Optimal commutation error compensation method for high-speed brushless DC motor without position sensor
Through improved co-frequency filter and segmented iterative compensation algorithm, the problem of phase commutation error in high-speed brushless DC motors is solved, and the compensation for impedance characteristic delay and short-time clamp voltage pulses is realized, which improves the rotation of the motor Moment and current ratio and operating efficiency.
Patent Information
- Application Number
- CN202411170421.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-26
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-26
AI Technical Summary
When the high-speed brushless DC motor is running without a position sensor, the commutation error caused by low-pass filter delay, software and hardware delay and impedance characteristic delay are difficult to effectively compensate, which affects the torque-current ratio and operating efficiency of the motor. .
Using an improved syndiotically modified filter and segmented iterative compensation algorithm, the phase correction is used to shape the opposite potential, eliminate short-time clamp voltage pulses, extract the fundamental signal of the opposite potential and phase current, and use the exclusive OR logic to process the phase. The difference is achieved to achieve phase synchronization, thereby determining the optimal compensation amount for commutation error compensation.
Accurate compensation for various commutation errors including impedance characteristic delay and short-time clamp voltage pulses is achieved, which improves the torque-current ratio and operating efficiency of the motor and ensures the optimal commutation of the motor.
Smart Images

Figure CN118677301B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a position sensorless commutation error compensation method, and in particular to a position sensorless optimal commutation error compensation method for a high-speed brushless DC motor. Background Art
[0002] In many industrial applications such as blowers, vacuum pumps and centrifugal compressors, high-speed brushless DC motors have become the main drive type due to their high power density, small size and light weight. However, when high-speed brushless DC motors are operated without mechanical position sensors, factors such as low-pass filter delay, software and hardware delay, and impedance characteristic delay will cause large commutation errors. Research on commutation error compensation of high-speed brushless DC motors is of great significance to improving the torque-to-current ratio and operating efficiency of the motors.
[0003] At present, the commutation error compensation methods mainly include open-loop compensation and closed-loop compensation. Since some commutation errors depend on motor parameters and are difficult to accurately model, the open-loop compensation method cannot accurately compensate for various commutation errors. The closed-loop compensation method can convert various commutation errors into a feedback quantity, and these commutation errors can be accurately compensated by making this feedback quantity converge to zero. At present, the closed-loop compensation methods for commutation errors are mainly divided into two categories, namely, the compensation method based on the symmetry of electrical signals and the compensation method based on the synchronization of opposite potential and phase current phase.
[0004] The first type of method mainly uses the symmetry of electrical signals such as the three-phase voltage or current of the motor before and after the commutation point to obtain the compensation amount. The non-conduction opposite potential, the conduction opposite potential, the voltage from the conduction phase to the midpoint of the DC side, the DC bus current or the non-commutation phase current can all be used to obtain the compensation amount. In order to prevent the sampling signal from having a phase delay, the above methods do not use a low-pass filter to filter the signal. In order to avoid the influence of high-frequency noise on double-point sampling, some scholars also use the integral difference of these electrical signals to obtain the compensation amount. For example, the integral difference of the non-conduction opposite potential between adjacent conduction intervals, the integral difference of the virtual third harmonic back potential at an electrical angle of 30° before and after the zero point, and the integral difference of the current in the two 30° half-cycles before and after the conduction period can all be used as feedback to compensate for the commutation error.
[0005] The compensation method based on the symmetry of electrical signals can compensate for the commutation error caused by the low-pass filter and the software and hardware delays by converging the feedback containing the commutation error information to zero. However, since it is essentially based on the symmetry of opposite potentials, that is, the control target is that the commutation point is located at the intersection of opposite potentials or the zero crossing point with a delay of 30° electrical angle, the influence of the impedance characteristic delay is not considered. Therefore, there is still room for improvement in the torque-to-current ratio and operating efficiency of the motor.
[0006] The phase difference between the opposite potential and the fundamental signal of the phase current is called the internal power factor angle. The basic idea of the compensation method based on the synchronization of the opposite potential and the phase current phase is to extract the internal power factor angle of the motor in real time, and to compensate for various commutation errors including impedance characteristic delay by converging the internal power factor angle to zero. Some scholars obtain the internal power factor angle by detecting the phase difference between the flux linkage and the phase current integral, and propose a structure of a second-order high-pass filter plus a low-pass filter to suppress the DC offset introduced by the pure integrator. In order to avoid the use of multiple filters, the same frequency filter is used to extract the fundamental signal of the flux linkage and the phase current integral, and then the internal power factor angle is obtained and compensated through the XOR logic or the phase-locked loop. In addition, the second-order generalized integrator is also used to extract the fundamental signal of the flux linkage and the phase current integral. In addition to the XOR logic and the phase-locked loop, the phase difference of the fundamental signal can also be obtained by trigonometric function transformation, thereby realizing the construction of the commutation error.
[0007] However, although the above compensation method based on the synchronization of the opposite potential and the phase current takes into account the influence of the impedance characteristic delay, it does not take into account the short-term clamping voltage pulse caused by the diode freewheeling, which will cause the extracted opposite potential fundamental signal to lead the ideal sinusoidal back potential in phase, ultimately leading to advanced commutation. Summary of the invention
[0008] In order to solve the problems existing in the background technology, the present invention provides a high-speed brushless DC motor position sensorless optimal commutation error compensation method.
[0009] The technical solution adopted by the present invention is:
[0010] The high-speed brushless DC motor position sensor-free optimal commutation error compensation method of the present invention comprises:
[0011] Step 1) The high-speed brushless DC motor is driven by a quasi-current source inverter and position sensorless control is performed. The phase consistency is determined by comparing the phase of the fundamental signal of the reverse potential containing a short-term clamping voltage pulse of the high-speed brushless DC motor with the phase of the ideal sinusoidal back potential.
[0012] Step 2) When the phases of the high-speed brushless DC motor are inconsistent, the reverse potential of the high-speed brushless DC motor in the commutation phase is shaped to obtain a shaped reverse potential to eliminate a short-time clamping voltage pulse.
[0013] Step 3) An improved synchronous frequency filter (ISFF) based on phase correction is established, and the phase current and the shaped reverse potential of the high-speed brushless DC motor are input into the improved synchronous frequency filter for processing. After processing, the improved synchronous frequency filter extracts the fundamental wave signals of the shaped reverse potential and phase current.
[0014] Step 4) A piecewise iterative compensation algorithm is used to converge the phase difference between the fundamental signal of the shaped opposite potential and the phase current, and the optimal compensation amount is determined at the same time. The commutation error of the position sensorless is compensated according to the optimal compensation amount, thereby realizing the optimal commutation error compensation of the high-speed brushless DC motor under position sensorless control.
[0015] In the step 1), the reverse potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse is subjected to Fourier decomposition to obtain a fundamental wave signal of the reverse potential of the short-time clamping voltage pulse, as follows:
[0016] In the step 1), the reverse potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse is subjected to Fourier decomposition to obtain a fundamental wave signal of the reverse potential of the short-time clamping voltage pulse, as follows:
[0017] e xs ( i )= A 1 cos( i - i e / 2+ α )+ A 2 cos( i - i e / 2)=( A 1 2 + A 2 2 +2 A 1 A 2 cos α ) 1 / 2 cos( I )
[0018] I = i - i e / 2+ c
[0019] c =tan -1 (( A 1 sin α ) / ( A 1 cos α + A 2 ))
[0020] A 1 =(12 n / π)(sin i e / 2)
[0021] A 2 =( E / 2π)(-6 i e +2π)
[0022] α =π / 3- i d
[0023] in, e xs ( )and I They are respectively the fundamental wave signal of the x-phase reverse potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse and its phase. The x-phase of the high-speed brushless DC motor is one of the a-phase, b-phase and c-phase. i is the rotor angle of the high-speed brushless DC motor; A 1 , A 2 and α are respectively the first, second and third intermediate variables of the fundamental wave signal of the x-opposite potential containing the short-time clamping voltage pulse of the high-speed brushless DC motor; i e is the commutation phase duration angle of the high-speed brushless DC motor; c is the lead angle caused by the short-term clamping voltage pulse; n The first intermediate variable of the fundamental signal of the x reverse potential containing a short-term clamping voltage pulse of the high-speed brushless DC motor A 1 The intermediate variable n = U d / 3, U d is the power supply voltage; E is the opposite potential amplitude; i d is the commutation error without position sensor.
[0024] Thus, the phase of the fundamental wave signal containing the x-phase reverse potential of the short-time clamping voltage pulse is extracted from the fundamental wave signal containing the x-phase reverse potential of the short-time clamping voltage pulse, and the phase of the fundamental wave signal containing the x-phase reverse potential of the short-time clamping voltage pulse is compared with the phase of the x-phase ideal sinusoidal reverse potential to obtain the leading angle caused by the short-time clamping voltage pulse. c , when the lead angle caused by the short clamping voltage pulse c When it is zero, it is judged that the phase of the high-speed brushless DC motor is consistent. When the leading angle caused by the short-term clamping voltage pulse c When it is not zero, it is judged that the phase of the high-speed brushless DC motor is inconsistent.
[0025] The short-term clamping voltage pulse causes the opposite potential fundamental signal to phase-lead the ideal sinusoidal back EMF ( c >0) is: 0< i e <π / 3,-π / 6< i d <π / 3. Before obtaining the phase of the fundamental signal of the opposite potential containing a short-time clamping voltage pulse of the high-speed brushless DC motor, the discontinuity of the opposite potential caused by the diode freewheeling of the quasi-current source inverter driven by the quasi-current source inverter is first analyzed, and then the phase of the fundamental signal of the opposite potential containing a short-time clamping voltage pulse is analyzed and quantitatively calculated.
[0026] In the step 2), the reverse potential of the high-speed brushless DC motor in the commutation phase is shaped to obtain the shaped reverse potential as follows:
[0027] e x re =- u z / 3, S inf =0
[0028] e x re =0, S inf =1
[0029] in, e x re is the x-phase phase potential after shaping of the high-speed brushless DC motor, and the x-phase of the high-speed brushless DC motor is one of the a-phase, the b-phase and the c-phase; u z is the line voltage difference of the high-speed brushless DC motor. When x=a, z=ca-ab, u ca-ab is the difference between the line voltage of phase c and phase a of the high-speed brushless DC motor and the line voltage of phase a and phase b. When x=b, z=ab-bc, u ab-bc is the difference between the line voltage of phase a and phase b of the high-speed brushless DC motor and the line voltage of phase b and phase c. When x=c, z=bc-ca,u bc-ca It is the difference between the line voltage of phase b and phase c of the high-speed brushless DC motor and the line voltage of phase c and phase a; S inf It is the first commutation mark. When the high-speed brushless DC motor is in the commutation stage, the first commutation mark S inf =1, when the high-speed brushless DC motor is in the non-commutation stage, the first commutation flag S inf =0.
[0030] Taking the a phase of the high-speed brushless DC motor as an example, the fundamental wave signal of the a phase potential after shaping is as follows:
[0031] e as re ( i )=( E / 2π)(-6 i e +2π)cos( i - i e / 2)
[0032] It can be seen that the phase of the fundamental wave signal of the shaped back electromotive force of phase a is consistent with the ideal sinusoidal back electromotive force of phase a.
[0033] Whether the high-speed brushless DC motor is in the commutation stage is determined by detecting whether the off-phase current of the high-speed brushless DC motor decays to zero, that is, voltage pulse detection is performed according to the off-phase current of the high-speed brushless DC motor, so as to determine whether the high-speed brushless DC motor is in the commutation stage according to the commutation mark, and then the opposite potential in the commutation stage is set to zero to eliminate the short-term clamping voltage pulse.
[0034] In the step 3), the improved same-frequency filter based on phase correction is as follows:
[0035] X out = e sin( P 2 ( t ))∫ X 1 ( t )d t + e cos( P 2 ( t ))∫ X 2 ( t )d t
[0036] ∫ X 1 ( t )d t =∫ X in sin( P 2 ( t ))d t
[0037] ∫ X 2 ( t )d t =∫ X in cos( P 2 ( t ))d t
[0038] P 2 ( t )= w 0 ´ t -Δ P ( t )
[0039] P 1 ( t )= w 0 t
[0040] Δ P ( t )=( w 0 ´- w 0 ) kT s
[0041] in, X in and X out They are the input signal and output signal of the improved same-frequency filter, respectively. The input signal is the x-phase current of the high-speed brushless DC motor. i x Or the x-phase potential after shaping e x re , the output signal is the x-phase current of the high-speed brushless DC motor i x The fundamental signal i xs Or the x-phase potential after shapinge x re The fundamental signal e xs re , the x phase of the high-speed brushless DC motor is one of the a phase, b phase and c phase; e is the gain of the improved same-frequency filter; t For time; P 1 ( t )and P 2 ( t ) are respectively t The reference signal phase before and after the frequency change of the time-modified same-frequency filter; X 1 ( t )and X 2 ( t ) are respectively t The first and second intermediate variables of the time-modified same-frequency filter; w 0 and w 0 'Respectively t The value before and after the frequency change of the reference signal of the improved same-frequency filter when the speed changes or the speed fluctuates; Δ P ( t )for t When the speed changes or the speed fluctuates, the phase of the reference signal of the traditional frequency filter changes suddenly. t = kT s , k is the sampling number, T s To control the cycle.
[0042] Assume that the input signal of the same frequency filter contains high-frequency harmonic components, as follows:
[0043] X in = V 0 sin( w 0 t + f 0 )+∑ V h sin( w h t + f h )
[0044] in,V 0 , w 0 and f 0 The input signals are X in The amplitude, frequency and phase of the fundamental component; V h , w h and f h The input signals are X in middle h Amplitude, frequency and phase of subharmonic components.
[0045] By using the orthogonality of trigonometric functions and performing integral operations, the signal can be obtained as follows:
[0046] ∫ X 1 ( t )d t =∫ X in sin( w 0 t )d t =( V 0 t cos f 0 ) / 2
[0047] ∫ X 2 ( t )d t =∫ X in cos( w 0 t )d t =( V 0 t sin f 0 ) / 2
[0048] Among them, sin( w 0 t ) and cos( w 0 t ) is the reference signal of the traditional frequency filter, and its frequency is obtained from the speed of the high-speed brushless DC motor.
[0049] The output signal of the traditional same-frequency filter is as follows:
[0050] X out = eV 0 t sin( w 0 t + f 0 ) / 2
[0051] Assumptions t = kT s When the speed changes or the speed fluctuates, the reference signal frequency of the traditional frequency filter is given by w 0 becomes w 0 ´, at this time, there is a phase mutation of the reference signal as follows:
[0052] Δ P ( t )=( w 0 ´- w 0 ) kT s
[0053] To ensure the phase continuity of the reference signal, the phase of the reference signal before and after the frequency change is as follows:
[0054] P 1 ( t )= w 0 t , t ∈[0, kT s )
[0055] P 2 ( t )= w 0 kT s + w 0 ´( t - kT s ), t ∈[ kT s ,∞)
[0056] The improved same-frequency filter corrects the phase when the reference signal frequency changes, and the phase of the reference signal before and after the frequency change after the phase correction can be obtained as follows:
[0057] P 1 ( t )= w 0 t
[0058] P 2 ( t )= w 0 ´ t -Δ P ( t )
[0059] After phase correction, the relationship between the reference signal phase before and after the frequency change is as follows:
[0060] P 2 ( t )= P 1 ( t )+ w 0 ´( t - kT s )
[0061] The improved same-frequency filter can ensure the phase continuity of the reference signal by correcting the phase when the reference signal frequency changes, so that the fundamental signal can be effectively extracted when the speed changes or the speed fluctuates.
[0062] In the step 4), the shaped reverse potential and phase current are processed by XOR logic and according to the leading and lagging commutation states of the high-speed brushless DC motor, the phase difference of the fundamental wave signal of the shaped reverse potential and phase current is obtained as follows:
[0063] i err =(-1) r m Δ i err
[0064] Δ i err =2π pn m T s / 60
[0065] in, i err is the phase difference between the fundamental wave signal of the shaped opposite potential and the phase current; rIt is the second commutation mark. When the high-speed brushless DC motor is in the advanced commutation state, the second commutation mark r =1, when the high-speed brushless DC motor is in the delayed commutation state, the second commutation flag r =0; m is the counting value when the fundamental wave signal of the shaped opposite potential and phase current has different signs, and the counting frequency is the control frequency of the high-speed brushless DC motor; Δ i err is the phase difference quantization unit; p is the number of pole pairs of the high-speed brushless DC motor; n m is the speed of the high-speed brushless DC motor; T s To control the cycle.
[0066] Phase difference between the fundamental wave signal of the phase current and the reverse potential after shaping i err It is zero upon convergence.
[0067] The fundamental wave signal extracted under digital control is not a continuous sine wave, but a sine wave that changes in steps with the control cycle as the unit. At this time, the phase difference between the phase current calculated by the exclusive OR logic and the shaped reverse potential fundamental wave signal is as follows:
[0068] i err = m Δ i err
[0069] The phase difference is calculated whenever the phase current and the shaped counter-potential fundamental signal have opposite signs, so the phase difference is updated every half electrical cycle of the motor operation.
[0070] The commutation state of the motor can be determined by comparing the order in which the phase current fundamental signal and the shaped reverse electromotive force fundamental signal cross zero, that is, the sign of the phase difference can be determined.
[0071] In the step 4), the optimal compensation amount is determined by using a segmented iterative compensation algorithm, and the segmented iterative compensation process is divided into a coarse adjustment stage and a fine adjustment stage. In the process of convergence of the phase difference between the shaped opposite potential and the fundamental wave signal of the phase current, the coarse adjustment stage compensation amount is obtained in the coarse adjustment stage according to the phase difference between the shaped opposite potential and the fundamental wave signal of the phase current in each iteration, and the fine adjustment stage compensation amount is obtained in the fine adjustment stage. Finally, the total compensation amount is obtained according to the coarse adjustment stage compensation amount and the fine adjustment stage compensation amount as follows:
[0072] i com = i 1 + i 2
[0073] in, i com is the total compensation amount; i 1 and i 2 They are the compensation amount in the coarse adjustment stage and the compensation amount in the fine adjustment stage respectively.
[0074] Until the phase difference between the fundamental wave signal of the shaped reverse potential and the phase current converges to zero, the total compensation amount at this time is obtained. i com As the optimal compensation amount, the sensorless commutation error is compensated. i err It will become larger with the increase of speed. For high-speed working conditions, in order to achieve more precise compensation, the compensation process is divided into two steps: coarse adjustment and fine adjustment.
[0075] The coarse adjustment stage compensation i 1 as follows:
[0076] i 1 = k iter1 ∑ N1=1 N1max ( i err ( i ) / N 1), m >1
[0077] in, k iter1 is the iteration coefficient of the coarse adjustment stage; N 1 and N 1 max are the number of iterations and the maximum number of iterations in the coarse adjustment stage respectively; i err ( i ) is the i The phase difference between the fundamental signal of the opposite potential and the phase current after the shaping of the iteration; m It is the count value when the fundamental wave signal of the shaped counter potential and the phase current has opposite signs.
[0078] Coarse adjustment stage: m >1, the phase difference is considered to be large. In order to make it converge quickly and avoid excessive accumulation of compensation and thus cross the optimal commutation point, an attenuation factor 1 / is set which decreases with the increase of the number of iterations. N 1 Make compensation.
[0079] The compensation amount in the fine adjustment stage is i 2 as follows:
[0080] i 2 = k iter2 ∑ N2=1 N2max ( i avg ( N 2)), m ≤1
[0081] i avg ( N 2)=(∑ i=1+Y(N2-1) YN2 ( i err ( i ))) / Y
[0082] in, k iter2 is the iteration coefficient of the fine-tuning stage; N 2 and N 2 max are the number of iterations and the maximum number of iterations in the fine-tuning phase, respectively; i avg ( ) is continuous Y The average value of the phase difference between the fundamental wave signal of the phase current and the reverse potential after the shaping is calculated; i err ( i ) is the i The phase difference between the fundamental wave signal of the phase current and the opposite potential after the shaping is calculated; m It is the count value when the fundamental wave signal of the shaped counter potential and the phase current has opposite signs.
[0083] Fine-tuning stage: m ≤1, since the quantization unit is relatively large at high speed and high base frequency, and the calculation of the fundamental signal phase difference usually has an error of one quantization unit, in order to make the compensation amount further converge to the optimal value, in the fine-tuning stage, the continuous Y The compensation amount is calculated by averaging the phase differences, which can filter the phase difference and thus suppress the influence of the phase difference calculation error on the compensation amount calculation.
[0084] The electronic device of the present invention comprises: a memory and a processor coupled to each other, wherein the memory stores program data, and the processor calls the program data to execute the method as described above.
[0085] The computer-readable storage medium of the present invention stores program data thereon, and when the program data is executed by a processor, the method described above is implemented.
[0086] The beneficial effects of the present invention are:
[0087] 1) The present invention quantitatively calculates the lead angle produced by the short-time clamping voltage pulse that causes the phase of the opposite potential fundamental signal to lead the ideal sinusoidal back EMF, providing a theoretical basis for improving the commutation error compensation strategy.
[0088] 2) The present invention perfectly eliminates the influence of the short-time clamping voltage pulse on the phase of the fundamental signal of the opposite potential by shaping the opposite potential, thus providing a guarantee for achieving phase synchronization of the opposite potential and the phase current and obtaining the optimal compensation amount.
[0089] 3) The present invention utilizes an improved same-frequency filter based on phase correction to solve the phase discontinuity problem when the reference signal frequency changes, so that the fundamental signal can be effectively extracted when the speed changes or the speed fluctuates.
[0090] 4) The present invention utilizes a segmented iterative compensation algorithm to enable the phase difference to converge quickly in the coarse adjustment stage, and to suppress inaccurate calculation of the compensation amount caused by a large quantization unit in the fine adjustment stage, thereby improving the speed and accuracy of the compensation process.
[0091] In summary, the present invention can accurately extract the fundamental signal of the opposite potential. By synchronizing the phase current with the phase of the opposite potential after shaping, various commutation errors including the lead angle caused by the impedance characteristic delay and the short-time clamping voltage pulse can be compensated, thereby achieving the optimal commutation of the motor and improving the torque-to-current ratio and operating efficiency of the motor. BRIEF DESCRIPTION OF THE DRAWINGS
[0092] Figure 1 It is the overall control block diagram of the present invention;
[0093] Figure 2 It is a structural block diagram of a traditional same-frequency filter and an improved same-frequency filter of the present invention;
[0094] Figure 3 It is a flow chart of the segmented iterative compensation algorithm of the present invention;
[0095] Figure 4 It is a schematic diagram of the experimental results of the commutation mark detection and reverse potential shaping of the present invention;
[0096] Figure 5 Schematic diagram of the reference signal and the extracted a-phase current fundamental signal when the reference signal frequency changes in a specific embodiment of the present invention, wherein: Figure 5(a) is a schematic diagram of experimental results of the reference signal and the extracted a-phase current fundamental signal when the reference signal frequency changes before phase correction in a specific embodiment of the present invention, Figure 5 (b) is a schematic diagram of experimental results of the reference signal and the extracted a-phase current fundamental signal when the reference signal frequency changes after phase correction in a specific embodiment of the present invention;
[0097] Figure 6 Schematic diagram of compensation results before and after opposite potential shaping in a specific embodiment of the present invention, wherein: Figure 6 (a) is a schematic diagram of the experimental results before the opposite potential shaping in a specific embodiment of the present invention, Figure 6 (b) is a schematic diagram of the experimental results after opposite potential shaping in a specific embodiment of the present invention. DETAILED DESCRIPTION
[0098] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0099] like Figure 1 As shown, it is the overall block diagram of the high-speed brushless DC motor (BLDCM) without position sensor control using the optimal commutation error compensation method of the present invention. First, the reference speed n ref and actual speed n The difference is passed through the proportional integral PI (proportional integral) controller to obtain the constant current inductor current reference value. i L ref , constant current inductor current reference value i L ref and the actual constant current inductor current i L The difference is passed through the PI controller to obtain the power switch tube S in the quasi-current source inverter chopper circuit. w1 and S w2 Duty cycle d , duty cycle d After pulse width modulation PWM (pulse width modulation), the control power switch tube S is obtained. w1 and S w2 The switch signal, U d is the power supply voltage, L 1 is the constant current inductor, D 1 is a freewheeling diode. Secondly, the three-phase voltage u a , u b andu c After passing through the low pass filter LPF (low pass filter) and the line back electromotive force zero crossing detection link, the initial commutation signal is obtained, and then the optimal commutation error compensation method of the present invention is used to correct the initial commutation signal and then pass through the commutation logic link to obtain the control of the six power switch tubes S in the three-phase inverter bridge. aH , S aL , S bH , S bL , S cH , S cL The optimal commutation error compensation method of the present invention mainly includes three steps: taking phase a as an example, firstly, by judging the shut-off phase current i a , i b or i c Whether the voltage pulse is detected after decaying to zero S inf ,use S inf The opposite potential is shaped to eliminate the short-term clamping voltage pulse; the shaped opposite potential is then e a re and phase current i a and the reference signal frequency obtained from the motor speed w 0 After the improved isotropic filter ISFF based on phase correction, the fundamental wave signal with opposite potential after shaping is obtained. e as re And the fundamental signal of phase current i as , Δ P is the phase correction amount; the fundamental wave signal of the opposite potential after shaping e as re And the fundamental signal of phase current i as The phase difference of the two fundamental wave signals is obtained through the XOR logic and phase difference sign judgment link i err Finally, the phase difference of the two fundamental signals is converged to zero through the segmented iterative compensation algorithm to achieve the phase synchronization of the opposite potential and phase current and obtain the optimal compensation amount. i com , thereby achieving the optimal commutation of the motor.
[0100] In the specific embodiment of the high-speed brushless DC motor position sensorless optimal commutation error compensation method of the present invention, the control unit in the experiment adopts the TMS320F28335 control unit and the EP1C6Q240C8 control unit, the power MOS (Metal-Oxide-Semiconductor) tube adopts the C3M0016120K MOS tube, the control frequency is set to 20kHz, and the switching frequency is set to 40kHz. In order to better highlight the influence of the short-time clamping voltage pulse on the commutation error compensation effect, an additional 0.25mH inductor is connected in series in the three-phase winding circuit of the high-speed brushless DC motor in the experiment to increase the width of the short-time clamping voltage pulse. In the experiment, the motor speed is 12000r / min, the load is 0.3Nm, and the basic parameters of the high-speed brushless DC motor are shown in Table 1 below:
[0101] Table 1
[0102]
[0103] The method of the present invention first drives the high-speed brushless DC motor through a quasi-current source inverter and performs position sensorless control, and then obtains the fundamental wave signal of the reverse potential containing the short-time clamping voltage pulse after Fourier decomposition of the reverse potential of the high-speed brushless DC motor, as follows:
[0104] e xs ( i )= A 1 cos( i - i e / 2+ α )+ A 2 cos( i - i e / 2)=( A 1 2 + A 2 2 +2 A 1 A 2 cos α ) 1 / 2 cos( I )
[0105] I = i - i e / 2+ c
[0106] c =tan -1 (( A 1 sin α ) / ( A 1 cos α + A 2 ))
[0107] A 1 =(12 n / π)(sin i e / 2)
[0108] A 2 =( E / 2π)(-6 i e +2π)
[0109] α =π / 3- i d
[0110] in, e xs ( )and I They are respectively the fundamental wave signal of the x-phase reverse potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse and its phase. The x-phase of the high-speed brushless DC motor is one of the a-phase, b-phase and c-phase. i is the rotor angle of the high-speed brushless DC motor; A 1 , A 2 and α are respectively the first, second and third intermediate variables of the fundamental wave signal of the x-opposite potential containing the short-time clamping voltage pulse of the high-speed brushless DC motor; i e is the commutation phase duration angle of the high-speed brushless DC motor; c is the lead angle caused by the short-term clamping voltage pulse; n The first intermediate variable of the fundamental signal of the x reverse potential containing a short-term clamping voltage pulse of the high-speed brushless DC motor A 1 The intermediate variable n = U d / 3, U d is the power supply voltage; E is the opposite potential amplitude; id is the commutation error without position sensor.
[0111] Then, the phase of the fundamental signal of the x-phase reverse potential containing the short-time clamping voltage pulse is extracted from the fundamental signal of the x-phase reverse potential containing the short-time clamping voltage pulse, and the phase of the fundamental signal of the x-phase reverse potential containing the short-time clamping voltage pulse is compared with the phase of the x-phase ideal sinusoidal reverse potential to obtain the leading angle caused by the short-time clamping voltage pulse. c , to determine the phase consistency, when the leading angle caused by the short clamping voltage pulse c When it is zero, it is judged that the phase of the high-speed brushless DC motor is consistent. When the leading angle caused by the short-term clamping voltage pulse c When it is not zero, it is judged that the phase of the high-speed brushless DC motor is inconsistent. The short-term clamping voltage pulse causes the phase of the opposite potential fundamental wave signal to lead the ideal sinusoidal back EMF ( c >0) is: i e <π / 3,-π / 6< i d <π / 3.
[0112] Before obtaining the phase of the fundamental signal of the opposite potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse, the discontinuous opposite potential caused by the diode freewheeling of the quasi-current source inverter driven by the quasi-current source inverter is first analyzed. Taking the a phase of the high-speed brushless DC motor as an example, the details are as follows:
[0113] When the rotor angle of the high-speed brushless DC motor i =0° when switching, when there is a commutation error without position sensor i d When the high-speed brushless DC motor is running, a short-term clamping voltage pulse contains a reverse potential. e a ( i ) Specifically: e a ( i )= n , i d ≤ i < i e + i d ; e a ( i )= e a0 ( i ), i e + id ≤ i <60°+ i d ; e a ( i )=- n ,60°+ i d ≤ i <60°+ i e + i d ; e a ( i )= e a0 ( i ),60°+ i e + i d ≤ i <120°+ i d ; e a ( i )=-2 n ,120°+ i d ≤ i <120°+ i e + i d ; e a ( i )= e a0 ( i ),120°+ i e + i d ≤ i <180°+ i d ; e a ( i )=- n ,180°+ i d ≤ i <180°+ i e + i d ; e a ( i )= e a0 ( i), 180°+ i e + i d ≤ i <240°+ i d ; e a ( i )= n , 240°+ i d ≤ i <240°+ i e + i d ; e a ( i )= e a0 ( i ), 240°+ i e + i d ≤ i <300°+ i d ; e a ( i )=2 n , 300°+ i d ≤ i <300°+ i e + i d ; e a ( i )= e a0 ( i ), 300°+ i e + i d ≤ i <360°+ i d ;in, e a0 ( i ) is the ideal sinusoidal back EMF of phase a, e a0 ( i )= E cos( i - i e / 2).
[0114] Then the phase of the opposite potential fundamental signal containing a short-time clamping voltage pulse is analyzed and quantitatively calculated.
[0115] After determining the phase consistency, when the phases of the high-speed brushless DC motor are inconsistent, the reverse potential of the high-speed brushless DC motor in the commutation phase is shaped to obtain the shaped reverse potential to eliminate the short-time clamping voltage pulse. The shaped reverse potential is as follows:
[0116] e x re =- u z / 3, S inf =0
[0117] e x re =0, S inf =1
[0118] in, e x re is the x-phase phase potential after shaping of the high-speed brushless DC motor, and the x-phase of the high-speed brushless DC motor is one of the a-phase, the b-phase and the c-phase; u z is the line voltage difference of the high-speed brushless DC motor. When x=a, z=ca-ab, u ca-ab is the difference between the line voltage of phase c and phase a of the high-speed brushless DC motor and the line voltage of phase a and phase b. When x=b, z=ab-bc, u ab-bc is the difference between the line voltage of phase a and phase b of the high-speed brushless DC motor and the line voltage of phase b and phase c. When x=c, z=bc-ca, u bc-ca It is the difference between the line voltage of phase b and phase c of the high-speed brushless DC motor and the line voltage of phase c and phase a; S inf It is the first commutation mark. When the high-speed brushless DC motor is in the commutation stage, the first commutation mark S inf =1, when the high-speed brushless DC motor is in the non-commutation stage, the first commutation flag S inf =0.
[0119] Taking the a phase of the high-speed brushless DC motor as an example, the fundamental wave signal of the a phase potential after shaping is as follows:
[0120] e asre ( i )=( E / 2π)(-6 i e +2π)cos( i - i e / 2)
[0121] It can be seen that the phase of the fundamental wave signal of the shaped back electromotive force of phase a is consistent with the ideal sinusoidal back electromotive force of phase a.
[0122] The method of the present invention determines whether the high-speed brushless DC motor is in the commutation stage by detecting whether the off-phase current of the high-speed brushless DC motor decays to zero, that is, voltage pulse detection is performed according to the off-phase current of the high-speed brushless DC motor, so as to determine whether the high-speed brushless DC motor is in the commutation stage according to the commutation mark. The relationship between the off-phase current and the commutation mark under different switching states is specifically shown in Table 2 below:
[0123] Table 2
[0124]
[0125] Among them, in the switching state, A+B- indicates that the upper bridge arm of phase A and the lower bridge arm of phase B in the quasi-current source inverter are turned on, A+C- indicates that the upper bridge arm of phase A and the lower bridge arm of phase C in the quasi-current source inverter are turned on, A+B-→A+C- indicates that the quasi-current source inverter is in the switching process from the A+B- switching state to the A+C- switching state, and the others are similar; i a 、i b and i c They are the a, b and c phase currents of the high-speed brushless DC motor respectively; a current value of - indicates that the current value is less than zero, and a current value of + indicates that the current value is greater than zero.
[0126] The short-term clamping voltage pulse is then eliminated by setting the opposite potential during the commutation phase to zero.
[0127] The invention processes the phase current and the shaped reverse potential of the high-speed brushless DC motor through an improved isofrequency filter ISFF based on phase correction, and then extracts the fundamental wave signals of the shaped reverse potential and the phase current.
[0128] Assume that the input signal of the same frequency filter contains high-frequency harmonic components, as follows:
[0129] X in = V 0 sin( w 0t + f 0 )+∑ V h sin( w h t + f h )
[0130] in, V 0 , w 0 and f 0 The input signals are X in The amplitude, frequency and phase of the fundamental component; V h , w h and f h The input signals are X in middle h Amplitude, frequency and phase of subharmonic components.
[0131] Then, by using the orthogonality of trigonometric functions and performing integral operations, the signal can be obtained as follows:
[0132] ∫ X 1 ( t )d t =∫ X in sin( w 0 t )d t =( V 0 t cos f 0 ) / 2
[0133] ∫ X 2 ( t )d t =∫ X in cos( w 0 t )d t =( V 0 t sin f 0 ) / 2
[0134] Among them, sin( w 0 t ) and cos( w 0 t ) is the reference signal of the traditional frequency filter, and its frequency is obtained from the speed of the high-speed brushless DC motor.
[0135] The output signal of the traditional same-frequency filter is as follows:
[0136] X out = eV 0 t sin( w 0 t + f 0 ) / 2
[0137] Assumptions t = kT s When the speed changes or the speed fluctuates, the reference signal frequency of the traditional frequency filter is given by w 0 becomes w 0 ´, at this time, there is a phase mutation of the reference signal as follows:
[0138] Δ P ( t )=( w 0 ´- w 0 ) kT s
[0139] To ensure the phase continuity of the reference signal, the phase of the reference signal before and after the frequency change is as follows:
[0140] P 1 ( t )= w 0 t , t ∈[0, kT s )
[0141] P 2 ( t )= w 0 kT s + w 0 ´(t - kT s ), t ∈[ kT s ,∞)
[0142] The improved same-frequency filter corrects the phase when the reference signal frequency changes, and the phase of the reference signal before and after the frequency change after the phase correction can be obtained as follows:
[0143] P 1 ( t )= w 0 t
[0144] P 2 ( t )= w 0 ´ t -Δ P ( t )
[0145] After phase correction, the relationship between the reference signal phase before and after the frequency change is as follows:
[0146] P 2 ( t )= P 1 ( t )+ w 0 ´( t - kT s )
[0147] Finally, the improved same-frequency filter based on phase correction is obtained as follows:
[0148] X out = e sin( P 2 ( t ))∫ X 1 ( t )d t + e cos( P 2 ( t ))∫ X 2 ( t )d t
[0149] ∫X 1 ( t )d t =∫ X in sin( P 2 ( t ))d t
[0150] ∫ X 2 ( t )d t =∫ X in cos( P 2 ( t ))d t
[0151] P 2 ( t )= w 0 ´ t -Δ P ( t )
[0152] P 1 ( t )= w 0 t
[0153] Δ P ( t )=( w 0 ´- w 0 ) kT s
[0154] in, X in and X out They are the input signal and output signal of the improved same-frequency filter, respectively. The input signal is the x-phase current of the high-speed brushless DC motor. i x Or the x-phase potential after shaping e x re , the output signal is the x-phase current of the high-speed brushless DC motor i x The fundamental signal i xs Or the x-phase potential after shaping e xre The fundamental signal e xs re , the x phase of the high-speed brushless DC motor is one of the a phase, b phase and c phase; e is the gain of the improved same-frequency filter; t For time; P 1 ( t )and P 2 ( t ) are respectively t The reference signal phase before and after the frequency change of the time-modified same-frequency filter; X 1 ( t )and X 2 ( t ) are respectively t The first and second intermediate variables of the time-modified same-frequency filter; w 0 and w 0 'Respectively t The value before and after the frequency change of the reference signal of the improved same-frequency filter when the speed changes or the speed fluctuates; Δ P ( t )for t When the speed changes or the speed fluctuates, the phase of the reference signal of the traditional frequency filter changes suddenly. t = kT s , k is the sampling number, T s To control the cycle.
[0155] like Figure 2 As shown in FIG. 1 , it is a control block diagram of a conventional same-frequency filter and an improved same-frequency filter of the present invention. The difference lies in the reference signal. The reference signal of the conventional same-frequency filter is sin( w 0 t ) and cos( w 0 t ), the reference signal of the improved same-frequency filter is sin( w 0 ´ t -Δ P ( t )) and cos( w 0 ´ t -Δ P ( t)). The improved same-frequency filter can ensure the phase continuity of the reference signal by correcting the phase when the reference signal frequency changes, so that the fundamental signal can be effectively extracted when the speed changes or the speed fluctuates.
[0156] Finally, the phase difference between the fundamental wave signal of the shaped opposite potential and the phase current is converged by using the segmented iterative compensation algorithm. The shaped opposite potential and the phase current are processed by XOR logic and according to the leading and lagging commutation states of the high-speed brushless DC motor, the phase difference between the fundamental wave signal of the shaped opposite potential and the phase current is obtained as follows:
[0157] i err =(-1) r m Δ i err
[0158] Δ i err =2π pn m T s / 60
[0159] in, i err is the phase difference between the fundamental wave signal of the shaped opposite potential and the phase current; r It is the second commutation mark. When the high-speed brushless DC motor is in the advanced commutation state, the second commutation mark r =1, when the high-speed brushless DC motor is in the delayed commutation state, the second commutation flag r =0; m is the counting value when the fundamental wave signal of the shaped opposite potential and phase current has different signs, and the counting frequency is the control frequency of the high-speed brushless DC motor; Δ i err It is the phase difference quantization unit; p is the number of pole pairs of the high-speed brushless DC motor; n m is the speed of the high-speed brushless DC motor; T s To control the cycle.
[0160] Phase difference between the fundamental wave signal of the phase current and the reverse potential after shaping i err It is zero upon convergence.
[0161] The fundamental wave signal extracted under digital control is not a continuous sine wave, but a sine wave that changes in steps with the control cycle as the unit. At this time, the phase difference between the phase current calculated by the exclusive OR logic and the shaped reverse potential fundamental wave signal is as follows:
[0162] i err = m Δ i err
[0163] The phase difference is calculated whenever the phase current and the shaped counter-potential fundamental signal have opposite signs, so the phase difference is updated every half electrical cycle of the motor operation.
[0164] The motor commutation state can be determined by comparing the zero crossing sequence of the phase current fundamental wave signal and the shaped reverse electromotive force fundamental wave signal, that is, the sign of the phase difference can be determined. Taking phase a as an example, the determination method is shown in Table 3 below:
[0165] Table 3
[0166]
[0167] in, i as ( k -1) and i as ( k ) are respectively the previous k -1 moment and current k The fundamental signal of phase a current at the moment; e as re ( k -1) and e as re ( k ) are respectively the previous k -1 moment and current k The fundamental signal of the opposite potential of a after moment shaping.
[0168] In specific implementation, in order to reduce program execution time and reduce processor computing burden, the method of the present invention only extracts a certain opposite potential of the high-speed brushless DC motor and the fundamental signal of the corresponding phase current to obtain the phase difference of the opposite potential and the phase current, so only a certain opposite potential needs to be processed.
[0169] At the same time, the optimal compensation amount is determined by using the segmented iterative compensation algorithm. The segmented iterative compensation process is divided into a coarse adjustment stage and a fine adjustment stage. In the process of convergence of the phase difference between the shaped opposite potential and the fundamental wave signal of the phase current, the coarse adjustment stage compensation amount is obtained in the coarse adjustment stage according to the phase difference between the shaped opposite potential and the fundamental wave signal of the phase current in each iteration, and the fine adjustment stage compensation amount is obtained in the fine adjustment stage. Finally, the total compensation amount is obtained according to the coarse adjustment stage compensation amount and the fine adjustment stage compensation amount as follows:
[0170] i com = i 1 + i 2
[0171] i 1 = k iter1 ∑ N1=1 N1max ( i err ( i ) / N 1), m >1
[0172] i 2 = k iter2 ∑ N2=1 N2max ( i avg ( N 2)), m ≤1
[0173] i avg ( N 2)=(∑ i=1+Y(N2-1) YN2 ( i err ( i ))) / Y
[0174] in, i com is the total compensation amount; i 1 and i 2 They are the compensation amount in the coarse adjustment stage and the compensation amount in the fine adjustment stage respectively; k iter1 is the iteration coefficient of the coarse adjustment stage; N 1 and N 1 max are the number of iterations and the maximum number of iterations in the coarse adjustment stage respectively; i err ( i ) is the i The phase difference between the fundamental signal of the opposite potential and the phase current after the shaping of the iteration; m It is the count value when the fundamental wave signal of the shaped opposite potential and phase current has opposite signs; k iter2 is the iteration coefficient of the fine-tuning stage; N 2 and N 2max are the number of iterations and the maximum number of iterations in the fine-tuning phase, respectively; i avg ( ) is continuous Y The average value of the phase difference between the fundamental wave signal of the phase current and the reverse potential after shaping is calculated. Y =100.
[0175] In the coarse adjustment stage of the segmented iterative compensation algorithm, m >1, the phase difference is considered to be large. In order to make it converge quickly and avoid excessive accumulation of compensation and thus cross the optimal commutation point, an attenuation factor 1 / is set which decreases with the increase of the number of iterations. N 1 for compensation. In the fine-tuning stage of the segmented iterative compensation algorithm: m ≤1, since the quantization unit is relatively large at high speed and high base frequency, and the calculation of the fundamental signal phase difference usually has an error of one quantization unit, in order to make the compensation amount further converge to the optimal value, in the fine-tuning stage, the continuous Y The compensation amount is calculated by averaging the phase differences, which can filter the phase difference and thus suppress the influence of the phase difference calculation error on the compensation amount calculation.
[0176] like Figure 3 As shown, the segmented iterative compensation algorithm first initializes the number of iterations in the coarse adjustment stage N 1 and the number of iterations in the fine-tuning phase N 2. Make N 1= N 2=0, and end mark D =1, when m > 1, the number of iterations in the coarse adjustment stage is increased N 1 plus 1, calculate the compensation amount in the coarse adjustment stage q 1 ;when m ≤1, the average value of the fundamental signal phase difference is calculated in the fine-tuning stage, and then it is determined whether the average value is zero. If the average value is not zero, the number of iterations in the fine-tuning stage is N 2 plus 1, calculate the compensation amount in the fine adjustment stage i 2 If the average value is zero, the end mark D Add 1, then determine the end mark D Is it greater than 10? If not, the number of iterations in the fine-tuning phase N 2 plus 1, calculate the compensation amount in the fine adjustment stage i 2 If so, the process ends and the compensation amount is calculated according to the coarse adjustment stage. i 1 And the compensation amount in the fine adjustment stage i2 Total compensation received i com = i 1 + i 2 .
[0177] Until the phase difference between the fundamental wave signal of the shaped reverse potential and the phase current converges to zero, the total compensation amount at this time is obtained. i com As the optimal compensation amount, the commutation error of the position sensorless motor is compensated, thereby realizing the optimal commutation error compensation of the high-speed brushless DC motor under position sensorless control. i err It will become larger with the increase of speed. For high-speed working conditions, in order to achieve more precise compensation, the compensation process is divided into two steps: coarse adjustment and fine adjustment.
[0178] like Figure 4 As shown in FIG. 1 , the commutation mark detection and reverse potential shaping experimental results of the method of the present invention can be seen that the commutation mark S of the motor can be effectively detected by judging whether the off-phase current decays to zero. inf , after detecting the commutation flag S inf After it is set to 1, the opposite potential can be shaped to obtain the opposite potential without the short-term clamping voltage pulse.
[0179] like Figure 5 As shown, the reference signal before and after phase correction and the extracted a-phase current fundamental wave signal are given when the speed fluctuation of the high-speed brushless DC motor of the present invention causes the reference signal frequency to change. Figure 5 As shown in (a), the reference signal phase of the traditional frequency filter changes suddenly when the speed fluctuates, causing the reference signal to be intermittent, thus affecting the extraction of the fundamental signal, and ultimately leading to inaccurate calculation of the fundamental phase difference and commutation error compensation. Figure 5 As shown in (b), the improved same-frequency filter of the present invention can ensure the phase continuity of the reference signal after phase correction, so that the fundamental signal can be effectively extracted when the frequency of the reference signal changes.
[0180] like Figure 6 As shown, the compensation results before and after the reverse potential shaping of the present invention are given. Since the impedance voltage drop at high speed is negligible relative to the reverse potential, the zero crossing point of the phase voltage is the ideal reverse potential zero crossing point. In the figure, u aN is the voltage of phase a, and Zoom means local magnification. Figure 6 As shown in (a), the short-term clamping voltage pulse will make the phase of the opposite potential fundamental signal lead the ideal sinusoidal back EMF by 9.43°. Figure 6As shown in (b), the method of the present invention can accurately extract the fundamental signal of the opposite potential. By aligning the phase current with the shaped fundamental signal of the opposite potential, the optimal commutation of the motor can be achieved. The peak-to-peak value PP (Peak-to-Peak) of the phase current is reduced from 14.78A to 14.51A by 0.27A.
[0181] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0182] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for compensating the optimal commutation error of a high-speed brushless DC motor without a position sensor, characterized in that: include: Step 1) driving a high-speed brushless DC motor through a quasi-current source inverter and performing position sensorless control, and judging the phase consistency according to the phase of a fundamental wave signal of the high-speed brushless DC motor containing an opposite potential of a short-term clamping voltage pulse; Step 2) When the phases of the high-speed brushless DC motor are inconsistent, the reverse potential of the high-speed brushless DC motor in the commutation phase is shaped to obtain a shaped reverse potential; Step 3) Establish an improved same-frequency filter based on phase correction, input the phase current and the shaped reverse potential of the high-speed brushless DC motor into the improved same-frequency filter for processing, and extract the fundamental wave signals of the shaped reverse potential and phase current after processing by the improved same-frequency filter; Step 4) A piecewise iterative compensation algorithm is used to converge the phase difference between the fundamental signal of the shaped opposite potential and the phase current, and the optimal compensation amount is determined at the same time. The commutation error of the position sensorless is compensated according to the optimal compensation amount, thereby realizing the optimal commutation error compensation of the high-speed brushless DC motor under position sensorless control.
2. The method for compensating the high-speed brushless DC motor position sensorless commutation error according to claim 1, characterized in that: In the step 1), the reverse potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse is subjected to Fourier decomposition to obtain a fundamental wave signal of the reverse potential of the short-time clamping voltage pulse, as follows: e xs ( θ )=( A 1 2 + A 2 2 +2 A 1 A 2cos α ) 1 / 2 cos( Θ ) Θ = θ - θ e / 2+ γ γ =so -1 (( A 1sin α ) / ( A 1cos α + A 2)) A 1=(12 n / π)(sin θ e / 2) A 2=( E / 2π)(-6 θ e +2π) α =π / 3- θ d in, e xs ()and Θ They are respectively the fundamental wave signal of the x-phase reverse potential of the high-speed brushless DC motor containing a short-time clamping voltage pulse and its phase. The x-phase of the high-speed brushless DC motor is one of the a-phase, b-phase and c-phase. θ is the rotor angle of the high-speed brushless DC motor; A 1. A 2 and α are respectively the first, second and third intermediate variables of the fundamental wave signal of the x-opposite potential containing the short-time clamping voltage pulse of the high-speed brushless DC motor; θ e is the commutation phase duration angle of the high-speed brushless DC motor; γ is the lead angle caused by the short-term clamping voltage pulse; n The first intermediate variable of the fundamental signal of the x reverse potential containing a short-term clamping voltage pulse of the high-speed brushless DC motor A 1 intermediate variable, n = U d / 3, U d is the power supply voltage; E is the opposite potential amplitude; θ d is the commutation error without position sensor; Thus, the phase of the fundamental wave signal containing the x-phase reverse potential of the short-time clamping voltage pulse is extracted from the fundamental wave signal containing the x-phase reverse potential of the short-time clamping voltage pulse, and the phase of the fundamental wave signal containing the x-phase reverse potential of the short-time clamping voltage pulse is compared with the phase of the x-phase ideal sinusoidal reverse potential to obtain the leading angle caused by the short-time clamping voltage pulse. γ , when the lead angle caused by the short clamping voltage pulse γ When it is zero, it is judged that the phase of the high-speed brushless DC motor is consistent. When the leading angle caused by the short-term clamping voltage pulse γ When it is not zero, it is judged that the phase of the high-speed brushless DC motor is inconsistent.
3. The method for compensating the high-speed brushless DC motor position sensorless commutation error according to claim 1, characterized in that: In the step 2), the reverse potential of the high-speed brushless DC motor in the commutation phase is shaped to obtain the shaped reverse potential as follows: e x re =- u z / 3, S inf =0 e x re =0, S inf =1 in, e x re is the x-phase phase potential after shaping of the high-speed brushless DC motor, and the x-phase of the high-speed brushless DC motor is one of the a-phase, the b-phase and the c-phase; u z is the difference in line voltage of the high-speed brushless DC motor; S inf It is the first commutation mark. When the high-speed brushless DC motor is in the commutation stage, the first commutation mark S inf =1, when the high-speed brushless DC motor is in the non-commutation stage, the first commutation flag S inf =0.
4. The high-speed brushless DC motor position sensorless optimal commutation error compensation method according to claim 1, characterized in that: In the step 3), the improved same-frequency filter based on phase correction is as follows: X out = ε sin( Ψ 2( t ))∫ X 1( t )d t + ε body( Ψ 2( t ))∫ X 2( t )d t ∫ X 1( t )d t =∫ X in sin( Ψ 2( t ))d t ∫ X 2( t )d t =∫ X in cos( Ψ 2( t ))d t Ψ 2( t )= w 0´ t -D Ψ ( t ) Ψ 1( t )= w 0 t D Ψ ( t )=( w 0´- w 0) kT s in, X in and X out They are the input signal and output signal of the improved same-frequency filter respectively. The input signal is the x-phase current of the high-speed brushless DC motor. i x Or the x-phase potential after shaping e x re , the output signal is the x-phase current of the high-speed brushless DC motor i x The fundamental signal i xs Or the x-phase potential after shaping e x re The fundamental signal e xs re , the x phase of the high-speed brushless DC motor is one of the a phase, b phase and c phase; ε is the gain of the improved same-frequency filter; t For time; Ψ 1( t )and Ψ 2( t ) are respectively t The reference signal phase before and after the frequency change of the time-modified same-frequency filter; X 1( t )and X 2( t ) are respectively t The first and second intermediate variables of the time-modified iso-frequency filter; w 0 and w 0´ is t The value before and after the frequency change of the reference signal of the improved frequency filter when the speed changes or the speed fluctuates; Δ Ψ ( t )for t When the speed changes or the speed fluctuates, the phase of the reference signal of the same frequency filter changes suddenly. t = kT s , k is the sampling number, T s To control the cycle.
5. The method for compensating the high-speed brushless DC motor position sensorless optimal commutation error according to claim 1, characterized in that: In the step 4), the shaped reverse potential and phase current are processed by XOR logic and according to the leading and lagging commutation states of the high-speed brushless DC motor, the phase difference of the fundamental wave signal of the shaped reverse potential and phase current is obtained as follows: θ err =(-1) r m D θ err D θ err =2π pn m T s / 60 in, θ err is the phase difference between the fundamental wave signal of the shaped opposite potential and the phase current; r It is the second commutation mark. When the high-speed brushless DC motor is in the advanced commutation state, the second commutation mark r =1, when the high-speed brushless DC motor is in the delayed commutation state, the second commutation flag r =0; m is the counting value when the fundamental wave signal of the shaped opposite potential and phase current has different signs, and the counting frequency is the control frequency of the high-speed brushless DC motor; Δ θ err It is the phase difference quantization unit; p is the number of pole pairs of the high-speed brushless DC motor; n m is the speed of the high-speed brushless DC motor; T s To control the cycle; Phase difference between the fundamental wave signal of the phase current and the reverse potential after shaping θ err It is zero upon convergence.
6. The high-speed brushless DC motor position sensorless optimal commutation error compensation method according to claim 1, characterized in that: In the step 4), the optimal compensation amount is determined by using a segmented iterative compensation algorithm, and the segmented iterative compensation process is divided into a coarse adjustment stage and a fine adjustment stage. In the process of convergence of the phase difference between the shaped opposite potential and the fundamental wave signal of the phase current, the coarse adjustment stage compensation amount is obtained in the coarse adjustment stage according to the phase difference between the shaped opposite potential and the fundamental wave signal of the phase current in each iteration, and the fine adjustment stage compensation amount is obtained in the fine adjustment stage. Finally, the total compensation amount is obtained according to the coarse adjustment stage compensation amount and the fine adjustment stage compensation amount as follows: θ com = θ 1+ θ 2 in, θ com is the total compensation amount; θ 1 and θ 2 are the compensation amounts in the coarse adjustment stage and the fine adjustment stage respectively; Until the phase difference between the fundamental wave signal of the shaped reverse potential and the phase current converges to zero, the total compensation amount at this time is obtained. θ com The sensorless commutation error is compensated as the optimal compensation amount.
7. The method for compensating the high-speed brushless DC motor position sensorless commutation error according to claim 6, characterized in that: The coarse adjustment stage compensation θ 1 is as follows: θ 1= k iter1 ∑ N1=1 N1max ( θ err ( i ) / N 1), m >1 in, k iter1 is the iteration coefficient of the coarse adjustment stage; N 1 and N 1 max are the number of iterations and the maximum number of iterations in the coarse adjustment stage respectively; θ err ( i ) is the i The phase difference between the fundamental signal of the opposite potential and the phase current after the shaping of the iteration; m It is the count value when the fundamental wave signal of the shaped counter potential and the phase current has opposite signs.
8. The method for compensating the high-speed brushless DC motor position sensorless optimal commutation error according to claim 6, characterized in that: The compensation amount in the fine adjustment stage is θ 2 is as follows: θ 2= k iter2 ∑ N2=1 N2max ( θ avg ( N 2)), m ≤1 θ avg ( N 2)=(∑ i=1+Y(N2-1) YN2 ( θ err ( i ))) / Y in, k iter2 is the iteration coefficient of the fine-tuning stage; N 2 and N 2 max are the number of iterations and the maximum number of iterations in the fine-tuning phase, respectively; θ avg () is continuous Y The average value of the phase difference between the fundamental wave signal of the phase current and the reverse potential after the shaping is calculated; θ err ( i ) is the i The phase difference between the fundamental wave signal of the phase current and the opposite potential after the shaping is calculated; m It is the count value when the fundamental wave signal of the shaped counter potential and the phase current has opposite signs.
9. An electronic device, characterized in that: include: A memory and a processor coupled to each other, wherein the memory stores program data, and the processor calls the program data to execute the method according to any one of claims 1 to 8.
10. A computer-readable storage medium having program data stored thereon, characterized in that: When the program data is executed by a processor, the method according to any one of claims 1 to 8 is implemented.
