Position-sensorless commutation optimization method and position-sensorless commutation optimization device

Abstract

The invention discloses a position-sensorless commutation optimization method and device. The method comprises the following steps: S1, calculating a voltage integral between a bus voltage midpoint and a virtual neutral point; and S2, according to the voltage integral, realizing position-sensorless commutation optimization through an online minimum search algorithm. By adopting the technical scheme of the invention, accurate commutation is realized, and the motor efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of brushless direct current motor control, and in particular relates to a position sensorless commutation optimization method and device for a BLDCM (Brushless Direct Current Motor). Background Art

[0002] Brushless DC motors are easy to control and highly efficient, and are widely used in scientific equipment. To control the rotation of the permanent magnet rotor of the motor, two commutation methods are usually used: commutation with position sensors and commutation without position sensors. In the commutation method with position sensors, Hall sensors and photoelectric encoders are usually used. However, installing position sensors not only increases the complexity of the system; mechanical processing and additional signal transmission lines increase costs and reduce reliability. In contrast, sensorless commutation obtains commutation signals by detecting motor signals, which has more advantages. Among sensorless commutation, the commutation method based on back electromotive force is the most widely used, and its commutation point is determined by offsetting the zero crossing point of the back electromotive force by a specific angle. However, due to the commutation error caused by delays in hardware and software, the commutation point is offset, resulting in a decrease in the maximum torque per ampere performance. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a position sensorless commutation optimization method and device.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A position sensorless commutation optimization method, comprising:

[0006] Step S1, calculating the voltage integral between the bus voltage midpoint and the virtual neutral point;

[0007] Step S2: According to the voltage integral, the position sensorless commutation optimization is realized by using an online minimum value search algorithm.

[0008] As a preferred embodiment, in step S1, a differential amplifier circuit is used to calculate the voltage difference between the bus voltage midpoint M and the virtual neutral point H to obtain u MH ,u MH The precision full-wave rectifier is used to obtain u abs , using a first-order low-pass filter to calculate u abs The integral of, that is,

[0009]

[0010] A T It is derived as follows:

[0011]

[0012] As a preferred embodiment, the bus voltage midpoint M is connected to the negative input terminal of the differential amplifier circuit composed of the operational amplifier U1 and the resistor, and the virtual neutral point H is connected to the positive input terminal of the differential amplifier circuit; the output signal u of the differential amplifier circuit MH Connect to the input end of the precision rectifier circuit; the precision rectifier circuit is composed of: u MH Connect to the inverting input of the operational amplifier U2, and the non-inverting input of the operational amplifier U2 is grounded through R6; the output of the operational amplifier U2 is connected to the inverting input of the operational amplifier U3, and the non-inverting input of the operational amplifier U3 is grounded through R10; the operational amplifier U3 outputs the rectified signal u abs Connect the low-pass filter input, the filter output u mea .

[0013] Preferably, in step S2, an online minimum search algorithm without using a model is used to minimize AT, compensate for the commutation error, and complete the closed-loop correction of the commutation error, that is,

[0014]

[0015] Where k is the number of iterations, φ com (k) is the compensation angle, η is the factor that adjusts the compensation increment for each iteration, φ inc (k) is the compensation angle increment.

[0016] Preferably, in step S2,

[0017] When the kth iteration, υ(k) = 1, the search direction is correct; φ rem (k-1)>0, the commutation is delayed; the compensation angle increment is negative, and the commutation error compensation is performed to reduce φ rem ;

[0018] When υ(k)=-1, the search direction is wrong; φ rem (k-1)>0, at this time, the commutation is advanced; at the kth iteration, the compensation angle increment becomes positive, and the commutation error compensation is performed to reduce φ rem ;

[0019] By compensating the error angle φ com , so that A T Reach the minimum value (A Tmin ), with φ rem =0 is the target, and error compensation is performed to achieve precise commutation.

[0020] The present invention also provides a position sensorless commutation optimization device, comprising:

[0021] FLAC circuit, used to calculate the voltage integral between the bus voltage midpoint and the virtual neutral point;

[0022] The commutation optimization circuit is used to realize position sensorless commutation optimization through an online minimum value search algorithm according to voltage integration.

[0023] As a preferred embodiment, the FLAC circuit uses a differential amplifier circuit to calculate the voltage difference between the bus voltage midpoint M and the virtual neutral point H to obtain u MH ,u MH The precision full-wave rectifier is used to obtain u abs , using a first-order low-pass filter to calculate u abs The integral of, that is,

[0024]

[0025] A T It is derived as follows:

[0026]

[0027] As a preferred embodiment, in the FLAC circuit, the bus voltage midpoint M is connected to the negative input terminal of the differential amplifier circuit composed of the operational amplifier U1 and the resistor, and the virtual neutral point H is connected to the positive input terminal of the differential amplifier circuit; the output signal u of the differential amplifier circuit MH Connect to the input end of the precision rectifier circuit; the structure of the precision rectifier circuit is: MH Connect to the inverting input of the operational amplifier U2, and the non-inverting input of the operational amplifier U2 is grounded through R6; the output of the operational amplifier U2 is connected to the inverting input of the operational amplifier U3, and the non-inverting input of the operational amplifier U3 is grounded through R10; the operational amplifier U3 outputs the rectified signal u abs Connect the low-pass filter input, the filter output u mea .

[0028] Preferably, the commutation optimization circuit adopts an online minimum search algorithm (DMC) without using a model to minimize AT, compensate for the commutation error, and complete the closed-loop correction of the commutation error, that is,

[0029]

[0030] Where k is the number of iterations, φ com (k) is the compensation angle, η is the factor that adjusts the compensation increment for each iteration, φ inc (k) is the compensation angle increment;

[0031] When the kth iteration, υ(k) = 1, the search direction is correct; φ rem (k-1)>0, the commutation is delayed; the compensation angle increment is negative, and the commutation error compensation is performed to reduce φ rem ;

[0032] When υ(k)=-1, the search direction is wrong; φ rem (k-1)>0, at this time, the commutation is advanced; at the kth iteration, the compensation angle increment becomes positive, and the commutation error compensation is performed to reduce φ rem ;

[0033] By compensating the error angle φ com , so that A T Reach the minimum value (A Tmin ), with φ rem =0 is the target, and error compensation is performed to achieve precise commutation.

[0034] The present invention first analyzes the voltage between the bus midpoint and the virtual neutral point, and proposes that under the condition of accurate phase switching, u MH The integral after rectification is the minimum value. By using FLAC and DMC, the commutation error is eliminated, making the rectified u MH The integral reaches the minimum value. Compared with the closed-loop compensation method of digital characterization, the present invention will not limit the compensation accuracy due to the sampling rate, so that the compensation accuracy of the motor at high speed is guaranteed. At the same time, compared with the traditional hardware integral commutation optimization method, the present invention does not need to use complex timing logic to determine the integral interval, and can obtain the commutation error characterization, and realizes accurate commutation through the minimum value search algorithm, and constructs a closed-loop error correction system. Thereby reducing power consumption, achieving high-efficiency operation of the motor, and improving reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0036] Figure 1 This is a flow chart of a position sensorless commutation optimization method according to an embodiment of the present invention;

[0037] Figure 2 The topological structure diagram of the brushless DC motor drive circuit based on Buck converter;

[0038] Figure 3 is the value u under different commutation error conditions MH and u abs The waveform of the phase shift lag (φ err >0), (b) is the commutation lead (φ err <0), (c) is the commutation accuracy (φ err =0);

[0039] Figure 4 is the induced flux component of the three-phase winding in [θ0,2π / 3] when there is a commutation error; where (a) is the change in the induced flux of phase B; (b) is the change in the induced flux of phase C; (c) is the change in the induced flux of phase A

[0040] Figure 5 When there is a commutation error, the three-phase winding induced flux components within [2π / 3,θ0+π / 3]; where (a) is the change in the induced flux of phase B; (b) is the change in the induced flux of phase C; (c) is the change in the induced flux of phase A;

[0041] Figure 6 This is a schematic diagram of the structure of FLAC;

[0042] Figure 7 is the convergence situation; where (a) is the convergence situation when υ(k) = 1 at the kth iteration; (b) is the convergence situation when υ(k) = -1 at the kth iteration. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the 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.

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Embodiment 1:

[0046] like Figure 1 As shown, an embodiment of the present invention provides a position sensorless commutation optimization method, comprising:

[0047] Step S1, calculating the voltage integral between the bus voltage midpoint and the virtual neutral point;

[0048] Step S2: According to the voltage integral, the position sensorless commutation optimization is realized by using an online minimum value search algorithm.

[0049] As an implementation method of the present invention, in step S1, Figure 2 As shown, a Buck converter is used for DC voltage regulation to achieve efficient control. PThe two ends are connected to a 28V DC power supply, and the Buck circuit composed of MOS tube, inductor, diode and capacitor is used to control the duty cycle of the input signal of T7 to achieve voltage control; the positive output end of the Buck circuit is connected to the source of T1, T3 and T5, and the negative output end is connected to the drain of T4, T6 and T2; the three-phase windings of the motor A, B and C are respectively connected to the drain-source of T1, T4, T3, T6, T5 and T2 to form a three-phase inverter circuit; through two resistors R M Construct the bus voltage midpoint M, connect the three-phase winding of the motor through three resistors to construct a virtual neutral point, and the voltage difference between the two is u MH The three-phase winding of the motor is connected in star shape, N is the neutral point of the three-phase winding, U is the DC is the DC bus voltage; u MH The expression is:

[0050]

[0051] Wherein, X represents the back electromotive force of the non-conducting phase, and Y and Z represent the back electromotive force of the conducting phase, respectively.

[0052] In an embodiment of the present invention, the transformation is defined as:

[0053] u abs =|u MH | (0.2)

[0054] like Figure 3 As shown, assuming that phase A is not conducting in the range [θ0,θ0+π / 3], then u abs The area enclosed by the x-axis can be expressed as:

[0055] A T =A L +A R (0.3)

[0056] when It can be seen that under different commutation conditions, A T Different values: delayed commutation error (φ err >0, Figure 3 (a)), leading commutation error (φ err <0, Figure 3 (b)) and accurate commutation (φ err =0, Figure 3 (c)).

[0057] The relationship between the back electromotive force (e) and the magnetic flux (ψ) generated by the rotating rotor magnetic field can be expressed as A T can be rewritten as:

[0058]

[0059] Among them, A_L ,Ψ B_L and C_L represents the change of three-phase ψ in the interval [θ0,2π / 3]; Ψ A_L ,Ψ B_L and C_L Represents the change of three-phase ψ in the interval [2π / 3,θ0+π / 3].

[0060] Figure 4 (a), (b), (c) and Figure 5 (a), (b), and (c) show the changes of the motor rotor magnetic potential vector in the angle range of [θ0, 2π / 3] and [2π / 3, θ0+π / 3] respectively under the condition of commutation lag. According to the geometric position relationship between the stator coil and the rotor magnetic field vector, the change of the magnetic flux induced by the three-phase winding can be expressed as:

[0061]

[0062] Among them, F rot is the magnetomotive force of the rotor, R mag Represents the magnetic resistance of each phase.

[0063] According to formulas (1.4), (1.5) and (1.6), A T can be simplified to:

[0064]

[0065] Similarly, for the case of phase leading, the same A can be obtained. T According to formula (1.7), when A T When the minimum value is reached The commutation error will be eliminated (φ err =0).

[0066] According to the conclusion, optimizing the commutation process needs to overcome two challenges. T and a self-optimizing algorithm to compensate for φ err , the purpose is to make A T Close to A Tmin .

[0067] like Figure 6 As shown, a hardware-based FLAC is designed to obtain A T The bus voltage midpoint M is connected to the negative input terminal of the differential amplifier circuit composed of the operational amplifier U1 and the resistor, and the virtual neutral point H is connected to the positive input terminal of the differential amplifier circuit; the output signal u MH Connect to the input of precision rectifier circuit; rectifier circuit output uabs Connect the low-pass filter input, the filter output u mea Specifically, the in-phase input terminal of the operational amplifier U1 is connected to the virtual neutral point H through the resistor R2 and to the ground through the resistor R3; the inverting input terminal of U1 is connected to the bus voltage midpoint M through the resistor R1 and to the output terminal of U1 through the resistor R4. The output signal u of U1 MH The inverting input of the operational amplifier U2 is connected through resistor R5; the non-inverting input of the operational amplifier U2 is connected to the ground through resistor R6; the output of the operational amplifier U2 is connected to the cathode of the diode D2; then the positive output signal of D2 is input to the inverting input of U2 through D1 and R7, and connected to the inverting input of U3 through R8. 10 Connect to ground; U3 inverting input terminal, connected to U1 output terminal through R9, 11 Connect to the output terminal of U3. The output signal of U3 is u abs By R 12 Input to the circuit output end, the circuit output end is connected to the ground through capacitor C to output signal u mea .

[0068] First, use the differential amplifier circuit to calculate the voltage difference between points M and H to get u MH ; Then, u MH After being processed by a precision full-wave rectifier, we get u abs ; Finally, a first-order low-pass filter (LPF) is used to calculate u abs The integral of this voltage is:

[0069]

[0070] A T It can be deduced as:

[0071]

[0072] Therefore, A T You can use u mea It is obtained by multiplying with π / 3. Compared with the traditional method, it not only avoids the integration operation of high sampling rate in software, but also realizes the integration based on hardware, and does not need to use complex timing logic to determine the integration interval.

[0073] As an implementation method of the present invention, it can be obtained from formula (1.7) that φ err With A T There is a nonlinear relationship between A and B, and the exact model of this relationship is difficult to obtain. Therefore, a model-free online minimum search algorithm (DMC) is proposed to make A TReach the minimum value, compensate for the commutation error, and complete the closed-loop correction of the commutation error. The algorithm is as follows:

[0074]

[0075] Where k is the number of iterations, φ com (k) is the compensation angle, η is the factor that adjusts the compensation increment for each iteration, φ inc (k) is the compensation angle increment.

[0076] Compensation angle φ com After (k-1), the remaining commutation error φ at the kth iteration is rem It can be expressed as:

[0077]

[0078] According to (1.11), υ(k) can be rewritten as:

[0079]

[0080] from Figure 6 It can be seen that the sign of the commutation error can be determined by A T The derivative υ(k) is used to represent it; therefore, the search process of the algorithm can be divided into two cases:

[0081] 1. As Figure 7 As shown in (a), when the kth iteration, υ(k) = 1, the search direction is correct; φ rem (k-1)>0, the commutation is delayed. Therefore, the compensation angle increment is negative (φ inc (k)<0), commutation error compensation is performed to reduce φ rem .

[0082] 2. If Figure 7 As shown in (b), when υ(k) = -1, the search direction is wrong; at this time φ rem (k-1)>0, the commutation is advanced. Therefore, at the kth iteration, the compensation angle increment becomes positive (φ inc (k)>0), commutation error compensation is performed to reduce φ rem .

[0083] By compensating the error angle φ com , so that A T Reach the minimum value (A Tmin ). In other words, rem =0 is the target, and error compensation is performed to achieve precise commutation.

[0084] Embodiment 2:

[0085] The embodiment of the present invention further provides a position sensorless commutation optimization device, comprising:

[0086] FLAC circuit, used to calculate the voltage integral between the bus voltage midpoint and the virtual neutral point;

[0087] The commutation optimization circuit is used to realize position sensorless commutation optimization through an online minimum value search algorithm according to voltage integration.

[0088] As an implementation method of the present invention, the FLAC circuit uses a differential amplifier circuit to calculate the voltage difference between the bus voltage midpoint M and the virtual neutral point H to obtain u MH ,u MH The precision full-wave rectifier is used to obtain u abs , using a first-order low-pass filter to calculate u abs The integral of, that is,

[0089]

[0090] A T It is derived as follows:

[0091]

[0092] As an implementation method of an embodiment of the present invention, in the FLAC circuit, the bus voltage midpoint M is connected to the negative input terminal of the differential amplifier circuit composed of an operational amplifier U1 and a resistor, and the virtual neutral point H is connected to the positive input terminal of the differential amplifier circuit; the output signal u of the differential amplifier circuit MH Connect to the input end of the precision rectifier circuit; the structure of the precision rectifier circuit is, u MH Connect the inverting input of the operational amplifier U2, and the non-inverting input of the operational amplifier U2 is grounded through R6; the output of the operational amplifier U2 is connected to the inverting input of the operational amplifier U3, and the non-inverting input of the operational amplifier U3 is grounded through R10; the output of the operational amplifier U3 is u abs Connect the low-pass filter input, the filter output u mea .

[0093] As an implementation method of an embodiment of the present invention, the commutation optimization circuit adopts an online minimum search algorithm without using a model to minimize AT, compensate for the commutation error, and complete the closed-loop correction of the commutation error, that is,

[0094]

[0095] Where k is the number of iterations, φ com (k) is the compensation angle, η is the factor that adjusts the compensation increment for each iteration, φ inc (k) is the compensation angle increment;

[0096] When the kth iteration, υ(k) = 1, the search direction is correct; φ rem (k-1)>0, the commutation is delayed; the compensation angle increment is negative, and the commutation error compensation is performed to reduce φ rem ;

[0097] When υ(k)=-1, the search direction is wrong; φ rem (k-1)>0, at this time, the commutation is advanced; at the kth iteration, the compensation angle increment becomes positive, and the commutation error compensation is performed to reduce φ rem ;

[0098] By compensating the error angle φ com , so that A T Reach the minimum value (A Tmin ), with φ rem =0 is the target, and error compensation is performed to achieve precise commutation.

[0099] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A position sensorless commutation optimization method, characterized in that: include: Step S1, calculating the voltage integral between the bus voltage midpoint and the virtual neutral point; Step S2: According to the voltage integral, the position sensorless commutation optimization is realized by using an online minimum value search algorithm.

2. The position sensorless commutation optimization method according to claim 1, characterized in that: In step S1, a differential amplifier circuit is used to calculate the voltage difference between the bus voltage midpoint M and the virtual neutral point H to obtain u MH ,u MH The precision full-wave rectifier is used to obtain u abs , using a first-order low-pass filter to calculate u abs The integral of, that is, A T It is derived as follows:

3. The position sensorless commutation optimization method according to claim 2, characterized in that: The bus voltage midpoint M is connected to the negative input terminal of the differential amplifier circuit composed of the operational amplifier U1 and the resistor, and the virtual neutral point H is connected to the positive input terminal of the differential amplifier circuit; the output signal u MH Connect to the input end of the precision rectifier circuit; the structure of the precision rectifier circuit is: MH Connect to the inverting input of the operational amplifier U2, the non-inverting input of the operational amplifier U2 is grounded through R6, the output of the operational amplifier U2 is connected to the inverting input of the operational amplifier U3, and the non-inverting input of the operational amplifier U3 is grounded through R10; the operational amplifier U3 outputs the rectified signal u abs Connect the low-pass filter input, the filter output u mea .

4. The method for optimizing commutation without position sensor according to claim 3, characterized in that: In step S2, an online minimum search algorithm without using a model is used to minimize AT, compensate for the commutation error, and complete the closed-loop correction of the commutation error, that is, Where k is the number of iterations, φ com (k) is the compensation angle, η is the factor that adjusts the compensation increment for each iteration, φ inc (k) is the compensation angle increment.

5. The method for optimizing commutation without position sensor according to claim 3, characterized in that: In step S2, When the kth iteration, υ(k) = 1, the search direction is correct; φ rem (k-1)>0, then the commutation is delayed; The compensation angle increment is negative, and the commutation error compensation is performed to reduce φ rem ; When υ(k)=-1, the search direction is wrong; φ rem (k-1)>0, at this time, the commutation is advanced; at the kth iteration, the compensation angle increment becomes positive, and the commutation error compensation is performed to reduce φ rem ; By compensating the error angle φ com , so that A T Reach the minimum value (A Tmin ), with φ rem =0 is the target, and error compensation is performed to achieve precise commutation.

6. A position sensorless commutation optimization device, characterized in that: include: FLAC circuit, used to calculate the voltage integral between the bus voltage midpoint and the virtual neutral point; The commutation optimization circuit is used to realize position sensorless commutation optimization through an online minimum value search algorithm according to voltage integration.

7. The position sensorless commutation optimization device according to claim 6, characterized in that: The FLAC circuit uses a differential amplifier circuit to calculate the voltage difference between the bus voltage midpoint M and the virtual neutral point H to obtain u MH ,u MH The precision full-wave rectifier is used to obtain u abs , using a first-order low-pass filter to calculate u abs The integral of, that is, A T It is derived as follows:

8. The position sensorless commutation optimization device according to claim 7, characterized in that: In the FLAC circuit, the bus voltage midpoint M is connected to the negative input terminal of the differential amplifier circuit composed of the operational amplifier U1 and the resistor, and the virtual neutral point H is connected to the positive input terminal of the differential amplifier circuit; the output signal u MH Connect to the input end of the precision rectifier circuit; the structure of the precision rectifier circuit is: MH Connect to the inverting input of the operational amplifier U2, the non-inverting input of the operational amplifier U2 is grounded through R6, the output of the operational amplifier U2 is connected to the inverting input of the operational amplifier U3, and the non-inverting input of the operational amplifier U3 is grounded through R10; the operational amplifier U3 outputs the rectified signal u abs Connect the low-pass filter input, the filter output u mea .

9. The position sensorless commutation optimization device according to claim 8, characterized in that: The commutation optimization circuit adopts an online minimum search algorithm without using a model to minimize AT, compensate for the commutation error, and complete the closed-loop correction of the commutation error, that is, Where k is the number of iterations, φ com (k) is the compensation angle, η is the factor that adjusts the compensation increment for each iteration, φ inc (k) is the compensation angle increment; When the kth iteration, υ(k) = 1, the search direction is correct; φ rem (k-1)>0, the commutation is delayed; the compensation angle increment is negative, and the commutation error compensation is performed to reduce φ rem ; When υ(k)=-1, the search direction is wrong; φ rem (k-1)>0, at this time, the commutation is advanced; at the kth iteration, the compensation angle increment becomes positive, and the commutation error compensation is performed to reduce φ rem ; By compensating the error angle φ com , so that A T Reach the minimum value (A Tmin ), with φ rem =0 is the target, and error compensation is performed to achieve precise commutation.

Images