Sensorless control unit and control method for permanent magnet electric machines

Abstract

Disclosed are a sensorless control unit and a control method for a permanent magnet motor. The sensorless control unit for a permanent magnet motor according to an embodiment of the present invention includes a first estimator that estimates first physical quantity information describing operation of the motor in a stationary coordinate system, a second estimator that estimates second physical quantity information representing a counter electromotive force of the motor in a synchronous coordinate system that is based on a first estimated value of an angle of a rotor of the motor obtained based on the first physical quantity information, and a position / speed detector that generates position and speed information of the rotor of the motor including compensation for an error in the angle of the first estimated value using the second physical quantity information.

Description

Technical Field

[0001] This invention relates to a sensorless control unit and method for a permanent magnet synchronous motor (PMSM). Specifically, the sensorless control involves estimating the information needed to control the motor from the voltage and current supplied to the motor without using physical sensors within the PMSM, and controlling the PMSM based on the estimated information. Background Technology

[0002] An electric motor is a device that converts electrical energy into mechanical energy by utilizing the force exerted on a conductor by an electric current in a magnetic field. Electric motors are classified into DC motors and AC motors based on the type of power source. AC motors are further divided into single-phase AC motors and three-phase AC motors, and each of these can be either induction motors or synchronous motors.

[0003] Since synchronous motors receive magnetic flux from permanent magnets attached to the rotor, vector control requires knowledge of the rotor's position. To obtain this position information, position sensors, such as Hall sensors, resolvers, or encoders, need to be attached to the motor shaft. However, position sensors are expensive and require additional complex hardware. Furthermore, attaching position sensors to the motor shaft increases the motor's size and inertia.

[0004] Furthermore, except in special circumstances, the position detection sensors used in this case have a significant disadvantage in environments where motors are widely used (e.g., environments requiring vibration resistance, shock resistance, corrosion resistance, temperature resistance, and humidity resistance).

[0005] Therefore, research is actively underway on sensorless control methods that can control motors even without position sensors.

[0006] Sensorless control algorithms for permanent magnet synchronous motors (PMSMs) are typically classified into harmonic injection (signal injection) and model-based methods. Harmonic injection exhibits excellent performance in the stop-low speed and low torque ranges, but it is unsuitable for high-speed motor control. Model-based methods, which perform sensorless control based on mathematical models, are advantageous for high-speed control. Conversely, model-based methods suffer from instability due to parameter variations in low-speed operation and rapid acceleration ranges, thus requiring improvements in responsiveness and other aspects.

[0007] As a model-based sensorless control algorithm, sensorless control methods based on back electromotive force (EMF) estimation can be cited. This method uses an electrical or electromechanical model of the motor to estimate the back EMF and obtain speed and rotor position information. Back EMF estimation-based sensorless control methods are divided into methods based on a stationary coordinate system and methods based on a synchronous coordinate system. For example, improved versions of the sensorless control method based on synchronous coordinate system back EMF estimation are described in Korean Patent Registration No. KR 10-1329132, "Sensorless Control Device for Permanent Magnet Motor," and Korean Patent Publication No. KR 10-2020-0046692, "Sensorless Control Method for Motor."

[0008] In another existing document, “Evaluation of Back-EMF Estimators for Sensorless Control of Permanent Magnet Synchronous Motors” (Lee, Kwang-Woon and Ha, Jung-Ik, Journal of Power Electronics, vol.12, no.4, pp.604-614, Jul.2012), the sensorless control methods of each of the back-EMF estimation methods based on the stationary coordinate system and the back-EMF estimation method based on the synchronous coordinate system are introduced and compared.

[0009] The problems and improvements of sensorless control based on flux estimation are disclosed in another existing document, “Active flux concept for motion-sensorless unified AC drives”, I. Boldea, M. Paicu and G. D. Andreescu, IEEE Trans. Power Electron., vol.23, no.5, pp.2612-2618, Sep. 2008. The flux estimation disclosed here is performed in a stationary coordinate system.

[0010] Sensorless control of PMSMs based on back EMF or magnetic flux estimation can be viewed as determining its characteristics according to a coordinate system representing the voltage equation corresponding to the mathematical model of the motor. Since traditional model-based sensorless control of PMSMs relies on one of the three mathematical models mentioned earlier, it is difficult to achieve stable sensorless control performance over a wide operating range, except at low and high speeds.

[0011] In Korean Patent Publication No. KR 10-1961106, entitled "Sensorless Control Method and Apparatus," which is another existing document, a hybrid sensorless control method is proposed as a way to detect whether phase separation has occurred in sensorless control, rather than detecting whether phase separation has occurred in precise sensorless control. This existing document adopts a sensorless control method based on back electromotive force estimation in a synchronous coordinate system, and in order to sense whether phase separation has occurred, a magnetic flux estimation method based on a stationary coordinate system is applied in parallel.

[0012] However, even with reference to the existing literature, the advantages and disadvantages of each traditional model-based sensorless control method for PMSM are clearly distinguished, and it is difficult to achieve stable sensorless control performance over a wide operating range simply by combining them.

[0013] Existing technical documents

[0014] Patent documents

[0015] (Patent Document 1) Korean Patent Registration No. KR 10-1329132 "Sensorless Control Device for Permanent Magnet Motor" (November 7, 2013)

[0016] (Patent Document 2) Korean Patent Publication No. KR 10-2020-0046692, “A Sensorless Control Method for an Electric Motor” (May 7, 2020)

[0017] (Patent Document 3) Korean Patent Registration No. KR 10-1961106 "Sensorless Control Method and Apparatus" (March 18, 2019)

[0018] Non-patent literature

[0019] (Non-Patent Document 1) "Evaluation of Back-EMF Estimators for Sensorless Control of Permanent Magnet Synchronous Motors" (Lee, Kwang-Woon and Ha, Jung-Ik, Journal of Power Electronics, vol.12, no.4, pp.604-614, Jul. 2012) (July 1, 2012)

[0020] (Non-Patent Document 2) "Active flux concept for motion-sensorless unified AC drives," I. Boldea, M.C. Paicu, and G.D. Andreescu, IEEE Trans. Power Electron., vol.23, no.5, pp.2612-2618, Sep. 2008 (September 1, 2008) Summary of the Invention

[0021] Technical problems to be solved

[0022] The present invention relates to a model-based sensorless control unit and method for estimating the rotor position and speed of a PMSM from the voltage and current supplied to the motor without using additional sensors, and in particular to a rotor position / speed estimation apparatus and method that can improve the performance of model-based sensorless control over a wide operating range of the PMSM.

[0023] The object of this invention is to provide a sensorless control unit and method for a PMSM that can stably ensure sensorless control performance over a wide operating range, including both low-speed and high-speed operating regions. The first embodiment presented in this invention is a hybrid approach combining a flux estimation method using the stationary coordinate system voltage equation and a back electromotive force estimation method using the synchronous coordinate system voltage equation. This approach complementaryly addresses the problems of flux estimators using integrators and HPFs (such as the reduced sensorless control performance in low-speed operating regions due to increased errors in sensorless position estimation caused by phase lead) and the problems of methods estimating back electromotive force from the synchronous coordinate system voltage equation (such as the inclusion of position errors in the estimated value due to inverter nonlinearity and PMSM parameter variations). This eliminates the phenomenon where errors accumulate in the PLL integrator, thus degrading sensorless control performance. Therefore, it allows for stable performance of the model-based sensorless control method of the PMSM across a wide operating range.

[0024] The second embodiment presented in this invention is a hybrid method that combines the back EMF estimation method using the voltage equation in the stationary coordinate system with the back EMF estimation method using the voltage equation in the synchronous coordinate system. This method can complementarily solve the shortcomings of both methods.

[0025] The object of this invention is to propose a novel sensorless control method that combines model-based sensorless control methods with mathematical models having different primary operating regions in a complementary manner, rather than simply combining them in parallel, and sequentially combining them to compensate for each other's shortcomings. The object of this invention is to propose several embodiments that combine model-based sensorless control methods with different primary operating regions in a complementary manner to achieve stable sensorless control performance over a wide operating range.

[0026] Problem-solving methods

[0027] This invention provides a configuration derived to achieve the stated objective. A sensorless control unit for a permanent magnet motor according to an embodiment of the invention includes a first estimator that estimates first physical quantity information describing the operation of the motor in a stationary coordinate system; a second estimator that estimates second physical quantity information representing the back electromotive force of the motor in a synchronous coordinate system based on a first estimated value of the angle of the motor rotor obtained based on the first physical quantity information; and a position / velocity detector that uses the second physical quantity information to generate position and velocity information of the motor rotor, including error compensation in the angle of the first estimated value. Here, the synchronous coordinate system of the second estimator can be understood as a synchronous coordinate system based on the angle of the first physical quantity information. Furthermore, the position / velocity detector can be understood as generating position and velocity information of the rotor including error compensation in the angle of the first physical quantity information.

[0028] The first physical quantity information may be the magnetic flux of the motor.

[0029] The first physical quantity information may be the back electromotive force of the motor.

[0030] When the first physical quantity information is the magnetic flux of the motor, the first estimator may include a magnetic flux estimator that estimates the magnetic flux of the motor as the first physical quantity information in the stationary coordinate system, and a magnetic flux angle calculator that generates the first estimated value of the angle of the rotor of the motor based on the first physical quantity information.

[0031] When the first physical quantity information is the back electromotive force of the motor, the first estimator may include a first back electromotive force estimator that estimates the back electromotive force of the motor in the stationary coordinate system as the first physical quantity information, and a back electromotive force angle calculator that generates the first estimated value of the angle of the rotor of the motor based on the first physical quantity information.

[0032] The second estimator may include a second back EMF estimator that estimates the second physical quantity information representing the back EMF of the motor in a synchronous coordinate system based on the first estimated value, an angular velocity calculator that calculates the angular velocity of the motor rotor based on the first physical quantity information and transmits the angular velocity to the second back EMF estimator, and an angle error calculator that calculates the error in the angle included in the first estimated value based on the angular velocity and the second physical quantity information.

[0033] The second estimator may further include an axis converter that transforms the voltage and current values ​​of the motor in the stationary coordinate system to the voltage and current values ​​of the motor in the synchronous coordinate system and transmits them to the second back EMF estimator.

[0034] The axis converter can set an estimated synchronization coordinate system based on the D-axis for the actual D-axis estimation of the motor rotor, and utilize the estimated synchronization coordinate system as the synchronization coordinate system for transforming the voltage and current values ​​of the motor. In this case, the axis converter can set the estimated synchronization coordinate system based on the first estimated value of the rotor angle of the motor obtained from the first physical quantity information.

[0035] According to an embodiment of the present invention, a sensorless control method for an electric motor is a model-based sensorless control method utilizing the voltage and current supplied in a permanent magnet motor (PMSM). The sensorless control method of the present invention includes a first estimation step of estimating first physical quantity information describing the operation of the motor in a stationary coordinate system; a second estimation step of estimating second physical quantity information representing the back electromotive force of the motor in a synchronous coordinate system based on a first estimated value of the rotor angle of the motor obtained based on the first physical quantity information; and a position / velocity detection step of using the second physical quantity information to generate position and velocity information of the motor rotor, including the angle in the first estimated value, which is compensated for.

[0036] The first estimation step may include a magnetic flux estimation step that estimates the magnetic flux of the motor as the first physical quantity information in the stationary coordinate system, and a magnetic flux angle calculation step that generates the first estimated value of the angle of the rotor of the motor based on the first physical quantity information.

[0037] The first estimation step may include a first back electromotive force estimation step, which estimates the back electromotive force of the motor as the first physical quantity information in the stationary coordinate system, and a back electromotive force angle calculation step, which generates the first estimated value of the angle of the rotor of the motor based on the first physical quantity information.

[0038] The second estimation step may include a second back electromotive force estimation step that estimates the second physical quantity information representing the back electromotive force of the motor in a synchronous coordinate system based on the first estimated value, an angular velocity calculation step that calculates the angular velocity of the motor rotor based on the first physical quantity information and transmits the angular velocity to the second back electromotive force estimation step, and an angle error calculation step that calculates the error in the angle included in the first estimated value based on the angular velocity and the second physical quantity information.

[0039] The second estimation step may further include an axis transformation step of transforming the voltage and current values ​​of the motor in the stationary coordinate system to the voltage and current values ​​of the motor in the synchronous coordinate system and transmitting them to the second back EMF estimator.

[0040] The axis transformation step can set an estimated synchronization coordinate system based on the D-axis for estimating the actual D-axis of the motor rotor, and use this estimated synchronization coordinate system as the synchronization coordinate system for transforming the voltage and current values ​​of the motor. In this case, the axis transformation step can set the estimated synchronization coordinate system based on the first estimated value of the rotor angle of the motor obtained from the first physical quantity information.

[0041] Invention Effects

[0042] According to the present invention, a rotor position / speed estimation apparatus and method are available that can improve the performance of model-based sensorless control in a wide operating range, including both the low-speed and high-speed operating regions of the PMSM.

[0043] According to a first embodiment of the present invention, a hybrid method combining a flux estimation method using the voltage equation in a stationary coordinate system and a back electromotive force estimation method using the voltage equation in a synchronous coordinate system can complementaryly solve the problems of flux estimators using integrators and HPFs (the phenomenon that sensorless control performance is reduced in the low-speed operating region due to increased errors in sensorless position estimation caused by phase lead) and the problems of methods for estimating back electromotive force from the voltage equation in a synchronous coordinate system (due to the inclusion of position errors in the estimated value caused by the nonlinearity of the inverter and the parameter variation of the PMSM). This eliminates the phenomenon where errors accumulate in the PLL integrator, thus degrading sensorless control performance. Therefore, it allows for stable performance of the model-based sensorless control method of the PMSM across a wide operating range.

[0044] According to a second embodiment of the present invention, by using a hybrid method that combines the back EMF estimation method using the voltage equation in the stationary coordinate system with the back EMF estimation method using the voltage equation in the synchronous coordinate system, the shortcomings of the back EMF estimation methods using the voltage equation in the stationary coordinate system and the back EMF estimation method using the voltage equation in the synchronous coordinate system can be solved in a complementary manner.

[0045] According to the present invention, a new sensorless control method is achieved by combining model-based sensorless control methods that are complementary to each other based on mathematical models with different main operating regions, rather than simply combining them in parallel, and by combining them sequentially to compensate for each other's shortcomings.

[0046] According to the present invention, the disadvantages of sensorless control methods based on complementary combinations of models can be overcome by leveraging the advantages of each other. For example, the advantages of synchronous coordinate system back EMF estimation methods can be utilized to overcome the disadvantages of stationary coordinate system back EMF estimation methods or stationary coordinate system magnetic flux estimation methods. In this way, stable sensorless control can be performed over a wide operating range, from low speed to high speed.

[0047] The model-based sensorless control technology according to the present invention is not limited to selectively or concurrently combining conventional sensorless control technologies, but rather combines them to achieve a new sensorless control technology. Therefore, the model-based sensorless control technology of the present invention is not selectively applied according to specific conditions, but can apply a single technology across a wide operating range, and can achieve stable sensorless control functionality across a wide operating range. Attached Figure Description

[0048] Figure 1 This is a block diagram illustrating a general PMSM sensorless vector control system within the technical field of the present invention.

[0049] Figure 2 This is a block diagram illustrating the functional elements of a model-based sensorless position / velocity estimator for a general PMSM in the technical field of the present invention.

[0050] Figure 3 It is a spatial vector diagram showing the correlation between the coordinates of the PMSM.

[0051] Figure 4 This is an embodiment of a detailed configuration of a position / velocity estimator utilizing estimated position errors according to an embodiment of the present invention.

[0052] Figure 5 This is a block diagram illustrating a sensorless control unit for a PMSM according to an embodiment of the present invention.

[0053] Figure 6 It is shown in detail Figure 5 A block diagram of a first embodiment of a sensorless control unit.

[0054] Figure 7 It is shown Figure 6 A block diagram of an embodiment of a detailed configuration of a magnetic flux estimator.

[0055] Figure 8 It is shown Figure 6 A block diagram of an embodiment of a detailed configuration of a position / velocity detector.

[0056] Figure 9 It is shown in detail Figure 5A block diagram of a second embodiment of a sensorless control unit.

[0057] Figure 10 The block diagram shown here is a sensorless control unit of a PMSM utilizing a back EMF estimator in a general synchronous coordinate system within the technical field of this invention, as a comparative example of the invention.

[0058] Figure 11 This is a waveform diagram showing the simulation results of sensorless control of a PMSM according to an embodiment of the present invention.

[0059] Explanation of reference numerals in the attached figures

[0060] 110: Rotor Position / Speed ​​Estimator

[0061] 510: First Estimator

[0062] 530: Second Estimator

[0063] 550: Position / Velocity Detector Detailed Implementation

[0064] In addition to the objectives stated herein, other objectives and features of the invention will become apparent from the description of embodiments with reference to the accompanying drawings.

[0065] Preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing the present invention, detailed descriptions of related well-known configurations or functions are omitted where such descriptions might obscure the essence of the invention.

[0066] This invention discloses a sensorless control unit and method for a permanent magnet synchronous motor (PMSM). Generally, sensorless control means estimating the information needed to control the motor from the voltage and current supplied to the motor without using physical sensors within the permanent magnet motor, and controlling the permanent magnet motor based on this estimated information.

[0067] The contents known to those skilled in the art prior to the application of this invention are described in this specification as part of the configuration of this invention, while the description of facts known to those skilled in the art may be omitted if it is believed that it may obscure the spirit of the invention. Furthermore, matters omitted in this specification can be found in existing literature cited in this application specification (e.g., Korean Patent Publication No. KR 10-1329132 "Sensorless Control Device for Permanent Magnet Motor", Korean Patent Publication No. KR 10-2020-0046692 "Sensorless Control Method for Motor", Korean Patent Publication No. KR 10-1961106 "Sensorless Control Method and Device", "Evaluation of Back-EMF Estimators for Sensorless Control of Permanent Magnet Synchronous Motors (Lee, Kwang-Woon and Ha, Jung-Ik, Journal of Power Electronics, Vol. 12, pp. 4, 604-614, July 2012)"). The terms “Electronics, vol. 12, no. 4, pp. 604-614, Jul. 2012.” and “Activeflux concept for motion-sensorless unified AC drives” (I. Boldea, M.C. Paicu and G.D. Andreescu, IEEE Trans. Power Electron., vol. 23, no. 5, pp. 2612-2618, Sep. 2008) are well known to those skilled in the art and can therefore be substituted in the description.

[0068] Some of the content disclosed in these existing documents is related to the problem that this invention aims to solve, and some of the solutions adopted by this invention are also commonly applied to these existing documents.

[0069] Figures 1 to 4 This is a diagram illustrating the constituent elements adopted in one embodiment of the present invention. Figures 1 to 4 At least a portion of it illustrates conventional techniques related to the present invention, and includes portions that are modified as needed for application to the configuration of the present invention, and thus may differ from conventional techniques recognized under patent law.

[0070] In the following Figures 1 to 11In order not to obscure the key points, matters that are considered to be well known in the technical field of the present invention may be omitted from the description as needed, or may be replaced by citing existing literature.

[0071] Furthermore, some or all of the configurations of the existing literature cited above and cited below may be related to the problem to be solved by the present invention, and some of the solutions adopted by the present invention may be borrowed from existing literature.

[0072] Matters commonly included in the prior art for the purpose of elaborating on the present invention can be considered as part of the configuration of the present invention.

[0073] The following is passed Figures 1 to 11 The embodiments described herein illustrate the details of the invention.

[0074] Figure 1 This is a block diagram illustrating a general PMSM sensorless vector control system 100 within the technical field of the present invention.

[0075] The control system 100, as a component for performing sensorless vector control of a permanent magnet synchronous motor (PMSM) 120, includes a rotor position / speed estimator 110, a speed controller 130, a current controller 132, and a PWM mode generator 140. Control commands generated in the PWM mode generator 140 are transmitted to the PMSM 120 via a three-phase inverter 142. A current sensor 150 senses the current in the PMSM 120 and transmits the phase current information to the rotor position / speed estimator 110. The rotor position / speed estimator 110 estimates the rotor position (electric angle) and speed of the PMSM 120 and transmits the estimated position / speed information to a DQ converter 152. The DQ converter 152 transmits the DQ-axis reference current information to the current controller 132 via DQ-axis transformation, and the speed information estimated in the rotor position / speed estimator 110 is transmitted to the speed controller 130. Speed ​​controller 130 generates DQ axis current commands based on the speed error between the commanded speed and the estimated speed and transmits them to current controller 132.

[0076] The PMSM 120 boasts higher energy efficiency and higher power density per unit volume compared to other motors, making it widely used across various industries, from household appliances like refrigerators, air conditioners, and washing machines to electric vehicles. Field-oriented control (FOC) is widely employed as the control method for the PMSM 120. For reference, FOC is also known as vector control. In FOC, the flux component current (hereinafter referred to as D-axis current) and the torque component current (hereinafter referred to as Q-axis current) can be controlled independently. This requires a process of detecting the D-axis and Q-axis currents from the phase current supplied to the PMSM 120 (called DQ conversion). The DQ conversion requires the electrical angle information of the PMSM rotor, which is typically detected using sensors such as Hall effect sensors, optical encoders, and resolvers.

[0077] As mentioned above, the use of this rotor position detection sensor leads to increased costs for the PMSM drive system and decreased reliability due to sensor failure. Therefore, as Figure 1 As shown, the present invention employs a sensorless control method that estimates the electrical angle of the rotor and controls the PMSM 120 by measuring the voltage and current supplied to the PMSM 120.

[0078] As mentioned above, several sensorless control methods for PMSM 120 have been proposed so far, which can be broadly classified into signal injection methods and model-based back electromotive force (or magnetic flux) estimation methods.

[0079] The signal injection method involves applying a high-frequency voltage signal, higher than the drive frequency of the PMSM 120, and using the resulting high-frequency current response to estimate the rotor's electrical angle. This sensorless control method utilizes the stator winding inductance variation of the PMSM 120 based on the rotor position change, and is known to be applicable only to PMSMs with an interior permanent magnet (IPM) structure where the permanent magnet is embedded in the rotor. The signal injection sensorless control method can estimate rotor position information even when the IPM type PMSM is stationary or operating at low speed. However, there is a problem of torque pulsation in the PMSM 120 due to the high-frequency signal injection, resulting in significant vibration and noise. Furthermore, there is a tendency for a significant increase in iron loss in the PMSM 120 due to the high-frequency voltage application. Therefore, the signal injection sensorless control method is known to be used only in the stationary and low-speed operating regions for IPM type PMSMs.

[0080] The model-based back electromotive force (or flux) estimation sensorless control method utilizes the voltage and current information applied to the PMSM 120 to estimate the back electromotive force (or flux) in the stator windings caused by the rotor's permanent magnets, and extracts the rotor electrical angle information included in the estimated back electromotive force (or flux) to control the PMSM 120. Since no additional high-frequency signals are injected in the model-based sensorless control method, torque pulses are relatively few, and it can be applied to both IPM-type and SPM (Surface Permanent Magnet)-type PMSMs. However, in the PMSM 120's stationary or low-speed operating region corresponding to 2 to 5% below rated speed, stable sensorless control is known to be very difficult to achieve due to the influence of electrical noise, such as that caused by the nonlinearity of the inverter used in the PMSM drive and the inverter's PWM. Therefore, the model-based sensorless control method is mainly used in regions other than the stationary and low-speed operating regions. Furthermore, research to improve the performance of the model-based sensorless control method is steadily progressing in academia and industry.

[0081] This invention proposes in Figure 1 The rotor position / speed estimator 110 is a sensorless control unit and method for estimating the position and speed of the rotor through a method further improved for sensorless control.

[0082] Figure 2 This is a block diagram illustrating the functional elements of a model-based sensorless position / velocity estimator for a general PMSM in the technical field of the present invention.

[0083] To date, several model-based sensorless position / velocity estimation methods for PMSM 120 have been suggested. Figure 2 This illustrates the common functional elements found in various model-based sensorless position / velocity estimation methods. Figure 2 In this process, the back electromotive force (or magnetic flux) estimator 220 estimates the back electromotive force (or magnetic flux) using the voltage equation corresponding to the mathematical model 210 of the PMSM and the voltage and current applied to the PMSM 120. The estimated back electromotive force (or magnetic flux) includes rotor position information. Figure 2 In this model, the rotor position / velocity estimator 230 uses position information included in the estimated back electromotive force (or magnetic flux) to estimate the rotor's position and velocity. The model-based sensorless position estimation method can be based on... Figure 2 The detailed characteristics vary depending on how each of the configuration elements (mathematical model 210, back EMF / magnetic flux estimator 220, rotor position / velocity estimator 230) is configured according to its respective function.

[0084] Examples of the configuration elements (mathematical model 210, back EMF / flux estimator 220, rotor position / velocity estimator 230) according to their respective functions can be found in the aforementioned existing literature: “Evaluation of Back-EMF Estimators for Sensorless Control of Permanent Magnet Synchronous Motors” (Lee, Kwang-Woon and Ha, Jung-Ik, Journal of Power Electronics, vol.12, no.4, pp.604-614, Jul. 2012) and “Active flux concept for motion-sensorless unified AC drive”. I. Bolda, M. Paicu and G.D. Andreescu, IEEE Trans. Power Electron., vol.23, no.5, pp.2612-2618, Sep.2008, etc. to identify the main features.

[0085] Figure 3 It is a spatial vector diagram showing the correlation between the coordinates of the PMSM. Figure 3 The correlation between coordinates used to perform vector control for the PMSM 120 of the present invention is shown.

[0086] exist Figure 3 In this system, the αβ axis is a stationary coordinate system, and the DQ axis (dq axis) is a coordinate system based on the actual D-axis (magnetic flux axis) of the rotor. The γδ axis is a coordinate system based on the estimated D-axis, which is used to estimate the actual D-axis by any method. From this specification onward, the γ axis means the estimated D-axis, and the δ axis means an axis orthogonal to the γ axis.

[0087] Therefore, in Figure 3 In the equation, θ represents the angle between the α-axis and the d-axis. r The electrical angle corresponding to the actual rotor, and θ as the angle between the α-axis and the γ-axis. r This means the estimated electrical angle of the rotor.

[0088] The sensorless control method adopted in the embodiments of the present invention will be described below based on a mathematical model. This sensorless control method and mathematical model may include not only facts known prior to this application, but also variations adapted to the configuration for applying the present invention.

[0089] A. Sensorless control method based on back electromotive force estimation using voltage equations in stationary coordinate system

[0090] In the stationary coordinate system, the voltage equations for the PMSM 120 are given as shown in Equations 1 and 2 below.

[0091] [Mathematical Expression 1]

[0092]

[0093] [Mathematical Expression 2]

[0094]

[0095] v α v β These represent the PMSM 120 stator winding voltages in the α and β axes, respectively, and i α i β These represent the PMSM 120 stator winding currents in the α and β axes, respectively. R is the stator winding resistance, p represents the differential operator, and λ PM This refers to the magnetic flux (back electromotive force constant) caused by the permanent magnets of the rotor. Ld and Lq are the D-axis and Q-axis inductances of the PMSM120 stator winding, respectively, and L0 and L1 are defined by mathematical formula 2.

[0096] θ r This refers to the rotor position, which is the electrical angle, and ω. r This refers to the rotor angular velocity.

[0097] When expressed using the following mathematical formula 3, which extends the aforementioned mathematical formula 1 with respect to the back electromotive force (Extended EMF) Eex, it is as shown in the following mathematical formula 4.

[0098] [Mathematical Expression 3]

[0099] E ex =ω r [(L d -L q )i d +λ PM ]-(L d -L q (pi) q )

[0100] [Mathematical Expression 4]

[0101]

[0102] In Equation 3, id and iq represent the stator voltage and stator current of the PMSM 120, expressed on the D-axis and Q-axis, respectively. The right-hand side of Equation 4 corresponds to the back electromotive force (EMF), which can be obtained from the previously mentioned existing literature: "Evaluation of Back-EMF estimators for Sensorless Control Of Permanent Magnet Synchronous Motors" (Lee, Kwang-Woon and Ha, Jung-Ik, Journal of Power Electronics, vol.12, no.4, pp.604-614, Jul. 2012) and "Active flux concept for motion-sensorless unified AC drive". The specific explanation will be omitted here, as it is clearly understood from the following: I. Boldea, M. Paicu and G. D. Andreescu, IEEE Trans. Power Electron., vol.23, no.5, pp.2612-2618, Sep. 2008.

[0103] Back electromotive force E in the α and β axes a E β It is possible to estimate from mathematical formula 4 as shown in mathematical formula 5 below using a sensorless controlled general observer (not shown). And the estimated value of the rotor's electrical angle The back electromotive force can be estimated using mathematical formula 6. Find the answer.

[0104] [Mathematical Expression 5]

[0105]

[0106]

[0107] [Mathematical Expression 6]

[0108]

[0109] The back electromotive force (EMF) estimation method using the stationary coordinate voltage equation can directly calculate the rotor electrical angle through ArcTangent calculations, as shown in Equation 6. The observer used in the back EMF estimation typically has a low-pass filter (LPF). In the LPF, the phase difference between the input and output signals increases with the frequency of the input signal. As shown in Equation 5, in the stationary coordinate system, since the back EMF is an AC signal, there is a phase difference between the back EMF estimated by the observer and the actual back EMF, and this phase difference increases with speed. Therefore, the sensorless control method that uses the stationary coordinate voltage equation to estimate the back EMF and then derives the rotor position from the estimated back EMF has the disadvantage that when the PMSM 120 is operated at a frequency exceeding the bandwidth of the back EMF observer, the phase error occurring in the back EMF estimator causes a significant increase in the sensorless position estimation error.

[0110] Since the relationship Ld=Lq holds true in the case of SPM type PMSM, the first term on the right side of the voltage equation in mathematical formula 4, which is related to velocity, contains ω. r (L d -L q The term becomes 0. However, in the case of IPM type PMSM where the values ​​of Ld and Lq are different from each other, ω cannot be ignored in the voltage equation. r (L d -L q Furthermore, in order to estimate the back electromotive force using the voltage equation in equation 4, the estimated speed needs to be used instead of the actual speed. Therefore, in the case of an IPM-type PMSM, the error that may occur between the actual speed and the estimated speed under transient or low-speed operating conditions can be considered a cause of performance degradation in sensorless control methods based on back electromotive force estimation using the voltage equation in a stationary coordinate system.

[0111] B. Sensorless control method based on back EMF estimation using synchronous coordinate system voltage equations

[0112] Refer again Figure 3 The voltage equation for PMSM 120 in the γδ axis is given by the following mathematical formula 7.

[0113] [Mathematical Expression 7]

[0114]

[0115] v γ v δ For γ δ The stator voltage of the PMSM 120 in the shaft, and i γ i δ For γ δStator current of PMSM 120 in shaft. For the estimated value of the electrical angle of the rotor The estimated angular velocity of the rotor is obtained.

[0116] e is the second term on the right side of mathematical expression 7. γ e δ The back electromotive force is shown with respect to the γδ axis, and specifically, for the extended back electromotive force Eex, it can be shown as shown in the following mathematical formula 8.

[0117] [Mathematical Expression 8]

[0118]

[0119] When estimating the back electromotive force from the synchronous coordinate system represented by Equation 8 of the observer based on Equation 7, the estimated value of the error Δθ between the actual electrical angle and the estimated electrical angle of the rotor is... It can be obtained through the following mathematical expression 9.

[0120] [Mathematical Expression 9]

[0121]

[0122] In mathematical expression 9, They respectively mean e γ e δ The estimated value. The estimated value of the position error is obtained using mathematical formula 9. Afterwards, using such Figure 4 The phase-locked loop (PLL) and other methods are used to estimate the rotor electrical angle θ for the PMSM120. r estimated value and for electric angular velocity ω r estimated value

[0123] Figure 4 This is an embodiment of a detailed configuration of a position / velocity estimator 230 utilizing estimated position errors according to an embodiment of the present invention.

[0124] Position error estimate The value is transmitted to the proportional-integral controller (PI controller) 410, and the estimated rotor electrical angular velocity is obtained by the proportional-integral controller 410. The data is transmitted to integrator 420, thereby deriving the estimated rotor electrical angle.

[0125] Refer again Figure 3 As shown in Equation 8, the back electromotive force is estimated from the voltage equation in the synchronous coordinate system. It contains position error information Δθ and estimates the back electromotive force under normal conditions. This can be considered a DC signal. Therefore, the phase delay effect of the observer used in the back EMF estimation can be almost ignored. Thus, compared to sensorless control based on back EMF estimation using the voltage equation in the stationary coordinate system, sensorless control based on back EMF estimation using the voltage equation in the synchronous coordinate system has the advantage of superior sensorless control performance at high speeds. However, due to inverter nonlinearity, PMSM 120 parameter variations, etc., including in the position error estimation value... Error components can accumulate in the integrator of the PLL used for position / velocity estimation, and this phenomenon is particularly detrimental to sensorless control performance in low-speed operating regions. Therefore, the bandwidth of the PLL used for position / velocity estimation should be set low in low-speed operating regions, resulting in unsatisfactory sensorless control performance in applications with high loads in low-speed operating regions.

[0126] The voltage equation in mathematical formula 7 contains the error corresponding to the difference between the actual speed and the estimated speed. Item. Since the relationship Ld = Lq does not hold in the case of SPM type PMSM, therefore, mathematical expression 7... The terms will actually be canceled out. However, in the case of IPM type PMSP, because... This will affect the voltage equation, so sensorless control performance may be reduced in transitional states where there is an error between the actual speed and the estimated speed, or under low-speed operating conditions.

[0127] C. Sensorless control method based on magnetic flux estimation using voltage equations in stationary coordinate system

[0128] In a stationary coordinate system, when the magnetic flux along the α-axis is λ α And the magnetic flux in the β axis is λ β At that time, the voltage equation for PMSM 120 is given as shown in the following mathematical formula 10.

[0129] [Mathematical Expression 10]

[0130]

[0131] At this time, the magnetic flux λ in the α axis α and magnetic flux λ in the β axis β With magnetic flux λ in the DQ axis d , λ q The following mathematical expression holds true between them.

[0132] [Mathematical Expression 11]

[0133]

[0134] Magnetic flux λ in the DQ axis d , λ q It is represented as shown in mathematical formula 12 below.

[0135] [Mathematical Expression 12]

[0136] λ d =L d i d +λ PM , λ q =L q i q

[0137] At this point, the active magnetic flux... It can be defined as shown in the following mathematical formula 13.

[0138] [Mathematical Expression 13]

[0139]

[0140] When the effective magnetic flux is used as defined in Equation 13, the magnetic flux λ in the α-axis α and magnetic flux λ in the β axis β It can be represented as shown in mathematical formula 14 below.

[0141] [Mathematical Expression 14]

[0142]

[0143] When using mathematical expression 14, mathematical expression 10 can be expressed as shown in mathematical expression 15 below.

[0144] [Mathematical Expression 15]

[0145]

[0146] From mathematical equation 15, the effective magnetic flux in the α-axis and effective magnetic flux in the β axis The results can be obtained as shown in mathematical expressions 16 and 17 below.

[0147] [Mathematical Expression 16]

[0148]

[0149] [Mathematical Expression 17]

[0150]

[0151] Utilizing the effective magnetic flux in the α axis and effective magnetic flux in the β axis Rotor electrical angle θ r The result is shown in the mathematical formula 18 below.

[0152] [Mathematical Expression 18]

[0153]

[0154] As shown in mathematical formula 17, in order to determine the effective magnetic flux in the α-axis and effective magnetic flux in the β axis Integration is required. When small error components included in the voltage and current accumulate in the integrator, errors will occur in the flux calculated using Equations 16 and 17. To address this issue, a high-pass filter (HPF) is typically used at the integrator output of Equations 16 and 17 to shield the DC component from the integrator output. Since a series connection of the integrator and HPF provides the same characteristics as an LPF, an LPF is used in practical implementations instead of both the integrator and HPF.

[0155] When an HPF is used to mask the DC component of the integrator in flux estimation, a large phase lead occurs between the input and output signals of the HPF below its shielding frequency. That is, when combining the integrator and HPF to estimate flux, an estimation error due to phase lead occurs between the actual and estimated flux. This error is significant when the operating frequency of the PMSM 120 is below the shielding frequency of the HPF, and almost no error due to phase lead occurs when the operating frequency of the PMSM 120 is sufficiently higher than the shielding frequency of the HPF.

[0156] Since the sensorless control method based on flux estimation using the voltage equation in the stationary coordinate system does not use the speed information from the PMSM 120, the effect caused by the error between the actual speed and the estimated speed cannot be ignored in the transient state. However, as mentioned above, in the low-speed operating region, there is a drawback that the sensorless control performance is reduced due to phase lead in the flux estimator.

[0157] To address the shortcomings of existing flux estimation methods, an existing paper (I. Boldea, M. Paicu, and G.D. Andreescu, “Active flux concept for motion-sensorless unified AC drives”, IEEE Trans. Power Electron., vol.23, no.5, pp.2612-2618, Sep. 2008) suggests calculating flux based on estimated rotor electrical angles using PMSM 120 parameters. This method utilizes the error between the calculated flux and the flux obtained from the integrator to prevent DC drift in the integrator used for flux estimation. Compared to simple integration-based flux estimation, this method exhibits superior performance at low speeds. However, it cannot accurately determine the actual parameters (Ld, Lq, and λ) of the PMSM 120 that vary with operating conditions. PM The value (etc.) and the need to add an additional controller to prevent integrator drift, which complicates its configuration.

[0158] Sensorless control of PMSMs based on back EMF or magnetic flux estimation can be viewed as determining its characteristics according to a coordinate system representing the voltage equation corresponding to the mathematical model of the motor. Since traditional model-based sensorless control of PMSMs is based on one of the three mathematical models mentioned above, it is difficult to achieve stable sensorless control performance over a wide operating range, up to the low-speed and high-speed operating regions.

[0159] For example, the back EMF estimation method based on the stationary coordinate system has the advantage of directly obtaining rotor position information from the estimated back EMF. However, due to the phase delay characteristics of the back EMF estimator, the position estimation error may increase with the operating speed of the PMSM 120. Furthermore, since the estimated speed is used for back EMF estimation in the case of an IPM-type PMSM, sensorless control performance can be controlled in the low-speed region. Therefore, there is a disadvantage that the bandwidth of the back EMF estimator and PLL may be limited in the low-speed region.

[0160] Furthermore, in the back EMF estimation method based on the synchronous coordinate system, since the estimated back EMF is a DC signal, there is an advantage of very little performance degradation due to phase delay in the back EMF estimator. However, noise signals included in the estimation error accumulate in the PLL integrator, which may reduce sensorless control performance in low-speed regions. Therefore, there is a disadvantage of limited bandwidth for both the back EMF estimator and the PLL in low-speed regions.

[0161] Furthermore, in the magnetic flux estimation method based on the stationary coordinate system, 1) since velocity information is not used in the magnetic flux estimation, there is no performance degradation due to errors in velocity estimation; and 2) since the magnetic flux is estimated through a combination of an integrator and a high-power field (HPF), there is no phase delay in the high-speed region exceeding the bandwidth of the HPF. Conversely, this method suffers from the disadvantage of significant position estimation errors due to phase lead in the low-speed region.

[0162] According to embodiments of the present invention, since the magnetic flux estimation (or back EMF estimation) based on the voltage equation in the stationary coordinate system and the back EMF estimation method based on the voltage equation in the synchronous coordinate system are combined into a hybrid configuration, their respective advantages can be flexibly utilized while their respective disadvantages can be complemented. As a result, according to embodiments of the present invention, there is no need to selectively apply sensorless control methods based on the operating region, and stable sensorless control performance can be achieved even when using a single sensorless control method spanning a wide operating region.

[0163] Figure 5 This is a block diagram illustrating a sensorless control unit for a PMSM according to an embodiment of the present invention.

[0164] The present invention relates particularly to the sensorless control unit of the PMSM. Figure 1 Rotor position / speed estimator 110.

[0165] The sensorless control unit for a PMSM of the present invention includes a first estimator 510 that estimates first physical quantity information describing the operation of a PMSM motor 120 in a stationary coordinate system; a second estimator 530 that estimates second physical quantity information representing the back electromotive force of the PMSM motor 120 in a synchronous coordinate system based on a first estimated value of the angle of the rotor of the PMSM motor 120 obtained based on the first physical quantity information; and a position / velocity detector 550 that uses the second physical quantity information to generate position and velocity information of the rotor of the PMSM motor 120, including errors in the angle of the first estimated value. Here, the synchronous coordinate system of the second estimator can be understood as a synchronous coordinate system based on the angle of the first physical quantity information. Furthermore, the position / velocity detector can be understood as generating position and velocity information of the rotor, including errors in the angle of the first physical quantity information.

[0166] The first physical quantity information can be the magnetic flux of the PMSM motor 120.

[0167] The first physical quantity information can be the back electromotive force of the PMSM motor 120.

[0168] Figure 6 It is shown in detail Figure 5 A block diagram of a first embodiment of a sensorless control unit.

[0169] Figure 6 The first embodiment is an embodiment in which the first physical quantity information is assumed to be the magnetic flux of the PMSM motor 120. The first estimator 510 may include a magnetic flux estimator 612 that estimates the magnetic flux of the PMSM motor 120 as the first physical quantity information in a stationary coordinate system, and a magnetic flux angle calculator 614 that generates a first estimated value of the angle of the rotor of the PMSM motor 120 based on the first physical quantity information.

[0170] The second estimator 530 may include a second back EMF estimator 632 that estimates second physical quantity information representing the back EMF of the PMSM motor 120 in a synchronous coordinate system based on the first estimated value, a (magnetic flux) angular velocity calculator 636 that calculates the angular velocity of the rotor of the PMSM motor 120 based on the first physical quantity information and transmits the angular velocity to the second back EMF estimator 632, and an angle error calculator 634 that calculates the error in the angle included in the first estimated value based on the angular velocity and the second physical quantity information.

[0171] The second estimator 530 may further include an axis converter 638 that transforms the voltage and current values ​​of the PMSM motor 120 in the stationary coordinate system into the voltage and current values ​​of the PMSM motor 120 in the synchronous coordinate system and transmits them to the second back EMF estimator 632.

[0172] The axis converter 638 can set an estimated synchronization coordinate system (γδ) based on the estimated D-axis (γ-axis) of the actual D-axis of the PMSM motor 120 rotor, and utilize the estimated synchronization coordinate system as a synchronization coordinate system for transforming the voltage and current values ​​of the PMSM motor 120. In this case, the axis converter can set the estimated synchronization coordinate system based on a first estimated value of the angle of the rotor of the PMSM motor 120 obtained from the first physical quantity information.

[0173] As shown in equations 15 to 17, the magnetic flux estimator 612 can estimate the effective magnetic flux in the α-axis using the voltage equation in the stationary coordinate system. and effective magnetic flux in the β axis The input to the magnetic flux estimator 612 is the voltage (v) in the αβ coordinate system. α v β ) and current (i α i β The output is the effective magnetic flux estimated based on the αβ coordinate system. Voltage (v) in the αβ coordinate system α v β The phase voltage command can be obtained from the three-phase inverter 142. Use mathematical expressions 19 and 20 below to find the answer.

[0174] [Mathematical Expression 19]

[0175]

[0176] [Mathematical Expression 20]

[0177]

[0178] Current (i) in the αβ coordinate system α i β The phase current (i) of the three-phase inverter 142 can be obtained. as i bs i cs Use mathematical expressions 21 and 22 below to find the answer.

[0179] [Mathematical Expression 21]

[0180]

[0181] [Mathematical Expression 22]

[0182]

[0183] Figure 7 It is shown Figure 6 A block diagram of an embodiment of a detailed configuration of the magnetic flux estimator 612.

[0184] Reference Figure 7 The integrator used to estimate the effective magnetic flux is implemented as an integrator and HPF LPF corresponding to the product Therefore, the operating frequency of PMSM 120 is the shielding frequency of LPF (=ω). c In the following low-speed operating range, Figure 6 The phase of the effective magnetic flux output by the magnetic flux estimator 612 is ahead of the actual value.

[0185] Refer again Figure 6 As shown in mathematical formula 18, the magnetic flux angle calculator 614 can perform an arctangent calculation on the output signal of the magnetic flux estimator 612 to calculate the magnetic flux angle (θ) in the stationary coordinate system αβ. flux ).

[0186] The magnetic flux angular velocity calculator 636 can calculate the magnetic flux angle (θ) flux angular velocity (ω) flux The magnetic flux angular velocity calculator 636 can calculate the magnetic flux angle (θ). flux The angular velocity (ω) can be obtained by taking the derivative of ω. flux Furthermore, when the noise caused by the derivative is significant, an additional PLL can be used to calculate the angular velocity (ω). flux The method of calculating angular velocity using a PLL based on angle information can be implemented using common knowledge in the field of motor drives, therefore a detailed explanation will be omitted.

[0187] The axis converter 638 can convert the magnetic flux angle (θ) flux The following mathematical expressions 23 and 24 are applied to perform the estimation of the synchronous coordinate system (γδ) reference voltage (v) based on magnetic flux. γf v δf ) and current (i γf i δf ) axis transformation.

[0188] [Mathematical Expression 23]

[0189]

[0190] [Mathematical Expression 24]

[0191]

[0192] The second back electromotive force estimator 632 uses the magnetic flux angle (θ) flux The estimation of the synchronous coordinate system (γδ) based on the voltage equation and the observer is used to estimate the rotor electrical angle (θ). r ) and magnetic flux angle (θ)flux Error information (Δθ) between f The back electromotive force of ).

[0193] At this point, as an observer used in back EMF estimation, various observers for estimating back EMF from voltage equations can be utilized, as disclosed in existing literature (Lee, Kwang-Woon and Ha, Jung-Ik, "Evaluation of Back-EMF Estimators for Sensorless Control of Permanent Magnet Synchronous Motors", Journal of Power Electronics, Vol. 12, No. 4, pp. 604-614, Jul. 2012). Such observers are considered generalized and commonly used knowledge in the field related to sensorless control, therefore, detailed descriptions thereof will be omitted in this invention.

[0194] The voltage equation that holds true in the synchronous coordinate system (γδ) can be expressed as shown in the following mathematical formula 25.

[0195] [Mathematical Expression 25]

[0196]

[0197] Estimate the back electromotive force component (e) of the synchronous coordinate system (γδ). γf e δf The extended back electromotive force Eex has the relationship shown in the following mathematical formula 26.

[0198] [Mathematical Expression 26]

[0199]

[0200] When the SPM type PMSM is represented by mathematical expression 25 for the relationship Ld = Lq, it can be simplified to mathematical expression 27.

[0201] [Mathematical Expression 27]

[0202]

[0203] Although in IPM type PMSM (L d -L q The component part is 0, but the last term on the right side of equation 25 includes the angular velocity (ω) of the motor rotor. r ) and from the first estimate (θ)flux The angular velocity component (ω) obtained flux The error (ω) between ) flux -ω r ) and (L d -L q The value of multiplying by θ. At this point, since the difference between Ld and Lq in the IPM type PMSM is small, the last term on the right-hand side of equation 25 is small enough to be ignored in the first estimated value (θ). flux The extent to which the estimation of the back electromotive force is affected in the synchronous coordinate system (γδ) is estimated based on γδ. Therefore, it is known that not only is (ωδ) eliminated as in Equation 27, but also... flux -ω r The influence of the term is negligible in SPM-type PMSMs and also negligible in IPM-type PMSMs (ω). flux -ω r The extent of the impact.

[0204] Angle error calculator 634 can perform arctangent calculation of the output signal of second back EMF estimator 632 as in mathematical formula 28 to obtain θ. r and the first estimate (θ) flux Error information Δθ between ) f =θ r -θ flux .

[0205] [Mathematical Expression 28]

[0206]

[0207] When the error information Δθ obtained through the mathematical formula 28 f =θ r -θ flux Compared with the first estimated value (θ) flux When these are added together, the desired position estimate θ can be obtained, as shown in equation 29. r .

[0208] [Mathematical Expression 29]

[0209] θ flux +Δθ f =θ r

[0210] As mentioned above, the first estimated value (θ) is obtained by combining the integrator and the HPF. flux This is the most common situation, therefore, when the operating frequency of the PMSM 120 is below the shielding frequency of the HPF, errors due to phase lead will occur. Information regarding this error due to phase lead includes information from the first estimated value (θ). fluxThe back electromotive force is estimated using the voltage equation in the synchronous coordinate system (γδ) as a reference. Therefore, when the rotor position θ is determined as in Equation 29... r In this way, it can compensate for the sensorless position estimation error caused by phase lead.

[0211] The position / velocity detector 550 can be applied to θ using PLL technology. flux +Δθ r To output estimated values ​​of rotor position and speed required for sensorless vector control of the PMSM 120.

[0212] Figure 8 It is shown Figure 6 A block diagram of one embodiment of a detailed configuration of the position / velocity detector 550.

[0213] The position / velocity detector 550 includes a receiver for θ. flux +Δθ f Furthermore, the estimated angular velocity value is obtained through proportional-integral (PI) control. The proportional-integral controller 810. Additionally, the position / velocity detector 550 may include values ​​estimated from angular velocity. Obtain location estimate The integrator 820. Figure 8 The detailed operating principle of the position / speed detector 550 is considered common knowledge in the field of motor control, therefore, a detailed description thereof will be omitted in this application specification.

[0214] Figure 9 It is shown in detail Figure 5 A block diagram of a second embodiment of a sensorless control unit.

[0215] Figure 9 The second embodiment is an embodiment in which the first physical quantity information is assumed to be the back electromotive force of the PMSM motor 120. Because... Figure 9 Position / velocity detector 550 and Figure 6 The operation of the position / velocity detector 550 is not significantly different, so repeated descriptions will be omitted.

[0216] When the first physical quantity information is the back electromotive force of the PMSM motor 120, the first estimator 510 may include a first back electromotive force estimator 912 that estimates the back electromotive force of the PMSM motor 120 as the first physical quantity information in a stationary coordinate system, and a first estimated value θ for the rotor angle of the PMSM motor 120 based on the first physical quantity information. EMF The back EMF angle calculator 914.

[0217] The first back electromotive force estimator 912 can obtain the back electromotive force estimate for the stationary coordinate system αβ from the observer using equation 4 and equation 5. As mentioned above, the estimated back electromotive force for the stationary coordinate system αβ is... This includes rotor position information. However, it is common for the observer to have LPF characteristics, therefore, as the operating frequency of the PMSM 120 increases, the back EMF estimate will also increase. The increase in error is due to the phase delay.

[0218] exist Figure 9 In the second embodiment, the first estimated value θ of the rotor position obtained from the first estimator 510 can be used. EMF The back electromotive force (EMF) estimated using the voltage equation in the reference synchronous coordinate system (γδ) is used to compensate for the back EMF estimation value. Errors caused by phase delay.

[0219] The first estimate θ based on the back electromotive force EMF The angular velocity calculator 936 can calculate the first estimate θ. EMF angular velocity ω EMF The axis converter 938 uses a first estimated value θ. EMF The voltage and current are transformed using the established estimated synchronous coordinate system (γδ) as a reference. This transformation process can be represented by the following mathematical formulas 30 and 31.

[0220] [Mathematical Expression 30]

[0221]

[0222] [Mathematical Expression 31]

[0223]

[0224] The second back EMF estimator 932, included within the second estimator 530, uses the first estimated value θ EMF Based on the estimated synchronous coordinate system (γδ), the voltage equation and the observer are used to decipher the rotor position θ. r Compared with the first estimated value θ EMF Error information Δθ E back electromotive force e γE ,eδ E .

[0225] Using the first estimated value θ EMF In the estimated synchronous coordinate system (γδ) based on the reference, the voltage equation can be expressed as shown in the following mathematical formula 32.

[0226] [Mathematical Expression 32]

[0227]

[0228] back electromotive force e γE e δE With error information Δθ E The same relationship as the mathematical expression 33 below.

[0229] [Mathematical Expression 33]

[0230]

[0231] The angle error calculator 934 can execute the output signal e of the second back electromotive force estimator 932. γE e δE The rotor position θ is determined by arctangent calculation. r Compared with the first estimated value θ EMF Error information Δθ E .

[0232] At this time, the error information Δθ F It is represented as shown in mathematical formula 34.

[0233] [Mathematical Expression 34]

[0234] Δθ E =θ r -θ EMF

[0235] exist Figure 9 In the embodiment, the error information Δθ E Comparable to the first estimated value θ EMF The values ​​are summed and input into position / speed detector 550, which can then use a PLL to output estimated values ​​of rotor position and speed required for sensorless vector control of the PMSM 120 from the input information.

[0236] As described above, in conventional sensorless control based on a stationary coordinate system that only utilizes the first estimator 510 of the present invention, for example, in a sensorless method based on back EMF estimation using the voltage equation of the stationary coordinate system, due to the characteristics of the back EMF observer, there is a disadvantage that the phase delay increases with the increase of speed, thereby increasing the position estimation error.

[0237] Furthermore, as another example of conventional technology that utilizes only the first estimator 510 of the present invention, the sensorless method based on magnetic flux estimation using the voltage equation of the stationary coordinate system suffers from the disadvantage that the magnetic flux estimator, which is composed of a combination of an integrator and an HPF, has a phase lead error at speeds lower than the bandwidth of the HPF.

[0238] Figure 10 The block diagram shown here is a sensorless control unit of a PMSM utilizing a back EMF estimator in a general synchronous coordinate system within the technical field of this invention, as a comparative example of the invention.

[0239] Figure 10 The comparative examples can be understood as utilizing only the present invention. Figures 5 to 9 The sensorless control unit of the second estimator 530 and position / velocity detector 550 is configured.

[0240] like Figure 10 The comparative example shown, namely the sensorless method based on back EMF estimation using the synchronous coordinate system voltage equation, suffers from a drawback in the low-speed region where the bandwidth of the PLL used for position / velocity estimation is limited, resulting in reduced low-speed operation performance.

[0241] exist Figure 10 In the comparative example, the axis converter 1038 converts the rotor winding voltage and current values ​​(V) in the stationary coordinate system. α v β i α ′i β The transformation is used to estimate the voltage and current values ​​(v) for the DQ axis. γ v δ i γ i δ Furthermore, the back electromotive force estimator 1032 estimates the back electromotive force based on the estimated DQ axis. The estimation process is performed by the aforementioned mathematical formulas 7 to 9.

[0242] Angle error calculator 1034 performs arctan calculation on the back EMF to generate the angle error, and then transmits the angle error (position error) to axis transformer 1038 via PLL processor 1050. At this point, back EMF estimator 1032 uses the estimated DQ axis (γδ axis), not the actual DQ axis, and the reduction of the error between the estimated DQ axis (γδ axis) reflected by the position error in PLL processor 1050 and the actual DQ axis is precisely... Figure 10 The target in the comparative example. In Figure 10 In the comparative example, the back electromotive force estimated based on the characteristics of the mathematical model indicated in Equations 7 to 9 can be considered to already reflect the position error information. Therefore, the effects of phase delay and other issues exhibited in conventional techniques based on a stationary coordinate system can be reduced. However, as explained above, in Figure 10 The PLL processor 1050 uses, for example Figure 4 With the integrator 420 configured, the position error may continue to accumulate even when the position error is already reflected.

[0243] Referring again to the present invention Figures 5 to 9 The configurations differ in that Figure 10 In the comparative example shown, position estimation information is fed back from PLL processor 1050 to shaft converter 1038, while in the present invention, the estimated rotor position (i.e., the first estimated value) is transmitted from first estimator 510 to shaft converters 638 and 938 of second estimator 530.

[0244] At this point, the present invention proposes a method in which the sensorless position estimation error caused by the phase delay (or phase lead) represented when estimating the back electromotive force (or magnetic flux) based on the voltage equation in the stationary coordinate system in the first estimator 510 can be compensated at once from the voltage equation in the synchronous coordinate system based on the estimated angle by the second estimator 530 using the physical laws included in the estimated back electromotive force.

[0245] exist Figure 10 In the comparative example, the PLL processor 1050 is typically like... Figure 4 As illustrated, the integrator 420 is used in this way, therefore, even when the position error has been reflected, there is a problem that the position error may continue to accumulate, as stated above. In the present invention... Figures 5 to 9 In this embodiment, an additional means of estimating position information is incorporated into a first estimator 510 to prevent the continuous accumulation of position errors. The first estimator 510 transmits a first estimated value for the position information to a second estimator 530, thereby shielding the position error within the mathematical model used by an additional third estimator 520, which provides a reference value for calculating the position error, to prevent its continuous accumulation. Furthermore, the first estimator 510 can generate a phase delay (back EMF) at high speeds or a phase lead (magnetic flux) at low speeds / low frequencies, depending on whether the first physical quantity information to be estimated is back EMF or magnetic flux. The first estimated value of the first estimator 510 is used to generate a second estimated value through the second estimator 530 simultaneously with the mathematical model of the second estimator 530, thereby resolving the problems of phase delay or phase lead.

[0246] This invention differs significantly from conventional techniques that operate multiple sensorless control methods using voltage equations in different coordinate systems while selectively utilizing the sensorless control method advantageous to the operating speed. For example, a simple combination of conventional techniques might involve using a back-EMF estimation method based on the stationary coordinate system voltage equation in the low-speed region and a magnetic flux estimation method based on the stationary coordinate system voltage equation in the high-speed region. While this combination of conventional techniques offers advantages, it cannot compensate for their respective disadvantages. In this combination, the stationary coordinate system back-EMF estimation method still has problems in the high-speed region, and the stationary coordinate system magnetic flux estimation method still has problems in the low-speed region.

[0247] As an example of simply combining and applying traditional sensorless control methods, there are also cases like the one described in Korean Patent Publication No. KR 10-1961106, "Sensorless Control Method and Apparatus," where one sensorless control method is the primary method and another is used as an auxiliary method. In KR10-1961106, the method of calculating the rotor position using the last-position algorithm in a synchronous rotating coordinate system is used as the primary sensorless control method, and a magnetic flux estimation method based on a stationary coordinate system is applied in parallel as an auxiliary means to detect phase separation during release. Since the parallel combination method in KR10-1961106 also compares the calculated rotor angle / position after independently operating the synchronous rotating coordinate system back EMF estimation method and the stationary coordinate system magnetic flux estimation method, the advantages and disadvantages of each are retained, and the disadvantages of each sensorless control technology are not solved by the parallel combination as in KR10-1961106.

[0248] Conversely, according to embodiments of the present invention, since the flux estimation (or back EMF estimation) based on the voltage equation in the stationary coordinate system and the back EMF estimation method based on the voltage equation in the synchronous coordinate system are combined into a hybrid configuration, their respective advantages are flexibly utilized while their respective disadvantages are complemented. As a result, according to embodiments of the present invention, there is no need to selectively apply sensorless control methods based on the operating region, and stable sensorless control performance can be achieved even when using a single sensorless control method across a wide operating region.

[0249] In an embodiment of the invention, the synchronous coordinate system set during back electromotive force estimation is based on a first estimated value of the rotor angle of the motor obtained based on the magnetic flux or back electromotive force estimated in the stationary coordinate system. The back electromotive force of the motor is estimated in this synchronous coordinate system, and the estimated back electromotive force is used to generate rotor position and speed information that compensates for errors in the angle of the first estimated value. While this embodiment of the invention utilizes the concept and configuration of conventional sensorless control, these configurations can be complementary and organically combined, and unlike conventional sensorless control, stable sensorless control performance can be achieved over a very wide operating range.

[0250] In addition, for Figure 10 Comparative examples and Figures 5 to 9 As can be seen from the comparison of the embodiments, the embodiments of the present invention organically combine various sensorless control techniques of the traditional technology. However, in this process, the sensorless control techniques of the traditional technology are not applied as is, but are partially modified to meet the purpose of the present invention. Therefore, the configuration of the present invention is different from that of the traditional technology.

[0251] Figure 11 This is a waveform diagram showing the simulation results of sensorless control of a PMSM according to an embodiment of the present invention.

[0252] Reference Figure 11 ,exist Figure 5 In the first embodiment, when implementing the invention based on magnetic flux estimation, the bandwidth of the PLL used for position / velocity estimation can be set slightly larger in the low-speed region than in other methods, thereby ensuring better sensorless control performance in applications with large load variations at low speeds (e.g., washing modes of drum washing machines using DD motors). Figure 11 The waveform diagram, as shown in the first embodiment of the invention, illustrates the sensorless control method using the washing load of a drum washing machine as a model for computer simulation. (Refer to...) Figure 11 It can be seen that the rotor position estimated using only the existing flux estimator in a sensorless control method takes 0.2 seconds and deviates significantly from the actual rotor position. Conversely, although the method described in this invention has a range where the position estimation error increases with repeated transition intervals, it can be seen that the sensorless position estimation error remains small overall.

[0253] The above has been approved. Figures 1 to 11A sensorless control unit for a motor according to an embodiment of the present invention is disclosed. Furthermore, the sensorless control technique of the present invention can also implement individual functional elements such as processors, controllers, and / or distributed design logic as hardware; however, functional elements performing the same function can be loaded into memory in the form of program instructions, and can also call and execute processors, controllers, and / or distributed design logic. This embodiment is yet another embodiment of the present invention as a model-based sensorless control method for permanent magnet motors (PMSMs).

[0254] For example, Figure 5 The operation of the first estimator 510 and its subordinate functional elements shown can be implemented as program instructions and executed as a first estimation step performed by a processor, controller, and / or logic of a distributed design.

[0255] also, Figure 5 The operation of the second estimator 530 and its subordinate functional elements shown can be implemented as program instructions and executed as a second estimation step performed by a processor, controller, and / or logic of a distributed design.

[0256] also Figure 5 The operation of the position / velocity detector 550 and its subordinate functional elements shown can be implemented as program instructions and executed as position / velocity detection steps performed by a processor, controller, and / or logic of a distributed design.

[0257] Figure 4 , Figures 6 to 9 The operation of the lower-level functional elements of the first estimator 510, the second estimator 530 and the position / velocity detector 550 shown can be implemented as program instructions and executed in the logic of a processor, controller and / or distributed design, etc., and each lower-level functional element can constitute the detailed steps of the first estimation step, the second estimation step or the position / velocity detection step according to their combination relationship.

[0258] A sensorless control method for an electric motor according to an embodiment of the present invention can be implemented as program commands executable by various computer devices and recorded in a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., individually or in combination. The program commands recorded in the medium are specifically designed and configured for the present invention, or may be known and used by those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs and DVDs; magneto-optical media such as floppy disks; and hardware devices specifically designed for storing and executing program instructions, such as read-only memory (ROM), random access memory (RAM), and flash memory. Examples of program commands include machine language code generated by a compiler and high-level language code executable by a computer using an interpreter, etc. The hardware device may be configured to run as one or more software modules to perform the operations of the present invention, and vice versa.

[0259] However, the present invention is not limited to or restricted by the embodiments. The same reference numerals shown in the various figures denote the same parts. For ease of understanding, lengths, heights, sizes, widths, etc., described in the embodiments and figures of the present invention may be exaggerated.

[0260] In the present invention described above, although specific details and limited embodiments and figures such as specific constituent elements have been presented, this is merely provided to facilitate a more comprehensive understanding of the invention. The invention is not limited to the described embodiments, and those skilled in the art can make various modifications and variations from this description. Therefore, the spirit of the invention should not be limited to the described embodiments, and the appended claims and their equivalent variations should be considered within the scope of the spirit of the invention.

Claims

1. A sensorless control unit for a permanent magnet motor, comprising a model-based sensorless control unit utilizing the voltage and current supplied in the permanent magnet motor, and including: A first estimator estimates first physical quantity information describing the operation of the motor in a stationary coordinate system, and generates a first estimated value for the angle of the motor rotor based on the first physical quantity information. A second estimator estimates a second physical quantity representing the back electromotive force of the motor in an estimation synchronization coordinate system set based on the first estimated value. An angular velocity calculator, which calculates the angular velocity of the motor rotor based on the physical quantity information and transmits the angular velocity to the second estimator; as well as An angle error calculator that calculates the error in the angle included in the first estimate based on the angular velocity and the second physical quantity information.

2. The sensorless control unit of the permanent magnet motor as described in claim 1, wherein the first physical quantity information is the magnetic flux of the motor.

3. The sensorless control unit of the permanent magnet motor as described in claim 1, wherein the first physical quantity information is the back electromotive force of the motor.

4. The sensorless control unit for the permanent magnet motor as described in claim 2, wherein the first estimator comprises: A magnetic flux estimator, wherein the magnetic flux estimator estimates the magnetic flux of the motor in the stationary coordinate system as the first physical quantity information; as well as A magnetic flux angle calculator that generates a first estimate of the angle of the rotor of the motor based on the magnetic flux.

5. The sensorless control unit for a permanent magnet motor as described in claim 3, wherein the first estimator comprises: A first back EMF estimator estimates the back EMF of the motor in the stationary coordinate system as the first physical quantity information. as well as A back EMF angle calculator, which generates a first estimate of the angle of the rotor of the motor based on the back EMF.

6. The sensorless control unit for the permanent magnet motor as described in claim 1, wherein the second estimator further comprises: An axis converter transforms the voltage and current values ​​of the motor in the stationary coordinate system into the voltage and current values ​​of the motor in the estimated synchronous coordinate system, and uses this transformation for estimating the second physical quantity information. The axis converter sets an estimated synchronization coordinate system based on the D-axis for the actual D-axis estimation of the motor rotor, and sets the estimated synchronization coordinate system based on the first estimated value.

7. A sensorless control method for a permanent magnet motor, which is a model-based sensorless control method utilizing the voltage and current supplied to the permanent magnet motor, comprising: The first estimation step involves estimating first physical quantity information describing the operation of the motor in a stationary coordinate system and generating a first estimated value for the rotor angle of the motor based on the first physical quantity information. A second estimation step involves estimating information about a second physical quantity representing the back electromotive force of the motor in an estimation synchronization coordinate system set based on the first estimated value. The angular velocity of the motor rotor is calculated based on the physical quantity information, and the angular velocity is transmitted to the angular velocity estimation step of the second estimator. as well as An angle error calculation step is performed to calculate the error in the angle included in the first estimated value based on the angular velocity and the second physical quantity information.

8. The sensorless control method for a permanent magnet motor as described in claim 7, wherein the first estimation step includes: The magnetic flux estimation step, which uses the first physical quantity information to estimate the magnetic flux of the motor in the stationary coordinate system; as well as The magnetic flux angle calculation step is used to generate the first estimated value of the angle of the rotor of the motor based on the magnetic flux.

9. The sensorless control method for a permanent magnet motor as described in claim 7, wherein the first estimation step includes: The first back electromotive force estimation step, which uses the first physical quantity information to estimate the back electromotive force of the motor in the stationary coordinate system; as well as The back electromotive force angle calculation step is to generate the first estimated value of the angle of the rotor of the motor based on the back electromotive force.

10. The sensorless control method for a permanent magnet motor as described in claim 7, wherein the second estimation step further includes: The axis transformation step involves transforming the voltage and current values ​​of the motor in the stationary coordinate system to the estimated voltage and current values ​​of the motor in the synchronous coordinate system, and then using these values ​​for estimating the second physical quantity information. The axis transformation step sets an estimated synchronization coordinate system based on the D-axis for the actual D-axis estimation of the motor rotor, and sets the estimated synchronization coordinate system based on the first estimated value.

Images