A speed sensorless control method for PMSM driven by MMC variable frequency drive

Abstract

The present invention discloses a method for controlling a PMSM with a speed sensorless driven by an MMC variable frequency drive, and belongs to the field of AC speed regulation. The method for multiplexing high-frequency signals and weighting functions in a speed sensorless control system based on an MMC driven permanent magnet synchronous motor proposed in this application is specifically as follows: the high-frequency signals in the pulse high-frequency voltage injection method of the speed sensorless in the zero and low-speed stage and the high-frequency signals required to suppress the voltage fluctuation of the submodule capacitor are multiplexed. At the same time, as the PMSM speed increases, that is, as the system frequency increases, the weighting function required for cutting off the injected high-frequency signal is multiplexed with the weighting function required for switching from the pulse high-frequency voltage injection method to the sliding mode observer without the speed sensor, thereby simplifying the system control structure and reducing costs. Ultimately, system control of the permanent magnet synchronous motor based on MMC drive within the full speed range can be achieved.

Description

[0001] manual

[0002] A speed sensorless control method for PMSM driven by MMC variable frequency drive Technical Field

[0003] The present invention relates to the field of AC speed regulation, and more specifically, to a method for multiplexing high-frequency signals and weighted functions in a speed sensorless control system for a permanent magnet synchronous motor (PMSM) driven by a modular multilevel converter (MMC) within a full speed range. Background Art

[0004] In recent decades, with the continuous improvement of industrialization level, the rapid development of permanent magnet materials, power electronics technology and motor control theory, permanent magnet synchronous motors have been increasingly widely used in CNC machine tools, aerospace, industrial robots and other fields due to their superior performance and robustness.

[0005] Permanent magnet synchronous motors (PMSMs) are widely used in AC speed control drives because they require no excitation current and offer high operating efficiency, torque inertia, and power density. However, their high-performance control requires precise rotor position and speed signals to achieve field orientation. Traditional motion control systems typically use photoelectric encoders or resolvers to detect rotor position and speed. However, these mechanical sensors are subject to limitations such as size and mass. In some environments, they are particularly susceptible to environmental factors, which can lead to system instability. Furthermore, position sensors increase system costs. Therefore, eliminating these sensors and adopting sensorless vector control technology for PMSMs has become a hot topic in the AC speed control field.

[0006] Currently, the topology for controlling motors is mostly based on traditional two-level frequency converters, which feature simple structures and relatively easy control methods, making them widely used. However, due to device voltage resistance and switching frequency limitations, these topologies struggle to meet the demands of medium-, high-voltage, and high-power applications. To address this challenge, researchers worldwide are seeking solutions in converter control methods and topologies. Due to its numerous advantages, MMCs have been widely used in areas such as flexible DC transmission and power quality management. In recent years, numerous researchers both domestically and internationally have begun applying them to high-power transmission applications, enabling high-power motor driving. However, when MMCs operate at low frequencies, the voltage fluctuations in the bridge arm submodule capacitors are significant. Since motor startup inevitably involves a low-frequency phase, minimizing these voltage fluctuations in the submodule capacitors during low-frequency operation presents a challenge.

[0007] Currently, MMCs typically inject high-frequency circulating current and high-frequency zero-sequence voltage signals during low-frequency operation to reduce submodule capacitor voltage fluctuations. Sensorless permanent magnet synchronous motors, on the other hand, employ pulsed high-frequency voltage injection and a sliding mode observer at low and medium speeds, respectively. To simplify the system's control structure and reduce costs, further research is needed to address the reuse of high-frequency signals at low and medium speeds in MMC-driven PMSM sensorless control systems, as well as the reuse of weighted functions for speed sensorless switching during full-speed operation. Ultimately, this will lead to the realization of a position sensorless vector control system for MMC-driven PMSMs across the full speed range. Summary of the Invention

[0008] In response to the shortcomings and problems of the existing technology, the present invention provides a speed sensorless control method for PMSM driven by MMC variable frequency drive. The method can realize position sensorless control within the full speed range. Its purpose is to reuse the pulsating high-frequency voltage injection method and the high-frequency signal required to suppress sub-module fluctuations in the zero-speed stage. As the system frequency increases, the MMC needs to gradually cut off the injected high-frequency signal. Correspondingly, the speed sensorless method also needs to switch from the pulsating high-frequency voltage injection method to the sliding mode observer. The switching processes of the two are similar, and their switching functions can be reused.

[0009] To achieve the above object, the technical solution adopted by the present invention is:

[0010] Step 1: Analyze the energy flow of the MMC bridge arm half-bridge sub-module to obtain the fluctuation ΔW of the sub-module fast capacitor voltage ap and DC bus voltage U dc , output current amplitude I m The relationship between and output angular frequency ω;

[0011] Step 2: Under low-frequency conditions, high-frequency circulating current and high-frequency zero-sequence voltage injection are used to suppress capacitor voltage fluctuations.

[0012] Step 3: Set the given speed n * and predicted speed By comparison, the q-axis given current i is obtained through the proportional integral link, i.e., the PI controller q * ;

[0013] Step 4: Set the dq axis to the given current i d * and i q * The i obtained by feedback from dq axis respectively d and i q By comparison, the dq axis voltage u is obtained through the PI current regulator dq ;

[0014] Step 5: dq axis voltage u dq The three-phase modulation voltage u is obtained by Park inverse transformation abc ;

[0015] Step 6: Convert the three-phase modulated voltage u abc The input is used into a reasonable modulation strategy to control the on and off of devices in the MMC inverter to control the permanent magnet synchronous motor.

[0016] Furthermore, in step 1, by analyzing the energy flow of the bridge arm submodule capacitor, the fluctuation ΔW of the submodule fast capacitor voltage can be obtained. ap and DC bus voltage U dc , output current amplitude I m The relationship between and the output angular frequency ω is calculated as follows:

[0017]

[0018] Furthermore, in step 2, the high-frequency circulating current and high-frequency zero-sequence voltage required to be injected are calculated as follows (taking phase a as an example):

[0019] Furthermore, in step 3, the speed is predicted by the pulse high frequency voltage injection method. and sliding mode observer to predict the speed Weighted calculation to get the predicted speed In this application, the high-frequency signal required by the pulsating high-frequency voltage injection method used in the low-speed stage of the permanent magnet synchronous motor and the high-frequency circulating current and high-frequency zero-sequence voltage signal injected when the MMC is running at low frequency can be multiplexed.

[0020] Furthermore, in step 3, the speed is predicted The calculation formula G1 and G2 are and The weighting factor of . The position sensorless control method within the full speed range of this application uses a weighting coefficient algorithm to switch between the pulse high-frequency voltage injection method and the sliding mode observation method. This weighting coefficient algorithm can also be used to switch the high-frequency signal injected during the low-frequency operation of the MMC. ω1-ω2 is the speed switching interval. When ω≤ω1, it is the low-speed operation stage. The speed sensorless method uses the pulse high-frequency voltage injection method, and the MMC operation uses the high-frequency signal injection. At this time, the weighting coefficient G1=1; when ω1<ω<ω2, it is the switching stage, and the speed is predicted at this time. The calculation formula The high-frequency signal injected in the MMC topology is u′=G1*i ah+G2*0, G1+G2=0, G1 and G2 vary linearly with speed. ω≥ω2 represents the medium- and high-speed operation phase. The speed sensor uses a sliding mode observer, and the MMC does not require high-frequency signal injection. In this case, G2=1. The specific speed switching range will be optimized through further experimentation. This will enable sensorless operation of the MMC-driven permanent magnet synchronous motor across the full speed range.

[0021] Further, in step 4, take i * d =0, the current i collected in step 6 abc After 3s / 2r coordinate transformation, the dq axis feedback current i is obtained dq .

[0022] Furthermore, the predicted angular velocity The predicted angle obtained by integration Input into the 3s / 2r coordinate transformation.

[0023] Furthermore, in step 5, the predicted angular velocity Integrate to get the predicted angle Input to the inverse park transform.

[0024] Furthermore, in step 6, a reasonable modulation strategy is adopted according to the three-phase modulation voltage to generate a power device pulse width signal, and then the power device pulse width signal is converted into a three-phase winding current i by a voltage source inverter. abc , sent to the permanent magnet synchronous motor.

[0025] Among them, according to the obtained modulated three-phase voltage u abc The power device pulse width signal is generated through a reasonable modulation strategy, and then the power device pulse width signal is converted into a three-phase winding current i by a voltage source inverter. abc , sent to the permanent magnet synchronous motor.

[0026] The modulation strategy adopts a reasonable modulation strategy to generate a power device pulse width signal according to the obtained three-phase modulation voltage; the voltage source inverter is used to generate a three-phase winding current according to the power device pulse width signal and send it to the permanent magnet synchronous motor.

[0027] Furthermore, based on the principle of pulse high-frequency voltage injection method: a sinusoidal high-frequency voltage signal is injected into the d-axis of the estimated synchronous rotating coordinate system to stimulate the motor to produce an inductor saturation effect, thereby showing the motor salient polarity. The high-frequency response current component is enhanced, and the corresponding demodulation is performed to obtain the rotor position information, that is, the predicted speed is obtained. Integrate the predicted speed to get the predicted angle

[0028] Furthermore, the sliding mode observer-based method can be used to obtain the state equation of the current error system by subtracting the mathematical model designed by the sliding mode observer in the synchronous rotating coordinate system of the three-phase PMSM:

[0029]

[0030] in, and is the current observation error;

[0031] The sliding mode observer is used to estimate the current and the sliding mode surface is defined as When the system enters the sliding mode, the dq-axis induced electromotive force can be obtained:

[0032]

[0033] It can be seen that the q-axis induced electromotive force contains the rotor speed information, that is, The predicted speed is obtained Integrate the predicted speed to get the predicted angle

[0034] Based on the above technical solution, this paper proposes a speed sensorless control multiplexing method for PMSM driven by MMC. The high-frequency signal required by the pulse high-frequency voltage injection method in the low-speed stage of the permanent magnet synchronous motor and the high-frequency circulating current and high-frequency zero-sequence voltage signal required by the MMC low-frequency operation can be reused. During the startup process of the PMSM, the weighted function of gradually cutting off the high-frequency signal injected by the MMC and the weighted function of switching between speed sensors are reused, and finally a PMSM speed sensorless control system based on MMC drive is realized in the full speed range. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a modular multi-level converter main circuit topology; Figure 2 compose a system control diagram for multiplexing high-frequency signals and weighting functions within the full speed range; Figure 3 is the weighted function control strategy diagram. DETAILED DESCRIPTION

[0036] In order to make the basic principles, technical solutions and advantages of the present invention more clearly understood, a method for speed sensorless control multiplexing based on MMC driving PMSM according to the present invention is described in detail below. It should be understood that the following description is merely exemplary and is not intended to limit the scope of protection and application of the present invention.

[0037] The present invention provides a method for controlling multiplexing of PMSM without speed sensor based on MMC. Figure 1, is a modular multi-level converter main circuit topology, consisting of three-phase six-bridge arms, each bridge arm has N sub-modules, the sub-module can be a full-bridge, half-bridge, etc. structure, but the present invention uses a half-bridge sub-module for analysis, the DC side of the MMC is connected to the DC source or the DC bus of the power grid, the DC side voltage is U dc , the AC side is connected to a permanent magnet synchronous motor, and the AC side can output N+1 level three-phase AC sinusoidal power. Figure 2 A system control structure that reuses high-frequency signals and weighted functions within the full speed operating range is proposed, and the sliding mode observer suitable for medium and high speeds is combined with the pulse high-frequency voltage injection method at zero and low speeds. In this process, the high-frequency signal required for injection into the MMC under zero and low speed conditions is multiplexed with the high-frequency signal required by the pulse high-frequency voltage method without speed sensor, and the weighted function for cutting off the MMC high-frequency signal from zero and low speed to medium and high speeds and the weighted function for switching without speed sensor are multiplexed.

[0038] The main system still adopts the speed and current double closed loop structure, and the given speed n * Predicted speed with feedback After comparison, the PI speed regulator is used to obtain the q-axis given current, where i is used. d * =0 control method, the given current of dq axis is respectively abc The dq axis feedback currents obtained by coordinate transformation are compared and the dq axis voltage u is obtained through the PI current regulator. dq , after Park inverse transformation, the three-phase voltage u is obtained abc The input is fed into a reasonable modulation strategy to control the on-off switching of the inverter, thereby controlling the permanent magnet synchronous motor.

[0039] In the picture is the speed predicted by the pulse high-frequency voltage injection method, is the speed predicted by the sliding mode observer, and the two are multiplied by the weighting factors G1 and G2 to obtain the predicted speed of feedback The predicted speed Integrating gives the predicted angle Then input it into Park transform and inverse transform; for the weighted coefficient G1G2, it can also be applied to the MMC topology, then the injected high frequency signal is u′=G1*i ah +G2*0, G1+G2=0, G1 and G2 change linearly with the speed.

[0040] The following describes how to implement a full-speed range position sensorless control method. The weighted function switching algorithm used here is both a method for switching between the two detection methods and an algorithm for removing the MMC high-frequency signal. The control strategy of the algorithm is as follows: Figure 3As shown, ω1-ω2 is the speed switching range. When ω≤ω1, it is the low-speed operation stage. The pulse high-frequency voltage injection method is used for speed sensorless operation, and the high-frequency signal injection is used for MMC operation. At this time, the weighting coefficient G1=1; when ω1<ω<ω2, it is the switching stage. At this time, the predicted speed The calculation formula The high-frequency signal injected in the MMC topology is u′=G1*i ah +G2*0, G1+G2=0, G1 and G2 vary linearly with speed. ω≥ω2 represents the medium- and high-speed operation phase. The speed sensor uses a sliding mode observer, and the MMC does not require high-frequency signal injection. In this case, G2=1. The specific speed switching range will be optimized through further experimentation. This will enable sensorless operation of the MMC-driven permanent magnet synchronous motor across the full speed range.

[0041] The present invention provides a method for multiplexing speed sensorless control of PMSMs driven by MMC. The advantages are: first, the present invention realizes a speed sensorless control system for permanent magnet synchronous motors driven by MMC topology within the full speed range, i.e., applying the speed sensorless control system from two levels to the MMC topology drive field. Second, the present invention considers injecting high-frequency signals under MMC low-frequency conditions to suppress submodule capacitor voltage fluctuations. This is consistent with the high-frequency signal required for the pulsed high-frequency voltage injection method of speed sensorless at zero low speed. For system cost and simplified system control, this method can be reused. In addition, during the system's transition from zero low speed to medium and high speed, the switching process of MMC high-frequency signal injection and speed sensorless switching is consistent, so the same weighting function can be used for switching, further simplifying the system structure and cost.

[0042] The above is a specific embodiment of the present invention and its advantages, but the scope of protection of the present invention is not limited thereto. Those skilled in the art may make changes and modifications to the above embodiment without departing from the technical spirit and principles described in the present invention, and such changes and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A PMSM speed sensorless control method for MMC variable frequency drive, characterized in that Under the low-frequency operation condition of the MMC, the high-frequency signal of the speed sensorless pulsating high-frequency voltage injection method is reused with the high-frequency signal injected by suppressing the capacitor voltage fluctuation, and the weighted function of the speed sensorless switching in the low-speed transition phase of the permanent magnet synchronous motor is reused, thereby reducing the reuse of control signals and controllers and simplifying the control system. The method includes the following steps: Step 1: Analyze the energy flow of the MMC bridge arm half-bridge sub-module to obtain the fluctuation ΔW of the sub-module fast capacitor voltage ap and DC bus voltage U dc , output current amplitude I m The relationship between and the output angular frequency ω; Step 2: Under low-frequency conditions, high-frequency circulating current and high-frequency zero-sequence voltage injection are used to suppress capacitor voltage fluctuations. Step 3: Set the given speed n * and predicted speed By comparison, the q-axis given current i is obtained through the proportional integral link, i.e., the PI controller q * ; Step 4: The collected AC current i abc The dq axis feedback current i is obtained by 3s / 2r coordinate transformation d and i q , the angle required for 3s / 2r coordinate transformation is the predicted angle The angular velocity is predicted by Calculate and set the dq axis given current i q * and i d * Respectively with i d and i q By comparison, the dq axis voltage u is obtained through the PI current regulator dq ; Step 5: dq axis voltage u dq The three-phase modulation voltage u is obtained by Park inverse transformation abc , the angle required for the park inverse transformation is the predicted angle The angular velocity is predicted by Calculated; Step 6: Convert the three-phase modulated voltage u abc Input a reasonable modulation strategy to control the on and off of devices in the MMC inverter to control the permanent magnet synchronous motor; In step 3, the speed is predicted by the pulse high-frequency voltage injection method. and sliding mode observer to predict the speed Then the predicted speed is calculated by weighted function Among them, the high-frequency signal required by the pulsating high-frequency voltage injection method used in the low-speed stage of the permanent magnet synchronous motor and the high-frequency circulating current and high-frequency zero-sequence voltage signal injected during the low-frequency operation of the MMC can be reused; In step 3, the speed is predicted The calculation formula G1 and G2 are and This weighting factor can not only be used to switch between the pulsating high-frequency voltage injection method and the sliding mode observation method, but also can be used as a switch for the high-frequency circulating current and high-frequency zero-sequence voltage signal injected during the transition phase from zero low speed to medium high speed of the entire system, thereby achieving multiplexing.

2. The method according to claim 1, characterized in that In step 1, by analyzing the energy flow of the bridge arm submodule capacitor, the fluctuation ΔW of the submodule fast capacitor voltage can be obtained. ap and DC bus voltage U dc , output current amplitude I m The relationship between and the output angular frequency ω is calculated as follows: Among them, P ap is the upper arm submodule power, m is the modulation index, is the power factor angle.

3. The method according to claim 2, characterized in that In step 2, the required high-frequency circulating current and high-frequency zero-sequence voltage to be injected are calculated as follows, taking phase a as an example: Among them, i a is the output current of phase a, K cm In order to introduce the coefficient that characterizes the amplitude margin of high-frequency common-mode voltage, it is set between 1.0 and 1.2, ω h is the high-frequency zero-sequence voltage angular frequency.

4. The speed sensorless control method according to claim 1, characterized in that: In step 4, take i * d =0.

5. The speed sensorless control method according to claim 1, characterized in that: In step 4, the angular velocity is predicted The predicted angle obtained by integration Input into the 3s / 2r coordinate transformation.

6. The speed sensorless control method according to claim 1, characterized in that: In step 5, the predicted angular velocity Integrate to get the predicted angle Input to the park inverse transform.

7. The speed sensorless control method according to claim 1, characterized in that: In step 6, according to the three-phase modulation voltage u abc A reasonable modulation strategy is adopted to generate a power device pulse width signal, and then the power device pulse width signal is converted into a three-phase winding current i by a voltage source inverter. abc , sent to the permanent magnet synchronous motor.

8. The speed sensorless control method according to claim 1, characterized in that: Based on the principle of the pulse high-frequency voltage injection method: a sinusoidal high-frequency voltage signal is injected into the d-axis of the estimated synchronous rotating coordinate system to stimulate the motor to produce an inductance saturation effect to show the motor salient polarity; the high-frequency response current component is increased and demodulated accordingly to obtain the rotor position information, that is, the predicted speed. Integrate the predicted speed to get the predicted angle 9. The speed sensorless control method according to claim 1, characterized in that: Method based on sliding mode observer: The state equation of the current error system can be obtained by subtracting the three-phase PMSM synchronous rotating coordinate system from the mathematical model designed by the sliding mode observer: in, and is the current observation error; The sliding mode observer is used to estimate the current and the sliding mode surface is defined as When the system enters the sliding mode, the dq-axis induced electromotive force can be obtained: It can be seen that the q-axis induced electromotive force contains the rotor speed information, that is, The predicted speed is obtained Integrate the predicted speed to get the predicted angle

Images