A phase current reconstruction method and device based on duty cycle model predictive control

Abstract

The present invention discloses a phase current reconstruction method and device based on duty cycle model predictive control, the method comprising: constructing a drive system with duty cycle model predictive control as a modulation mode; reconstructing the system phase current using a multi-position coupling sampling method; obtaining a reconstruction blind area when the modulation mode is duty cycle model prediction according to the minimum sampling time; the reconstruction blind area mainly includes a low modulation blind area and an over-modulation blind area; eliminating the reconstruction blind area to realize a phase current reconstruction method based on duty cycle model predictive control. The present invention improves the duty cycle model predictive control algorithm to reduce the amount of calculation, and achieves a good reconstruction effect through a multi-position coupling sampling method and optimizing the action time of two zero voltage vectors.

Description

Technical Field

[0001] The present invention belongs to the technical field of permanent magnet synchronous motor control technology, and in particular relates to a phase current reconstruction method and device based on duty cycle model measurement and control. Background Art

[0002] In recent years, with the continuous research and application of high-performance permanent magnet materials, the volume and mass of permanent magnet synchronous motor (PMSM) have been greatly reduced, and its power density has been greatly improved. It has been widely developed and applied in aerospace, military, medical equipment, new energy electric vehicles and other fields. In the PMSM drive control system, there are generally two methods for motor winding current detection: high-precision sampling resistors and Hall current sensors. These two detection methods have their own advantages and disadvantages. High-precision sampling resistors are cheap and can reduce system costs. However, since they need to be connected to the circuit for use, when a large current flows through, it will cause heat and bring additional losses to the controller. The Hall-type current sensor does not need to be electrically connected to the motor drive controller circuit. As long as the line to be tested passes through the perforation of the current sensor, the line current can be detected. It has higher sampling accuracy and reliability, but its installation requires a certain space and the price cost is also higher. The above two methods require two or three detection lines in vector control. Multiple lines are not only costly but also have poor reliability. In order to solve the above problems, the phase current reconstruction method of the single detection line came into being.

[0003] The phase current reconstruction currently used is mainly based on the traditional permanent magnet synchronous motor vector control method based on the space vector modulation technology (SVPWM) algorithm. This method will cause some inherent errors when reconstructing the phase current, and it is difficult to meet the requirements of low calculation amount and high accuracy at the same time. As a result, the phase current reconstruction method cannot be used to obtain three-phase current information in scenarios with high requirements for current information, and the application scenarios are limited. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a phase current reconstruction method and device based on duty cycle model predictive control to solve the above technical problems.

[0005] In a first aspect, an embodiment of the present application provides a phase current reconstruction method based on a duty cycle model prediction algorithm, the method comprising:

[0006] The drive system is constructed using duty cycle model predictive control as the modulation method;

[0007] The multi-position coupled sampling method is used to reconstruct the system phase current;

[0008] According to the minimum sampling time, a reconstruction blind area when the modulation mode is predicted by the duty cycle model is obtained; the reconstruction blind area includes a low modulation blind area and an over modulation blind area;

[0009] Eliminate the reconstruction blind area and realize the phase current reconstruction method based on duty cycle model predictive control.

[0010] Preferably, the duty cycle model predictive control includes:

[0011] Establish a predictive stator current model for permanent magnet synchronous motor in a stationary coordinate system;

[0012] The reference current at the next moment and the predicted stator current model are used to obtain the angle of the reference voltage vector, and the optimal voltage vector is selected according to the angle of the reference voltage vector;

[0013] According to the relationship between the reference current vector and the predicted current vector at the next moment, the duty cycle triangle is obtained, and the optimal effective voltage vector action time is calculated with the help of the cosine theorem.

[0014] Preferably, the optimal effective voltage vector action time is:

[0015]

[0016] In the formula, λ k+1 is a vector value proportional to the reference voltage vector, is the effective voltage vector u4 acting at time k, L S is the synchronous inductance, θ n is the effective voltage vector u4 and λ k+1 The angle between.

[0017] Preferably, the reconstruction blind zone of the low modulation area is a reconstruction blind zone caused by the effective voltage vector action time being less than 2 times of the minimum sampling time; the reconstruction blind zone of the overmodulation area is a reconstruction blind zone caused by the zero voltage vector action time being less than 2 times of the minimum sampling time.

[0018] Preferably, the elimination of the reconstruction blind zone includes: utilizing the mutual cooperation of two zero voltage vectors to eliminate the reconstruction blind zone of the low modulation area and part of the over modulation area; selecting a suitable pulse width modulation period to exclude the reconstruction blind zone of the over modulation area from the linear modulation area.

[0019] Preferably, the mutual cooperation of the two zero vectors is used to eliminate the reconstruction blind area of ​​the low modulation area and part of the over modulation area; specifically including:

[0020] The zero voltage vector of each cycle is completed by the joint action of U0 (000) and U7 (111). The action time T7 of the zero voltage vector U7 (111) is greater than the minimum sampling time T min , the action vector of each cycle consists of 2 zero voltage vectors and 1 effective voltage vector;

[0021] According to whether the total action time of the zero voltage vector can make the action time T0 of the zero voltage vector U0 (000) and the action time T7 of the zero voltage vector U7 (111) meet the sampling requirements at the same time, the entire linear modulation area is divided into a double zero vector area and a single zero vector area;

[0022] Select available voltage vectors according to the divided modulation areas:

[0023] When in the double zero vector region, the two available vectors are U0 (000) and U7 (111);

[0024] When in the single zero vector region, the two available vectors are U7(111) and the effective voltage vector U i ;

[0025] When in the reconstruction blind area, no vectors are available.

[0026] Preferably, the selecting of a suitable pulse width modulation period to exclude the reconstruction blind area of ​​the overmodulation area from the linear modulation area specifically includes:

[0027] Let time T M Greater than the minimum duration T min , according to the knowledge of plane geometry, time T M for

[0028]

[0029] T M >T min

[0030] Where, T M It is the difference between the maximum action time of the space voltage vector in the overmodulation area and the maximum action time in the linear modulation area.

[0031] Minimum duration T min With PWM switching period T s The relationship is:

[0032]

[0033] Choose the appropriate PWM switching period T s , phase current reconstruction is performed in the entire linear modulation region.

[0034] Preferably, the method further includes: by establishing an incremental model of the stator current prediction value of the permanent magnet synchronous motor at time k+1, obtaining the current prediction value of the aligned sampling point, so as to reduce the phase current reconstruction error of the permanent magnet synchronous motor; the incremental model of the stator current prediction value at time k+1 is:

[0035]

[0036] In the formula, i d 、i q is the stator current of dq axis, u d 、u q is the stator voltage of dq axis, R S , L S , f are stator resistance, synchronous inductance, permanent magnet flux, ω e is the electrical angular velocity, T s is the pulse width modulation period, is the average current value of the dq axis at time k, is the average value of the applied voltage vector in the k time period of the dq axis.

[0037] In a second aspect, an embodiment of the present application provides a phase current reconstruction device based on duty cycle model predictive control, the device comprising:

[0038] A duty cycle model prediction module is used to realize the construction of a drive system using duty cycle model prediction control as a modulation method;

[0039] A phase current reconstruction module is used to reconstruct the phase current of a drive system constructed with duty cycle model predictive control as a modulation mode by adopting a multi-position coupled sampling method;

[0040] A reconstruction blind zone elimination module is used to obtain the reconstruction blind zone when the modulation mode is predicted by the duty cycle model according to the minimum sampling time, and eliminate the reconstruction blind zone;

[0041] The reconstruction blind area includes an undermodulation blind area and an overmodulation blind area.

[0042] Preferably, the device also includes a reconstruction error reduction module, which is used to reduce the permanent magnet synchronous motor phase current reconstruction error by establishing an incremental model of the stator current prediction value of the permanent magnet synchronous motor at time k+1 to obtain the current prediction value of the aligned sampling point.

[0043] Beneficial effects:

[0044] The phase current reconstruction method based on optimal duty cycle model predictive control proposed in the present invention improves the optimal duty cycle model prediction algorithm in order to avoid the problem of excessive system calculation. Compared with the traditional algorithm, this algorithm can more efficiently select the optimal voltage vector and calculate the three-phase duty cycle; in order to solve the problem of phase current reconstruction blind spot, it is proposed to use a multi-position coupling sampling method to reconstruct the phase current. This strategy can detect effective current in 8 output vectors, making phase current reconstruction based on optimal duty cycle model predictive control possible; the phase current reconstruction blind spot is reduced by using zero voltage vectors U0 (000) and U7 (111), and by selecting a suitable pulse width modulation period, there is no reconstruction blind spot in the entire linear modulation area. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0046] Figure 1 A flow chart of a phase current reconstruction method based on optimal duty cycle model predictive control provided in one embodiment.

[0047] Figure 2 It is a duty cycle calculation vector diagram provided by an embodiment of the present disclosure.

[0048] Figure 3 This is a diagram of the installation position of a single current sensor for multi-position coupled sampling provided in an embodiment of the present disclosure.

[0049] Figure 4 This is the phase current reconstruction blind area based on the prediction of the optimal duty cycle model provided by the embodiment of the present disclosure.

[0050] Figure 5 This is a phase current reconstruction vector diagram improved by the method of the present invention.

[0051] Figure 6 It is the BC phase current within a PWM cycle of the control case based on the optimal duty cycle model prediction of the present invention.

[0052] Figure 7 A schematic diagram of a phase current reconstruction device based on a duty cycle model prediction algorithm provided in one embodiment of the present invention.

[0053] Figure 8 It is a schematic diagram of a phase current reconstruction device based on a duty cycle model prediction algorithm provided in yet another embodiment of the present invention.

[0054] Fig. 9 This is a current error diagram after current reconstruction processing is performed using the present invention.

[0055] Fig.10 It is the current error diagram after the reconstruction error reduction process. DETAILED DESCRIPTION

[0056] The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them.

[0057] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The optional embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0059] In recent years, the model predictive control strategy of permanent magnet synchronous motor has been widely used. The phase current reconstruction based on model predictive control not only solves the problem of reconstruction blind area, but also solves the problem of excessive system calculation. The present invention improves the optimal duty cycle model predictive control algorithm to reduce the amount of calculation, and achieves a good reconstruction effect through multi-position coupling sampling and optimization of the action time of two zero voltage vectors. Figure 1 As shown, the embodiment of the present disclosure provides a phase current reconstruction method based on a duty cycle model prediction algorithm, and the control method specifically includes the following method steps:

[0060] S100: constructing a drive system using duty cycle model predictive control as a modulation method. Specifically, this step includes:

[0061] Establish a predictive stator current model for permanent magnet synchronous motor in a stationary coordinate system;

[0062] The reference current at the next moment and the predicted stator current model are used to obtain the angle of the reference voltage vector, and the optimal voltage vector is selected according to the angle of the reference voltage vector;

[0063] According to the relationship between the reference current vector and the predicted current vector at the next moment, the vector triangle is obtained, and the optimal effective voltage vector action time is calculated with the help of the cosine theorem.

[0064] In the embodiment of the present application, a predicted stator current model of a permanent magnet synchronous motor in a stationary coordinate system is established, specifically:

[0065] In the stationary αβ coordinate system, the surface mounted permanent magnet synchronous motor model with equal d-axis and q-axis inductance can be expressed in complex vector form:

[0066]

[0067]

[0068] In the formula, is the rotor flux vector; R S , L S , f ,θ e , p are stator resistance, synchronous inductance, permanent magnet flux, electric rotor angle, and pole pair number respectively; u s ,i S and T e are the stator voltage vector, stator current vector and electromagnetic torque respectively, ω r =dθ e / dt is the motor rotor speed.

[0069] By discretizing the voltage mathematical model in the above formula, the stator current prediction model at the next sampling moment can be obtained as follows:

[0070]

[0071] In the embodiment of the present application, the reference current at the next moment and the predicted stator current model are used to obtain the angle of the reference voltage vector, and the optimal voltage vector is selected according to the angle of the reference voltage vector. Specifically:

[0072] When the duty cycle model predictive control strategy is adopted, the action vector of one cycle includes an effective voltage vector and a zero voltage vector. When the effective voltage vector acts, the stator predicted current is:

[0073]

[0074] When the zero voltage vector acts, the predicted stator current is:

[0075]

[0076] Where, t on is the effective voltage vector action time.

[0077] In the above formula, i s ,ω r ,θ e The change within the period has little effect on the second term, so we can get

[0078]

[0079] The predicted stator current after a complete cycle is:

[0080]

[0081] make The latter formula becomes:

[0082]

[0083] like but like but is the reference current at time k+1, is the predicted current at time k+1. The relationship is as follows Figure 2 shown.

[0084] Predict the current at time k+1 When k+1, the reference current Known, λ is calculated by the following formula k+1 The vector value of .

[0085]

[0086] λ is calculated by the following formula k+1 The angle θ in the αβ coordinate system is used to select the optimal effective voltage vector The angle θ is:

[0087] θ=arctan(λ k+1 ).

[0088] In the embodiment of the present application, the vector triangle is obtained based on the relationship between the reference current vector and the predicted current vector at the next moment, and the optimal effective voltage vector action time is calculated by means of the cosine theorem. Specifically:

[0089] Using Lambda k+1 The projection on the selected effective voltage vector determines the time t for which the effective voltage vector needs to be applied. on This case is based on Figure 2 A brief analysis of the situation shown in Figure 2 The optimal effective voltage vector selected according to θ is U4(011), and the effective voltage vector U4(011) is proportional to λ k+1 The angle between them is θ n ,from Figure 2 It can be found that k+1 With t on different, and at the same time and The distance between them will also change. and Applying the cosine theorem of triangles to the triangles composed of and The distance between k+1 for:

[0090]

[0091] when and The distance between k+1 When the minimum value is obtained, the system obtains the optimal duty cycle, and the optimal effective voltage vector action time t is calculated at this time on for:

[0092]

[0093] In the formula, λ k+1 is a vector value proportional to the reference voltage vector, is the effective voltage vector u4 acting at time k, L S is the synchronous inductance, θ n is the effective voltage vector u4 and λ k+1 The angle between.

[0094] This embodiment proposes a new duty cycle model predictive control algorithm, which can efficiently select the optimal voltage vector and calculate the three-phase duty cycle without listing all candidate voltage vectors.

[0095] S200: Reconstruct system phase current using multi-position coupled sampling method.

[0096] When the action time is greater than the minimum sampling time T min When there are less than two available voltage vectors, a reconstruction blind spot will appear. When the optimal duty cycle model prediction algorithm is used, there are only two voltage vectors in one cycle, and one of the vectors is fixed as a zero voltage vector. In order to better realize the phase current reconstruction, this case adopts the multi-position coupling sampling method of the optimal selection combination to reconstruct the system phase current. The single current sensor position of the multi-position coupling sampling is installed as follows: Figure 3 This combination method can detect effective current when eight voltage vectors are output, so the eight vectors are all available vectors. The corresponding relationship between the detection current of the single current sensor and the three-phase winding current during the eight vectors is shown in Table 1.

[0097] Table 1 shows the relationship between the current detected by the current sensor and the three-phase currents A, B, and C in different switching states

[0098]

[0099] The phase current reconstruction strategy based on duty cycle model prediction of multi-position coupled sampling reduces the current reconstruction blind area of ​​single current sensor. This strategy can detect effective current in 8 output vectors, making phase current reconstruction based on optimal duty cycle model prediction algorithm possible.

[0100] S300: According to the minimum sampling time, a reconstruction blind area when the modulation mode is predicted by the duty cycle model is obtained; the reconstruction blind area includes an under-modulation blind area and an over-modulation blind area.

[0101] In the embodiment of the present application, when the action time is greater than the minimum sampling time T min When there are less than two available voltage vectors, a reconstruction blind zone occurs. When the optimal duty cycle model prediction algorithm is used, there are only two voltage vectors in one cycle, and one of the vectors is fixed as a zero voltage vector.

[0102] The reconstruction blind area caused by the zero voltage vector action time being less than 2 times the minimum sampling time is called the overmodulation area; the reconstruction blind area caused by the effective voltage vector action time being less than 2 times the minimum sampling time is called the undermodulation area. Figure 4As shown, the gray area is called the overmodulation area, which is caused by the zero voltage vector action time being less than 2 times the minimum sampling time; the light gray area is called the undermodulation area, which is caused by the effective voltage vector action time being less than 2 times the minimum sampling time.

[0103] S400: Eliminate the reconstruction blind area and implement a phase current reconstruction method based on duty cycle model predictive control.

[0104] In the embodiment of the present application, analysis of Table 1 shows that when the effective voltage vector is U3 (001), the three-phase current reconstruction cannot be completed based on only the two vectors U3 (001) and U0 (000), and in the low modulation area, this is caused by insufficient action time of the effective voltage vector. In order to solve these two problems, the present invention completes the zero voltage vector of each cycle by the joint action of U0 (000) and U7 (111).

[0105] The action time of the zero voltage vector U7 (111) is T7>T min Therefore, there will be three vector effects in each cycle. Figure 5 As shown, the available voltage vector selection method for current reconstruction is as follows:

[0106] (1) When the action times T0 and T7 of the zero voltage vector U0 (000) and the zero voltage vector U7 (111) meet the requirements, the zero voltage vector is preferred, and the sampling point position is the same as the traditional zero vector voltage detection method;

[0107] (2) When the total zero voltage vector action time cannot make T0 and T7 meet the sampling requirements at the same time, except when the effective voltage vector is U6 (101), T0 is preferentially satisfied to meet the sampling requirements. In other cases, T7 is preferentially satisfied to meet the sampling requirements. At this time, the effective voltage vector action time must meet the sampling requirements. Therefore, there are still two voltage vectors that meet the sampling requirements.

[0108] (3) When the total zero voltage vector action time makes T0 and T7 unable to meet the sampling requirements, only the effective voltage vector action time meets the sampling requirements, so a reconstruction blind area appears.

[0109] Specifically, the two zero voltage vectors are used to cooperate with each other to eliminate the reconstruction blind area in the low modulation area and part of the over modulation area. Specifically, the following steps are included:

[0110] The zero voltage vector of each cycle is completed by the joint action of U0 (000) and U7 (111). The action time T7 of the zero voltage vector U7 (111) is greater than the minimum sampling time T min , the action vector of each cycle consists of 2 zero voltage vectors and 1 effective voltage vector;

[0111] According to whether the total action time of the zero voltage vector can make the action time T0 of the zero voltage vector U0 (000) and the zero voltage vector U7 (111) meet the sampling requirements at the same time, the entire linear modulation area is divided into a double zero vector area, a single zero vector area and a reconstruction blind area;

[0112] Select available voltage vectors according to the divided modulation areas:

[0113] When in the double zero vector region, the two available vectors are U0 (000) and U7 (111);

[0114] When in the single zero vector region, the two available vectors U7 (111) and the effective voltage vector U i ;

[0115] When in the reconstruction blind area, no vectors are available.

[0116] The available voltage vectors in different regions are shown in Table 2.

[0117] Table 2 shows the available voltage vectors in different regions

[0118]

[0119] If the reconstruction blind area is always outside the linear modulation area, the entire linear modulation area can always be guaranteed to have two available and different voltage vectors. At this time, by selecting a suitable pulse width modulation period, the reconstruction blind area is always outside the linear modulation area, specifically including:

[0120] Let time T M Greater than the minimum duration T min , according to the knowledge of plane geometry, time T M for:

[0121]

[0122] T M >T min

[0123] Where, T M It is the difference between the maximum action time of the space voltage vector in the overmodulation area and the maximum action time in the linear modulation area.

[0124] In order to make the reconstruction blind area always in the overmodulation area, the minimum duration T min With PWM switching period T s The relationship is simplified to:

[0125]

[0126] Therefore, as long as the appropriate PWM switching period T is selected s, the prerequisite for phase current reconstruction in the entire linear modulation area (two available different voltage vectors) can be met.

[0127] In this embodiment, the zero voltage vectors U0 (000) and U7 (111) are used together to eliminate the reconstruction blind spot of the phase current reconstruction low modulation area; in addition, a reasonable pulse width modulation period is selected to ensure that there is no reconstruction blind spot in the entire linear modulation area; finally, the three-phase current is reconstructed according to the relationship between the coupling current and the phase current under different voltage vectors output by the inverter, forming a complete phase current reconstruction method based on duty cycle model predictive control.

[0128] In an example of the method provided by the above invention, the DC bus voltage Udc is 24V, the stator winding Rs per phase is 0.958Ω, the stator inductance Ls is 4.67mH, the rotor permanent magnet flux ψf is 0.1827Wb, the motor model load is 2N·m, the speed is 80r / min, the switching frequency is 10kHz, and the minimum current sampling time is assumed to be 5us. Using Matlab simulation, the reconstructed current error is about 0.15A, and the error rate is about 8%.

[0129] S500: An incremental model of a stator current prediction value of the permanent magnet synchronous motor at time k+1 is established to obtain a current prediction value of an aligned sampling point, thereby reducing a phase current reconstruction error of the permanent magnet synchronous motor.

[0130] Specifically, the discrete mathematical model of the permanent magnet synchronous motor dq axis is:

[0131]

[0132] In the formula, i dq is the stator current of dq axis, u dq is the stator voltage of dq axis, R S , L S , f are stator resistance, synchronous inductance, permanent magnet flux, ω e is the electrical angular velocity, T s is the pulse width modulation period.

[0133] When the current information i at time k is known dq (k) The predicted current value of T1 at time k is:

[0134]

[0135] The predicted current value of T2 at time k is:

[0136]

[0137] Combining the above two equations, we get

[0138]

[0139] When there are n points between time k and time k+1, the predicted current at time k+1 is predicted to be:

[0140]

[0141] In the formula,

[0142]

[0143]

[0144] Approximate calculation

[0145]

[0146] In the formula, is the average value of the d-axis and q-axis currents at time k, is the average value of the voltage vector applied to the d-axis and q-axis at time k.

[0147] Therefore, the current value prediction model at time k+1 is obtained as follows:

[0148]

[0149] From the above formula, the current value prediction model at time k can be obtained as:

[0150]

[0151] Combining the above two equations, the incremental model for calculating the current prediction value at time k+1 is:

[0152]

[0153] From the above formula, we can know that to calculate the current prediction value at time k+1, we need to get the difference between the initial value of each cycle and the average value of the two cycles before and after. The current slope calculation formula of the permanent magnet synchronous motor is:

[0154]

[0155] In the formula, i s is the stator current, u s is the stator voltage, R S , L S , r They are stator resistance, synchronous inductance, flux linkage, ω r is the electrical angular velocity.

[0156] The initial and average values ​​for each cycle are calculated using Figure 6 For example, the C phase current is sampled at point 1 to obtain the sampled current value iC ,i C is the midpoint current of the zero voltage vector U0 (000), the PWM waveform is symmetrical within the cycle, and the average value of the C phase current in this cycle is the midpoint current of the zero voltage vector U7 (111). The current slope P when each voltage vector acts can be calculated by the above formula c0 and P c1 After finding the slope of each current waveform, the initial value of phase C current in this cycle is calculated by vector action time and current slope:

[0157] i c0 =i C -P c0 T1 / 2

[0158] The average value of phase A current in this cycle is:

[0159] i cv =i C +P c1 T2+P c0 T3 / 2

[0160] In the formula, i cv is the average value of the C phase current in one control cycle.

[0161] Figure 6 The B phase current in the sample point 2 is obtained by sampling current i B ,i B is the midpoint current of the zero voltage vector U7 (111). The PWM waveform is symmetrical within the cycle. The average value of the B-phase current in this cycle is the midpoint current of the zero voltage vector U7 (111), which is the sampling current value i B .

[0162] i bv =i B

[0163] In the formula, i bv is the average value of the B-phase current in one control cycle.

[0164] The current slope P when each voltage vector acts is calculated by the current slope formula b0 and P b1 , the initial value of phase B current in this cycle is:

[0165] i b0 =i B -P b0 T1-P b1 T2-P b0 T3 / 2

[0166] The initial value and average value of each cycle are calculated, and the current is transformed from the abc static three-phase coordinate system to the dq rotating coordinate system through coordinate transformation. Then, the current prediction value i of the aligned sampling point can be obtained by substituting it into the permanent magnet synchronous motor prediction current increment model. a5 and i b5 , reducing the current reconstruction error.

[0167] In the process of phase current reconstruction, this embodiment reduces the amount of calculation caused by reducing the reconstruction error as much as possible, improves the accuracy of phase current reconstruction, and ensures the stability of the system.

[0168] The following is an embodiment of the phase current reconstruction device based on duty cycle model predictive control of the present invention, which can be used to execute the embodiment of the phase current reconstruction method based on duty cycle model predictive control of the present invention. For details not disclosed in the embodiment of the phase current reconstruction device of the present invention, please refer to the embodiment of the phase current reconstruction method of the present invention.

[0169] like Figure 7 As shown, it shows a schematic diagram of the structure of a control device for a phase current reconstruction method based on duty cycle model predictive control provided by an exemplary embodiment of the present invention. The device includes:

[0170] A duty cycle model prediction module is used to realize the construction of a drive system using duty cycle model prediction control as a modulation method;

[0171] A phase current reconstruction module is used to reconstruct the phase current of a drive system constructed with duty cycle model predictive control as a modulation mode by adopting a multi-position coupled sampling method;

[0172] The phase current reconstruction module also includes a reconstruction blind area elimination module;

[0173] A reconstruction blind zone elimination module is used to obtain the reconstruction blind zone when the modulation mode is predicted by the duty cycle model according to the minimum sampling time, and eliminate the reconstruction blind zone;

[0174] The reconstruction blind area includes an undermodulation blind area and an overmodulation blind area.

[0175] Optionally, the duty cycle model prediction module may be used to perform the following method steps: establishing a predicted stator current model of the permanent magnet synchronous motor in a stationary coordinate system;

[0176] The reference current at the next moment and the predicted stator current model are used to obtain the angle of the reference voltage vector, and the optimal voltage vector is selected according to the angle of the reference voltage vector;

[0177] According to the relationship between the reference current vector and the predicted current vector at the next moment, the duty cycle triangle is obtained, and the optimal effective voltage vector action time is calculated with the help of the cosine theorem.

[0178] Furthermore, the optimal effective voltage vector action time is:

[0179]

[0180] In the formula, λ k+1 is a vector value proportional to the reference voltage vector, is the effective voltage vector u4 acting at time k, L S is the synchronous inductance, θ n is the effective voltage vector u4 and λ k+1 The angle between.

[0181] Optionally, the module for eliminating reconstruction blind spots can utilize the cooperation of two zero voltage vectors to eliminate reconstruction blind spots in low modulation areas and partial overmodulation areas; and can select a suitable pulse width modulation period to exclude reconstruction blind spots in overmodulation areas from the linear modulation area.

[0182] Optional, such as Figure 8 As shown, the device also includes: a reconstruction error reduction module, which is used to reduce the permanent magnet synchronous motor phase current reconstruction error by establishing an incremental model of the stator current prediction value of the permanent magnet synchronous motor at time k+1 to obtain the current prediction value of the aligned sampling point.

[0183] Furthermore, the incremental model of the stator current prediction value at time k+1 is:

[0184]

[0185] In the formula, i d 、i q is the stator current of dq axis, u d 、u q is the stator voltage of dq axis, R S , L S , f are stator resistance, synchronous inductance, permanent magnet flux, ω e is the electrical angular velocity, T s is the pulse width modulation period, is the average current value of the dq axis at time k, is the average value of the applied voltage vector in the k time period of the dq axis.

[0186] In the example provided above, the simulation parameters are the same as the previous case, the DC bus voltage Udc is 24V, the stator phase winding Rs is 0.958Ω, the stator inductance Ls is 4.67mH, the rotor permanent magnet flux Ψf is 0.1827Wb, the load is 2N·m, the speed is 80r / min, the switching frequency is 10kHz, and the minimum current sampling time is assumed to be 5us. Using Matlab simulation, the reconstructed current error is about 0.15A, and the error rate is about 8%, such as Fig. 9 shown. Fig.10 This is a current error diagram after the reconstruction error is reduced by the present invention. The reconstructed current error is about 0.06A, and the error rate is about 3%. The effect of reducing the reconstruction error is obvious.

[0187] It should be noted that the phase current reconstruction device based on duty cycle model predictive control provided in the above embodiment only uses the division of the above functional modules as an example when executing the phase current reconstruction method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the phase current reconstruction device based on duty cycle model predictive control provided in the above embodiment and the embodiment of the phase current reconstruction method based on duty cycle model predictive control belong to the same concept. The implementation process thereof is detailed in the embodiment of the phase current reconstruction method based on duty cycle model predictive control, which will not be repeated here.

[0188] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A phase current reconstruction method based on duty cycle model predictive control, characterized in that: The method includes: The drive system is constructed using duty cycle model predictive control as the modulation method; The duty cycle model predictive control comprises: Establish a predictive stator current model for permanent magnet synchronous motor in a stationary coordinate system; The reference current at the next moment and the predicted stator current model are used to obtain the angle of the reference voltage vector, and the optimal voltage vector is selected according to the angle of the reference voltage vector; According to the relationship between the reference current vector and the predicted current vector at the next moment, the vector triangle is obtained, and the optimal effective voltage vector action time is calculated by means of the cosine theorem; When the optimal effective voltage vector selected is U4(011), the optimal effective voltage vector action time is: In the formula, λ k+1 is a vector value proportional to the reference voltage vector, is the effective voltage vector u4 acting at time k, L S is the synchronous inductance, θ n is the effective voltage vector u4 and λ k+1 The angle between The multi-position coupled sampling method is used to reconstruct the system phase current; According to the minimum sampling time, a reconstruction blind area when the modulation mode is predicted by the duty cycle model is obtained; the reconstruction blind area includes a low modulation blind area and an over modulation blind area; Eliminate the reconstruction blind area and realize the phase current reconstruction method based on duty cycle model predictive control.

2. The phase current reconstruction method based on duty cycle model predictive control according to claim 1 is characterized in that: The reconstruction blind area in the low modulation area is the reconstruction blind area caused by the effective voltage vector action time being less than 2 times the minimum sampling time; The reconstruction blind area of ​​the overmodulation region is the reconstruction blind area caused by the zero voltage vector action time being less than 2 times of the minimum sampling time.

3. The phase current reconstruction method based on duty cycle model predictive control according to claim 2 is characterized in that: The step of eliminating the reconstruction blind area includes: By using the cooperation of two zero voltage vectors, the reconstruction blind areas in low modulation areas and some over-modulation areas are eliminated; An appropriate pulse width modulation period is selected to exclude the reconstruction blind area of ​​the overmodulation area from the linear modulation area.

4. The phase current reconstruction method based on duty cycle model predictive control according to claim 3 is characterized in that: The method utilizes the mutual cooperation of two zero vectors to eliminate the reconstruction blind area of ​​the low modulation area and part of the over modulation area; specifically includes: The zero voltage vector of each cycle is completed by the joint action of U0 (000) and U7 (111). The action time of the zero voltage vector U7 (111) is T7> the minimum sampling time T min , the action vector of each cycle consists of 2 zero voltage vectors and 1 effective voltage vector; According to whether the total action time of the zero voltage vector can make the action time T0 of the zero voltage vector U0 (000) and the action time T7 of the zero voltage vector U7 (111) meet the sampling requirements at the same time, the entire linear modulation area is divided into a double zero vector area and a single zero vector area; Select available voltage vectors according to the divided modulation areas: When in the double zero vector region, the two available vectors are U0 (000) and U7 (111); When in the single zero vector region, the two available vectors U7(111), the effective voltage vector U i ; When in the reconstruction blind area, no vectors are available.

5. The phase current reconstruction method based on duty cycle model predictive control according to claim 4 is characterized in that: The selecting of a suitable pulse width modulation period to exclude the reconstruction blind area of ​​the overmodulation area from the linear modulation area specifically includes: Let time T M Greater than the minimum sampling time T min , according to the knowledge of plane geometry, time T M for: T M >T min Where, T M It is the difference between the maximum action time of the space voltage vector in the overmodulation region and the maximum action time in the linear modulation region; Minimum sampling time T min With PWM switching period T s The relationship is: Choose the appropriate PWM switching period T s , phase current reconstruction is performed in the entire linear modulation region.

6. The phase current reconstruction method based on duty cycle model predictive control according to claim 1, characterized in that: The method further includes: establishing an incremental model of the stator current prediction value of the permanent magnet synchronous motor at time k+1 to obtain the current prediction value of the aligned sampling point, thereby reducing the permanent magnet synchronous motor phase current reconstruction error; The incremental model of the stator current prediction value at time k+1 is: In the formula, i d 、i q is the stator current of dq axis, u d 、u q is the stator voltage of dq axis, R S , L S , f are stator resistance, synchronous inductance, permanent magnet flux, ω e is the electrical angular velocity, T s is the pulse width modulation period, is the average current value of the dq axis at time k, is the average value of the applied voltage vector in the k time period of the dq axis.

7. A phase current reconstruction device based on duty cycle model predictive control, characterized in that: The device includes: The duty cycle model prediction module is used to realize the construction of a drive system using the duty cycle model prediction control as a modulation mode; the duty cycle model prediction control includes: Establish a predictive stator current model for permanent magnet synchronous motor in a stationary coordinate system; The reference current at the next moment and the predicted stator current model are used to obtain the angle of the reference voltage vector, and the optimal voltage vector is selected according to the angle of the reference voltage vector; According to the relationship between the reference current vector and the predicted current vector at the next moment, the vector triangle is obtained, and the optimal effective voltage vector action time is calculated by means of the cosine theorem; When the optimal effective voltage vector selected is U4(011), the optimal effective voltage vector action time is: In the formula, λ k+1 is a vector value proportional to the reference voltage vector, is the effective voltage vector u4 acting at time k, L S is the synchronous inductance, θ n is the effective voltage vector u4 and λ k+1 The angle between A phase current reconstruction module is used to reconstruct the phase current of a drive system constructed with duty cycle model predictive control as a modulation mode by adopting a multi-position coupled sampling method; A reconstruction blind zone elimination module is used to obtain the reconstruction blind zone when the modulation mode is predicted by the duty cycle model according to the minimum sampling time, and eliminate the reconstruction blind zone; The reconstruction blind area includes an undermodulation blind area and an overmodulation blind area.

8. The phase current reconstruction device based on duty cycle model predictive control according to claim 7, characterized in that: The device also includes a reconstruction error reduction module, which is used to reduce the permanent magnet synchronous motor phase current reconstruction error by establishing an incremental model for predicting the stator current value of the permanent magnet synchronous motor at time k+1 to obtain the current prediction value of the aligned sampling point.

Images