System for controlling a synchronous machine and method thereof

Abstract

This invention relates to a system (100) and a method (200) for controlling a synchronizer (110). The system includes: a flux linkage estimation module (120) for estimating one or more values ​​of flux linkage (Ψ) in the synchronizer (110) based on a regression-based model; an iron loss estimation module (130) for estimating one or more values ​​of iron loss (PFe) in the synchronizer (110); and a control module (140). The control module (140) receives the one or more values ​​of flux linkage (Ψ) from the flux linkage estimation module (120) and the one or more values ​​of iron loss (PFe) from the iron loss estimation module (130), and determines, based on the values ​​of flux linkage (Ψ) and iron loss (PFe), an optimal current setpoint (Id*) along a vertical axis and an optimal current setpoint (Iq*) along a horizontal axis, thereby controlling the synchronizer (110).

Description

Technical Field

[0001] This invention relates to a system and method for controlling a synchronizer. Prior Technology

[0002] In conventional systems and methods for torque control of interior permanent magnet synchronous motors (IPMSMs), look-up table (LUT) methods are generally employed. LUT methods for IPMSM control are computationally simple because they do not require complex real-time calculations. However, constructing such look-up tables before their use remains a time-consuming and costly process. Furthermore, for LUT methods to be effective for torque control of IPMSMs, large memory usage and highly accurate interpolation algorithms are required.

[0003] In other attempts to use torque control for IPMSM, the least squares method is known for coefficient fitting in the algorithm. However, these methods can introduce and easily estimate errors.

[0004] Efficient control of IPMSMs requires loss-minimizing control and machine efficiency map generation. Advances in these loss-minimizing control and machine efficiency map generation necessitate accurate prediction of iron losses. However, conventional methods for torque control of IPMSMs using the Maximum Torque per Ampere (MTPA) do not consider iron loss estimation. In particular, the accuracy of existing methods for iron loss prediction differs significantly from the midpoint of the torque-speed range of the speed-torque envelope. This limitation severely restricts the use of conventional methods because iron losses may not be estimated over a wide range of torque and speed. Considering core saturation and cross-coupling across various speed ranges of synchronous machines, it is crucial that iron losses are taken into account for accurate torque control of synchronous machines.

[0005] It has been observed that iron losses are significant at high speeds compared to copper losses. Therefore, the inclusion of iron loss effects for MTPA-based torque control of synchronizers becomes necessary, especially for the application of synchronizers in electric vehicles (which frequently subject them to high-speed operation). More specifically, accurate iron loss estimation across the entire torque-speed range is essential during machine operating conditions and the experimental verification phase of the electric vehicle.

[0006] Specifically, electric vehicles require torque control systems and methods that have low computational burden and can therefore provide a fast and dynamic response to torque demand and a response with high torque estimation accuracy.

[0007] Therefore, this technology requires a system and method for controlling the synchronizing machine, which solves at least the aforementioned problems. Summary of the Invention

[0008] In one embodiment, the present invention relates to a system for controlling a synchronizer. The system comprises a flux linkage estimation module for estimating one or more values ​​of flux linkage in the synchronizer based on a regression-based model, an iron loss estimation module for estimating one or more values ​​of iron loss in the synchronizer, and a control module. The control module receives one or more values ​​of flux linkage from the flux linkage estimation module and one or more values ​​of iron loss from the iron loss estimation module, and determines an optimal current setpoint (Id*) along the direct axis and an optimal current setpoint (Iq*) along the horizontal axis based on these values ​​of flux linkage and iron loss. The control module then controls the synchronizer based on the determined optimal current setpoints.

[0009] In one embodiment of the present invention, the synchronizing machine is an internal permanent magnet synchronous motor.

[0010] In another embodiment of the invention, in the steady state prior to the operation of the synchronizer, the flux linkage estimation module based on the regression model is trained using one or more training values ​​of flux linkage (Ψd) in the direct axis of the synchronizer rotor and one or more training values ​​of flux linkage (Ψq) in the transverse axis of the synchronizer rotor.

[0011] In another embodiment of the invention, in the steady state prior to the operation of the synchronizing machine, the training values ​​of the flux linkage (Ψd) in the direct axis and the flux linkage (Ψq) in the transverse axis are based on the training values ​​of the flux linkage corresponding to six dissimilar equidistant distribution values ​​of the current (Iq) in the transverse axis of the rotor and five dissimilar equidistant distribution values ​​of the current (Id) in the direct axis of the rotor.

[0012] In another embodiment of the invention, during the operation of the synchronizer, the flux linkage estimation module estimates the value of flux linkage (Ψd) in the direct axis and the value of flux linkage (Ψq) in the horizontal axis based on a combination of input values ​​of current (Id) in the direct axis of the rotor and current (Iq) in the horizontal axis of the rotor.

[0013] In another embodiment of the invention, the iron loss estimation module estimates the iron loss based on at least two of the following: the iron loss value under short-circuit conditions, the iron loss value under open-circuit conditions, and the iron loss value under the base speed maximum torque-to-power ratio (MTPF) condition. In this document, if the torque demand is greater than 5% of the peak torque of the synchronizer, the iron loss estimation module estimates the iron loss based on the following: the iron loss value under short-circuit conditions and the iron loss value under the base speed MTPF condition. Conversely, if the torque demand is less than 5% of the peak torque of the synchronizer, the iron loss estimation module estimates the iron loss based on the following: the iron loss value under short-circuit conditions and the iron loss value under open-circuit conditions.

[0014] In another embodiment of the invention, the system further includes a first coordinate converter for converting a combination of input values ​​of current (Ia) in the first phase of the synchronous machine, current (Ib) in the second phase of the synchronous machine, and current (Ic) in the third phase of the synchronous machine into a combination of input values ​​of current (Id) in the direct axis of the rotor and current (Iq) in the transverse axis of the rotor.

[0015] In another embodiment of the invention, the system has a second coordinate converter for converting a combination of input values ​​of voltage (Va) in the first phase of the synchronous machine, voltage (Vb) in the second phase of the synchronous machine, and voltage (Vc) in the third phase of the synchronous machine into a combination of input values ​​of voltage (Vd) in the direct axis of the rotor and voltage (Vq) in the transverse axis of the rotor.

[0016] In another embodiment of the invention, the system has a voltage command generator for converting the determined optimal current setpoints (Id*, Iq*) into optimal voltage setpoints (Vd*, Vq*).

[0017] In another embodiment of the invention, the system has a third coordinate converter for converting the optimal voltage setpoint (Vd*, Vq*) into the optimal voltage setpoint (Vα*, Vβ*) in the α-β frame.

[0018] In another embodiment of the invention, the system has a space vector pulse width modulation power converter configured to perform the following operation: receive optimal voltage setpoints (Vα*, Vβ*) via a DC power supply for driving a synchronous machine.

[0019] In another embodiment, the present invention relates to a method for controlling a synchronizer. The method comprises the following steps: estimating one or more values ​​of flux linkage in the synchronizer based on a regression-based model using a flux linkage estimation module; estimating one or more values ​​of iron loss in the synchronizer using an iron loss estimation module; receiving one or more values ​​of flux linkage from the flux linkage estimation module and one or more values ​​of iron loss from the iron loss estimation module via a control module; and determining an optimal current setpoint (Id*) along the direct axis and an optimal current setpoint (Iq*) along the horizontal axis based on the values ​​of flux linkage and iron loss, thereby controlling the synchronizer via the control module.

[0020] In another embodiment, the present invention relates to a method for flux linkage estimation in a synchronizer. The method comprises the following steps: In a steady state prior to operation of the synchronizer, receiving one or more training values ​​of flux linkage (Ψd) in the direct axis of the synchronizer's rotor, based on six dissimilar equidistant distributions of rotor current (Iq) in the horizontal axis and five dissimilar equidistant distributions of rotor current (Id) in the vertical axis, and one or more training values ​​of flux linkage (Ψq) in the horizontal axis of the synchronizer's rotor. A second step involves sampling the training values ​​of the flux linkage. A third step involves splitting the sampled training values ​​into a first training dataset having known training values ​​of flux linkage (Ψd) in the vertical axis corresponding to the values ​​of rotor currents (Id, Iq), and a second training dataset having known training values ​​of flux linkage (Ψq) in the horizontal axis corresponding to the values ​​of rotor currents (Id, Iq). A fourth step involves using the first and second training datasets to train a regression-based model. The fifth step involves evaluating the regression-based model using a flux linkage estimation module based on R-scores. The final step involves estimating one or more values ​​of flux linkage (Ψd) in the direct axis of the rotor and one or more values ​​of flux linkage (Ψq) in the transverse axis of the rotor, based on the regression-based model, for a combination of input values ​​of the rotor current (Id) in the direct axis and the rotor current (Iq) in the transverse axis.

[0021] In another embodiment, the present invention relates to a method for estimating iron losses in a synchronous machine. The method includes a first step of receiving values ​​of iron losses under short-circuit conditions, iron losses under open-circuit conditions, and iron losses under the base speed maximum torque-to-full-pump (MTPF) condition. A second step involves checking whether the torque demand is higher or lower than 5% of the synchronous machine's peak torque. Based on the result of the second step, a third step involves estimating the iron loss based on the values ​​of iron losses under short-circuit conditions and iron losses under the base speed MTPF condition if the torque demand is higher than 5% of the peak torque. Alternatively, the third step involves estimating the iron loss based on the values ​​of iron losses under short-circuit conditions and iron losses under open-circuit conditions if the torque demand is lower than 5% of the peak torque. Simple Explanation of the Diagram

[0022] Reference will be made to embodiments of the invention, examples of which may be illustrated in the accompanying drawings. These drawings are intended to be illustrative and not restrictive. Although the invention is generally described in the context of these embodiments, it should be understood that the scope of the invention is not intended to be limited to these specific embodiments. [Figure 1] is a block diagram illustrating a system for controlling a synchronizer according to an embodiment of the present invention. [Figure 2] illustrates a method for controlling a synchronizer according to an embodiment of the present invention. [Figure 3] is a block diagram illustrating the initialization and implementation of a flux linkage estimation module in a system for controlling a synchronizer according to an embodiment of the present invention. [Figure 4] illustrates a method for estimating flux linkage in a synchronizer according to an embodiment of the present invention. [Figure 5] is a block diagram illustrating the initialization and implementation of an iron loss estimation module in a system for controlling a synchronizing machine according to an embodiment of the present invention. [Figure 6] illustrates a method for estimating iron loss in a synchronizing machine according to an embodiment of the present invention. Implementation

[0023] This invention relates to a system and method for controlling a synchronizer.

[0024] Figure 1 illustrates a block diagram of a system 100 for controlling a synchronizer 110. As shown in Figure 1, the system 100 for controlling the synchronizer 110 includes a flux linkage estimation module 120. The flux linkage estimation module 120 estimates one or more values ​​of flux linkage (Ψ) in the synchronizer 110. The flux linkage estimation module 120 utilizes a regression-based model for flux linkage estimation. The system 100 further includes an iron loss estimation module 130 for estimating one or more values ​​of iron loss (PFe) in the synchronizer 110. A control module 140 receives one or more values ​​of flux linkage from the flux linkage estimation module 120 and one or more values ​​of iron loss from the iron loss estimation module 130, and then determines an optimal current setpoint (Id*) along the direct axis of the rotor of the synchronizer and an optimal current setpoint (Iq*) along the transverse axis of the rotor based on the values ​​of flux linkage and iron loss. The optimal current setpoints are then used to control the synchronizer 110. In one embodiment of the invention, the synchronous motor 110 is an internal permanent magnet synchronous motor (IPMSM) driven by a DC power supply 180. In another embodiment of the invention, the IPMSM is adapted for electric vehicle applications.

[0025] As further illustrated in Figure 1, in an embodiment of the present invention, the control module 140 receives inputs from the flux linkage estimation module 120, the iron loss estimation module 130, and the operation target command generator module 170. The operation target command generator module 170 generates the required torque (Te*) for specific drive conditions of the IPMSM. The control module 140 further receives inputs via the speed calculation unit 168 of the maximum current (Im) of the synchronizer 110, the maximum voltage (Vdc) of the DC power supply 180, and the instantaneous angular velocity (ωe) of the synchronizer rotor, and based on all the aforementioned inputs, the control module 140 generates an instantaneous optimal current setpoint for driving the synchronizer 110.

[0026] In one embodiment of the invention shown in FIG1, the system 100 further includes a voltage command generator 150 for converting the determined optimal current setpoints (Id*, Iq*) along the direct and transverse axes of the rotor into optimal voltage setpoints (Vd*, Vq*) along the direct and transverse axes of the rotor. Subsequently, a third coordinate converter 156 of the system 100 converts the optimal voltage setpoints (Vd*, Vq*) into optimal voltage setpoints (Vα*, Vβ*) in the α-β frame. Additionally, the system includes a space vector pulse width modulation power converter 158 configured to receive the optimal voltage setpoints (Vα*, Vβ*) in the α-β frame for driving the synchronous machine 110 via a DC power supply 180 by converting this input (Vα*, Vβ*) into a three-phase input (Va, Vb, Vc) for voltages along the first, second, and third phases of the synchronous machine 110.

[0027] Correspondingly, the method 200 for controlling the synchronizer 110 is illustrated in FIG2. In step 2A, the flux linkage estimation module 120 (shown in FIG1) estimates one or more values ​​of flux linkage (Ψ) in the synchronizer 110 based on a regression-based model. Additionally, the iron loss estimation module 130 (shown in FIG1) estimates one or more values ​​of iron loss (PFe) in the synchronizer. In step 2B, the control module 140 (shown in FIG1) receives one or more values ​​of flux linkage from the flux linkage estimation module 120 and one or more values ​​of iron loss from the iron loss estimation module 130. In step 2C, the control module 140 determines the optimal current setpoint (Id*) along the vertical axis and the optimal current setpoint (Iq*) along the horizontal axis for controlling the synchronizer 110 based on the values ​​of flux linkage and iron loss, thereby driving the synchronizer 110.

[0028] Figure 3 illustrates a block diagram of the initialization and implementation of the flux linkage estimation module 120 in the system 100 used to control the synchronous machine 110. As shown in Figure 3, during operation of the synchronous machine 110, the flux linkage estimation module 120 receives input values ​​of the current (Id) in the direct axis of the rotor and the current (Iq) in the transverse axis of the rotor from a first coordinate converter 152. The first coordinate converter 152 converts the combination of the input values ​​of the current (Ia) in the first phase of the synchronous machine 110, the current (Ib) in the second phase of the synchronous machine 110, and the current (Ic) in the third phase of the synchronous machine 110 into a combination of the input values ​​of the current (Id) in the direct axis of the rotor and the current (Iq) in the transverse axis of the rotor. During operation of the synchronous machine 110, the flux linkage estimation module 120 further receives the values ​​of the voltage (Vd) in the direct axis of the rotor and the voltage (Vq) in the transverse axis of the rotor from a second coordinate converter 154. The second coordinate converter 154 converts the combination of input values ​​of the voltage (Va) in the first phase, the voltage (Vb) in the second phase, and the voltage (Vc) in the third phase of the synchronous machine 110 into a combination of input values ​​of the voltage (Vd) in the direct axis of the rotor and the voltage (Vq) in the transverse axis of the rotor. The flux linkage estimation module 120 further receives the input of the instantaneous angular velocity 110 (ωe) of the rotor of the synchronous machine via the speed calculation unit 168, and estimates the values ​​of flux linkage (Ψd) in the direct axis of the rotor and flux linkage (Ψq) in the transverse axis of the rotor based on all the aforementioned inputs.

[0029] As previously mentioned, the flux linkage estimation module 120 is based on a regression-based model. As is known, regression-based models predict the dependent variable based on the independent variable and need to be trained before implementation. In one embodiment of the invention, a multinomial regression model has been used to estimate the values ​​of flux linkage (Ψd) in the d-axis and flux linkage (Ψq) in the q-axis of the rotor of the synchronous machine 110, and the values ​​of inductance (Ld) in the d-axis and inductance (Lq) in the q-axis of the rotor of the synchronous machine 110. For the purposes of this invention, considering that the values ​​of flux linkage in the d-axis and q-axis contain information on saturation and cross-saturation, the relationship between flux linkage and stator current in a rotating dq coordinate system is given as follows: Where ψd and ψq represent the flux linkages along the d-axis and q-axis of the rotor of the synchronous machine 110, respectively, and ψm represents the flux linkage in the permanent magnet of the synchronous machine 110 affected by the d-axis current and q-axis current. Furthermore, Ldd and Lqq represent the self-inductance of the synchronous machine 110 along the d-axis and q-axis, respectively, and Ldq and Lqd represent the mutual inductance. The relationship between mutual inductance and flux is explained as follows.

[0030] As previously mentioned, the regression-based model needs to be trained before the operation of the synchronizer 110 for the above-described situation is implemented during the operation of the synchronizer 110. To train the regression-based model of the flux linkage estimation module 120, as shown in FIG3, in the steady state before the operation of the synchronizer 110, the first target command generator 172 generates one or more training values ​​for the flux linkage (Ψd) in the direct axis of the rotor (not shown) of the synchronizer 110 and one or more training values ​​for the flux linkage (Ψq) in the transverse axis of the rotor. In one embodiment of the invention, in the steady state before the operation of the synchronizer 110, the training values ​​for the flux linkage (Ψd) in the direct axis and the training values ​​for the flux linkage (Ψq) in the transverse axis are based on the training values ​​of the flux linkage corresponding to six dissimilar equidistant distribution values ​​of the current (Iq) in the transverse axis of the rotor and five dissimilar equidistant distribution values ​​of the current (Id) in the direct axis of the rotor.

[0031] Six distinct equidistant distributions of the current (Iq) in the transverse axis of the rotor and five distinct equidistant distributions of the current (Id) in the direct axis of the rotor provide 30 distinct operating points for training the regression-based model of the flux linkage estimation module 120. The distinct values ​​of Id and Iq used for training are fed to the flux linkage estimation module 120 via a voltage command generator 150, an SVPWM power converter 158, a first coordinate converter 152, and a second coordinate converter 154.

[0032] The 30 distinct operating points of Id and Iq have corresponding values ​​of Vd and Vq, and the flux linkage at these operating points in the steady state is calculated based on the following formula: ,

[0033] The regression model is therefore trained by feeding the values ​​of the independent and dependent variables into the model. The trained regression-based model is then used to estimate the flux linkage values ​​during the operation of the synchronizer 110. The errors in estimating the flux linkage values ​​in the d-axis and q-axis using the trained regression-based model are as low as 8.5 × 10⁻³% and 2.7 × 10⁻³%, respectively.

[0034] Correspondingly, the method 300 for estimating the flux linkage of the synchronizer 110 is illustrated in FIG. 4. The method involves the steps mentioned below. In step 4A, in the steady state prior to operation of the synchronizer 110 (shown in FIG. 3), one or more training values ​​of the flux linkage (Ψd) in the direct axis of the synchronizer's rotor, based on six dissimilar equidistant distribution values ​​of the rotor current (Iq) in the horizontal axis and five dissimilar equidistant distribution values ​​of the rotor current (Id) in the vertical axis, and one or more training values ​​of the flux linkage (Ψq) in the horizontal axis of the synchronizer's rotor are received by the flux linkage estimation module 120 (shown in FIG. 3). In step 4B, the training values ​​of the flux linkage are sampled by the flux linkage estimation module 120. In step 4C, the flux linkage estimation module 120 splits the sampled training values ​​into a first training dataset with known training values ​​of flux linkage (Ψd) on the vertical axis corresponding to rotor currents (Id, Iq), and a second training dataset with known training values ​​of flux linkage (Ψq) on the horizontal axis corresponding to rotor currents (Id, Iq). Next, in step 4D, the flux linkage estimation module 120 uses the first and second training datasets to train a regression-based model. In step 4E, the flux linkage estimation module 120 evaluates the training quality of the regression-based model based on the R² score. If the model's R² score is higher than a certain threshold (e.g., 0.95), the method proceeds; otherwise, the method reverts to step 4D and the degree of the regression-based model increases.

[0035] Finally, in step 4F, during the operation of the synchronizer 110, based on a regression-based model, the flux linkage estimation module 120 estimates one or more values ​​of flux linkage (Ψd) in the direct axis of the rotor and one or more values ​​of flux linkage (Ψq) in the transverse axis of the rotor using a combination of input values ​​of the rotor current (Id) in the direct axis and the rotor current (Iq) in the transverse axis.

[0036] Figure 5 illustrates a block diagram of the initialization and implementation of the iron loss estimation module 130 in the system 100 (shown in Figure 1) used to control the synchronizer 110. As shown in Figure 5, the iron loss estimation module 130 receives inputs representing the values ​​of iron loss (P Fe_SC) under short-circuit conditions, iron loss (P Fe_OC) under open-circuit conditions, and iron loss (P Fe_MTPF) under the base speed maximum torque flux ratio (MTPF) condition, which are associated with the magnetization flux path. A second target command generator 174 generates input commands to generate iron loss values ​​Id and Iq at various operating points for initializing the iron loss estimation module 130. In this paper, the iron loss under open-circuit conditions is generated by running the synchronizer 110 under no-load conditions at a base speed where Id and Iq are zero. Furthermore, iron losses under short-circuit conditions are generated by running the synchronizer 110 at a basic speed where Id is taken as the negative of the maximum current (Idmax) in the direct axis of the rotor and Iq is zero. Additionally, iron losses under MTPF conditions are generated by running the synchronizer 110 at a basic speed where Id is taken as the negative of the maximum current (Idmax) in the direct axis of the rotor and Iq is taken as the negative of the maximum current (Iqmax) in the transverse axis of the rotor. The iron loss module further receives the instantaneous angular velocity (ωe) of the rotor of the synchronizer 110 and the values ​​of the flux linkage (Ψd) in the d-axis and the flux linkage (Ψq) in the q-axis from the flux linkage estimation module 120 via the speed calculation unit 168.

[0037] The iron loss function can be described as: The coefficients (ah, ae, ax), (bh, be, bx), and (ch, ce, cx) are determined by measurement data under open-circuit, short-circuit, and MTPF operation conditions, respectively.

[0038] As mentioned above, the necessary values ​​for Vm, Vda, and Vmf used in calculating iron loss are calculated as follows: ……(7) Wherein ψd_MTPF and ψq_MTPF are the d-axis and q-axis flux linkages obtained by the flux linkage estimation module 120 under the MTPF condition at the basic velocity.

[0039] For any given operating point of the synchronizer 110, the iron loss is the sum of PFe_oc and PFe_sc or PFe_MTPF and PFe_sc. In the implementation of the iron loss estimation module 130 in the system 100 for controlling the synchronizer 110, the iron loss estimation module 130 also receives input from the operating target command generator 170 (shown in FIG. 1) for the torque demand (Te*) at the instantaneous operating point. If the torque demand is greater than 5% of the peak torque (Te peak) of the synchronizer 110, the iron loss estimation module 130 estimates the iron loss based on the following: the value of the iron loss (PFe_sc) under short-circuit conditions, and the value of the iron loss (PFe_MTPF) under the basic speed MTPF conditions. Alternatively, if the torque demand (Te*) is less than 5% of the peak torque (Te peak) of the synchronous machine 110, the iron loss estimation module 130 estimates the iron loss based on the following: the value of the iron loss (PFe_sc) under short-circuit conditions and the value of the iron loss (PFe_oc) under open-circuit conditions. The relative error of the iron loss estimation is only 2% to 4% in the torque-speed range in which the synchronous machine operates continuously.

[0040] Correspondingly, a method 400 for estimating the iron loss in the synchronizer 110 (shown in FIG. 5) is illustrated in FIG. 6. Method 400 involves the steps mentioned below. In step 6A, the iron loss estimation module 130 (shown in FIG. 5) receives the values ​​of the iron loss (PFe_sc) under short-circuit conditions, the iron loss (PFe_oc) under open-circuit conditions, and the iron loss (PFe_MTPF) under the base speed maximum torque-to-power ratio (MTPF) condition. In step 6B, the iron loss estimation module 130 checks whether the torque demand (Te*) is higher or lower than 5% of the peak torque (Te peak) of the synchronizer 110. Based on the result of step 6B, in step 6C, if the value of the torque demand (Te*) is greater than 5% of the peak torque (Te peak), then the iron loss estimation module 130 estimates the iron loss based on the following: the value of the iron loss (PFe_sc) under short-circuit conditions, and the value of the iron loss (PFe_MTPF) under the basic speed MTPF conditions. Alternatively, in step 6C, if the torque demand (Te*) is less than 5% of the peak torque (Te peak), then the iron loss estimation module 130 estimates the iron loss based on the following: the value of the iron loss (PFe_sc) under short-circuit conditions, and the value of the iron loss (PFe_oc) under open-circuit conditions.

[0041] Finally, after initializing the flux linkage estimation module and the iron loss estimation module, when implemented in the system to control the synchronous machine, the control module determines the optimal current setpoint for the given torque requirement. The control system for the synchronous machine based on MTPA is achieved by the following equation: Where T is the torque corresponding to the maximum torque-Ampere ratio control and P is the number of pole pairs in the synchronous machine.

[0042] Advantageously, the present invention provides a system and method for controlling a synchronizer with a highly accurate flux linkage estimation module that takes into account the effects of saturation and cross-coupling under the real-time operating conditions of the synchronizer. Furthermore, the proposed iron loss estimation module can estimate the iron loss of a high-speed, high-power IPMSM with low complexity based on data from only three measurement points. The low computation time and low memory utilization required for flux linkage and iron loss estimation for synchronizer control according to the present invention result in a rapid dynamic response to torque demands, making the current system suitable for electric vehicle applications.

[0043] In this invention, the error of flux estimation is within 1% and the error of iron loss estimation is within 4%, which is highly necessary during MTPA control implementation to drive machines with different torque requirements.

[0044] Furthermore, the method proposed in this invention readily achieves the real-time operating conditions of the synchronous machine, and the proposed method is applicable to the control and characteristics of the synchronous machine without requiring any information regarding the geometry and the placement of magnets in the rotor.

[0045] While the invention has been described with respect to certain embodiments, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the scope of the invention as defined by the following claims.

[0046] 100: System 110: Synchronizer 120: Magnetic flux estimation module 130: Iron Loss Estimation Module 140: Control Module 150: Voltage command generator 152: First Coordinate Converter 154: Second Coordinate Converter 156: Third Coordinate Converter 158: Space Vector Pulse Width Modulation Power Converter 168: Speed ​​Calculation Unit 170: Operation Target Command Generator Module 172: First Target Command Generator 180: DC power supply 200, 300, 400: Method 2A, 2B, 2C, 4A, 4B, 4C, 4D, 4E, 4F, 6A, 6B, 6C: Steps Ψ,Ψ d,Ψ q: Magnet chain I d*: Optimal current setpoint along a straight axis I q*: Optimal current setpoint along a horizontal axis I q: Current in the horizontal axis Id: Current in the direct axis I a: Current in the first phase of the synchronous machine Ib: Current in the second phase of the synchronous machine Ic: The current in the third phase of the synchronous machine. P Fe: Iron loss P Fe_sc: Iron loss P Fe_oc: Iron loss T e*: Torque requirement Peak torque: Va: Voltage in the first phase of the synchronous machine. Vb: The voltage in the second phase of the synchronous machine. Vc: The voltage in the third phase of a synchronous machine. Vd*, Vq*: Optimal voltage setpoint Vα*, Vβ*: Optimal voltage setpoints within the α-β framework.

Claims

1. A system (100) for controlling a synchronizer (110), comprising: a flux linkage estimation module (120) for estimating one or more values ​​of flux linkage (Ψ) in the synchronizer (110) based on a regression-based model; an iron loss estimation module (130) for estimating one or more values ​​of iron loss (PFe) in the synchronizer (110); and a control module (140) for receiving the one or more values ​​of flux linkage (Ψ) from the flux linkage estimation module (120) and the one or more values ​​of iron loss (PFe) from the iron loss estimation module (130), and determining, based on the values ​​of flux linkage (Ψ) and iron loss (PFe), an optimal current setpoint (Id*) along a vertical axis and an optimal current setpoint (Iq*) along a horizontal axis, thereby controlling the synchronizer (110). The iron loss estimation module (130) estimates the iron loss (PFe) based on the following: the value of the iron loss (PFe_sc) under short-circuit conditions and the value of the iron loss (PFe_OC) under open-circuit conditions, wherein the synchronous machine (110) is an internal permanent magnet synchronous motor, and wherein a polynomial regression model is used to estimate one or more values ​​of the flux linkage in the direct axis and the flux linkage in the transverse axis of the rotor of the synchronous machine (110).

2. The system (100) of claim 1, wherein in a steady state prior to operation of the synchronizer (110), the flux linkage estimation module (120) based on the regression-based model is trained using one or more training values ​​of the flux linkage (Ψd) in the direct axis of the rotor of the synchronizer (110) and one or more training values ​​of the flux linkage (Ψq) in the transverse axis of the rotor of the synchronizer (110).

3. The system (100) of claim 2, wherein in a steady state prior to operation of the synchronizer (110), the training values ​​of the flux linkage (Ψd) in the direct axis and the training values ​​of the flux linkage (Ψq) in the transverse axis are based on the training values ​​of the flux linkage (Ψ) corresponding to six dissimilar equidistant distribution values ​​of the current (Iq) in the transverse axis of the rotor and five dissimilar equidistant distribution values ​​of the current (Id) in the direct axis of the rotor.

4. The system (100) of claim 2, wherein during operation of the synchronizer (110), the flux linkage estimation module (120) estimates the values ​​of the flux linkage (Ψd) in the direct axis and the values ​​of the flux linkage (Ψq) in the horizontal axis for a combination of input values ​​of the current (Id) in the direct axis of the rotor and the current (Iq) in the horizontal axis of the rotor.

5. The system (100) of claim 1, wherein the iron loss estimation module (130) further estimates the iron loss (PFe) based on the value of the iron loss (PFe_MTPF) under the condition of base speed maximum torque flux ratio (MTPF).

6. The system (100) of request item 5, wherein the torque demand (Te*) is greater than 5% of the peak torque (Tepeak) of the synchronous machine (110).

7. The system (100) of request item 5, wherein the torque demand (Te*) is less than 5% of the peak torque (Tepeak) of the synchronous machine (110).

8. The system (100) of claims 4 and 5, comprising a first coordinate converter (152) for converting a combination of input values ​​of current (Ia) in a first phase of the synchronous machine, current (Ib) in a second phase of the synchronous machine and current (Ic) in a third phase of the synchronous machine into the same combination of input values ​​of current (Id) in the direct axis of the rotor and current (Iq) in the transverse axis of the rotor.

9. The system (100) of claim 4, comprising a second coordinate converter (154) for converting a combination of input values ​​of voltage (Va) in a first phase of the synchronous machine, voltage (Vb) in a second phase of the synchronous machine and voltage (Vc) in a third phase of the synchronous machine into a combination of input values ​​of voltage (Vd) in the direct axis of the rotor and voltage (Vq) in the transverse axis of the rotor.

10. The system (100) of claim 1 includes a voltage command generator (150) for converting the determined optimal current setpoints (Id*, Iq*) into an optimal voltage setpoint (Vd*, Vq*).

11. The system (100) of claim 10 includes a third coordinate converter (156) for converting the optimal voltage setpoint (Vd*, Vq*) into one of the optimal voltage setpoints (Vα*, Vβ*) in an α-β frame.

12. The system (100) of claim 11 includes a space vector pulse width modulation power converter (158) configured to perform the following operations: receive the optimal voltage setpoint (Vα*, Vβ*) for driving the synchronous machine (110) via a DC power supply (180).

13. A method (200) for controlling a synchronizer (110), comprising the steps of: estimating one or more values ​​of flux linkage (Ψ) in the synchronizer (110) based on a regression-based model by a flux linkage estimation module (120); estimating one or more values ​​of iron loss (PFe) in the synchronizer (110) by an iron loss estimation module (130); receiving the one or more values ​​of flux linkage (Ψ) from the flux linkage estimation module (120) and the one or more values ​​of iron loss (PFe) from the iron loss estimation module (130) by a control module (140); and determining, by the control module (140) based on the values ​​of flux linkage (Ψ) and iron loss (PFe), an optimal current setpoint (Id*) along a vertical axis and an optimal current setpoint (Iq*) along a horizontal axis, thereby controlling the synchronizer (110). The iron loss estimation module (130) estimates the iron loss (PFe) based on the following: the value of the iron loss (PFe_sc) under short-circuit conditions and the value of the iron loss (PFe_OC) under open-circuit conditions, wherein the synchronous machine (110) is an internal permanent magnet synchronous motor, and wherein a polynomial regression model is used to estimate one or more values ​​of the flux linkage in the direct axis and the flux linkage in the transverse axis of the rotor of the synchronous machine (110).

14. A method (300) for estimating flux linkage in a synchronizer (110), comprising the steps of: in a steady state prior to operation of the synchronizer (110), receiving, by means of a flux linkage estimation module (120), one or more training values ​​of flux linkage (Ψd) in the direct axis of the rotor of the synchronizer (110) based on six dissimilar equidistant distribution values ​​of rotor current (Iq) in a horizontal axis of a rotor and five dissimilar equidistant distribution values ​​of rotor current (Id) in a vertical axis of the rotor; and sampling the training values ​​of flux linkage (Ψ) by means of the flux linkage estimation module (120); The flux linkage estimation module (120) splits the sampled training values ​​into a first training data set with known training values ​​of the flux linkage (Ψd) in the vertical axis corresponding to the rotor current values ​​(Id, Iq) and a second training data set with known training values ​​of the flux linkage (Ψq) in the horizontal axis corresponding to the rotor current values ​​(Id, Iq); the flux linkage estimation module (120) uses the first training data set and the second training data set to train a regression-based model; The flux linkage estimation module (120) evaluates the regression-based model based on an R-score; and during the operation of the synchronizer (110), based on the regression-based model, the flux linkage estimation module (120) estimates one or more values ​​of the flux linkage (Ψd) in the direct axis of the rotor and one or more values ​​of the flux linkage (Ψq) in the horizontal axis of the rotor for a combination of input values ​​of the rotor current (Id) in the direct axis and the rotor current (Iq) in the horizontal axis of the rotor.

15. A method (400) for estimating iron losses in a synchronous machine (110), comprising the steps of: receiving, by means of an iron loss estimation module (130), the values ​​of the iron loss (PFe_sc) under short-circuit conditions, the iron loss (PFe_oc) under open-circuit conditions, and the iron loss (PFe_MTPF) under basic speed maximum torque-to-power ratio (MTPF) conditions; and by means of the iron loss estimation module (130) checking whether a torque demand (Te*) is higher or lower than 5% of a peak torque (Tepeak) of the synchronous machine; If the torque demand (Te*) is higher than 5% of the peak torque (Tepeak), the iron loss (PFe) is estimated by the iron loss estimation module (130) based on the following: the iron loss (PFe_sc) under the short circuit condition and the iron loss (PFe_MTPF) under the basic speed MTPF condition; and if the torque demand (Te*) is lower than 5% of the peak torque (Tepeak), the iron loss (PFe) is estimated by the iron loss estimation module (130) based on the following: the iron loss (PFe_sc) under the short circuit condition and the iron loss (PFe_oc) under the open circuit condition.

Images