magnet synchronous motor, characterized in that: The sliding mode observer calculates the stator current estimation values ​​of the motor in the α-axis and β-axis directions using the stator current equation in the α-β stationary coordinate system, wherein the α-axis stator current estimation value has an α-axis current error compared to the α-axis actual current, the β-axis stator current estimation value has a β-axis current error compared to the β-axis actual current, and the α-axis and β-axis extended back electromotive force estimation values ​​are related to the stator current estimation values ​​of the corresponding axes; and When the absolute value of the α-axis current error does not exceed the first switching point, the α-axis extended back electromotive force estimation value satisfies the following first relationship: α =(KR s )s α +η|s α |sgn(s α ); When the absolute value of the α-axis current error exceeds the first switching point, increasing the gain coefficient before the sign function sgn(·) in the first relationship, wherein the first switching point is a positive number and is less than the maximum absolute value of the α-axis current error; When the absolute value of the β-axis current error does not exceed the second switching point, the β-axis extended back electromotive force estimation value satisfies the following second relationship: β =(KR s )s β +η|s β |sgn(s β ); When the absolute value of the β-axial current error exceeds the second switching point, increasing the gain coefficient before the sign function sgn(·) in the second relationship, wherein the second switching point is a positive number and is less than the maximum absolute value of the β-axial current error; Among them, e α and e β are the estimated values ​​of the α-axis extended back electromotive force and the β-axis extended back electromotive force respectively, K and η are positive and adjustable gain parameters, s α and s β are the sliding mode functions of the α-axis and β-axis, R s is the motor resistance.

2. The sliding mode observer according to claim 1, wherein When the absolute value of the α-axis current error exceeds the first switching point, the α-axis extended back electromotive force estimation value satisfies the following third relationship: e α = (K - R s )s α + (η'|s α | + (η - η')|Δ1|) sgn(s α ),η'>η,η' is a positive number and The adjustable gain parameter, Δ1, is the first switching point.

3. The sliding mode observer according to claim 1, wherein When the absolute value of the α-axis current error exceeds the first switching point, the α-axis extended back electromotive force estimation value satisfies the following fifth relationship: n>1, Δ1 is the first switching point.

4. The sliding mode observer according to claim 1, wherein When the absolute value of the β-axis current error exceeds the second switching point, the β-axis extended back electromotive force estimation value satisfies the following fourth relationship: e β =(K-Rs)s β +(η'|s β |+(η-η')|Δ2|)sgn(s β ), η'>η, η' is a positive and adjustable gain parameter, and Δ2 is the second switching point.

5. The sliding mode observer according to claim 1, wherein When the absolute value of the β-axis current error exceeds the second switching point, the β-axis extended back electromotive force estimation value satisfies the following sixth relationship: n>1, Δ2 is the second switching point.

6. The sliding mode observer according to claim 1, wherein The first switching point is equal to the second switching point.

7. The sliding mode observer according to any one of claims 1 to 6, characterized in that: The sliding mode observer calculates a rotor position estimate using the α-axial extended back electromotive force estimate and the β-axial extended back electromotive force estimate. The rotor position estimate satisfies in, is the estimated value of the rotor position.

8. The sliding mode observer according to claim 7, wherein: The sliding mode observer calculates a speed estimate using the rotor position estimate, and the speed estimate satisfies in, is the estimated value of the rotational speed, s is the differential operator, and Ts is the filtering time constant.

9. The sliding mode observer according to claim 8, wherein: The jitter frequency of the speed estimation value is four times the fundamental frequency of the current.

10. The sliding mode observer according to any one of claims 1 to 6, characterized in that: The stator current equation is: in, and are the estimated values ​​of the stator current in the α-axis and β-axis, L s is the motor inductance, V α and V β are the stator voltages in the α-axis and β-axis directions, e α and e β They are the α-axial extended back electromotive force estimated value and the β-axial extended back electromotive force estimated value respectively.

Images