Fractional slot motor torque observation virtual displacement calculation deflection angle setting method
Through the combined method of derivation and experiment, the virtual displacement deflection angle of the fractional slot motor is determined, which solves the problem of low torque calculation accuracy of the fractional slot motor, and achieves the improvement of high-frequency torque observation and system reliability.
Patent Information
- Application Number
- CN202411049314.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-19
AI Technical Summary
Due to the high-frequency harmonic pole content of the motor torque calculation in fractional slot, the torque calculation is difficult, the accuracy is low, and the imaginary displacement deflection angle is difficult to determine, resulting in low calculation accuracy or large error.
By ignoring the cogging effect, it is approximately believed that the magnetic density distribution of the motor circumference is an ideal rectangle, and a stator current analysis expression with the rotor speed equal to zero is derived, a stator current sampling matrix during the electrical period is calculated, and a different imaginary displacement is set in combination with the magnetic field energy analysis. The deflection angle is obtained through experimental and theoretical calculations, and the standard mean square deviation calculation is performed to determine the set value of the imaginary displacement deflection angle.
High-frequency torque pulse vibration is realized, errors caused by too small or too large virtual displacement deflection angle are avoided, system torque observation accuracy and reliability are improved, and current sampling noise and data cutoff errors are avoided.
Smart Images

Figure CN118984088A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of power electronics and motor control, and in particular to a virtual displacement calculation and deflection angle setting method for torque observation of a fractional slot motor. Background Art
[0002] There are two main technical routes for megawatt-class wind power generation systems at present: doubly-fed induction motors and synchronous direct-drive motors. The stators and rotors of traditional doubly-fed motors both adopt a winding structure, and the rotor-side converter only needs to provide slip power, which has high reliability and is relatively economical. However, the leakage inductance of the doubly-fed motor increases significantly with the increase in the number of pole pairs, and the power factor decreases. It is difficult to achieve direct drive and high-speed gears are required to increase the speed, which increases the complexity of the mechanical transmission chain. Although the synchronous direct-drive motor does not rely on high-speed gears, the motor has a large diameter, which limits the flexibility of installation and application, and requires a full-power converter, so the system cost is relatively high. In summary, if the doubly-fed wind turbine can solve the key problem of low-speed and high-torque to achieve direct drive, it will effectively improve the torque density, reliability and safety of the motor and the transmission system. At the same time, the motor volume and cost can also be significantly improved.
[0003] Therefore, the fractional slot structure is applied to the doubly-fed wind turbine to achieve multi-pole with fewer slots, improve the torque density of the motor, and realize direct drive or semi-direct drive, which has certain engineering practical value. Torque calculation is the key technology of torque observation. Due to the high content of harmonic pole pairs in fractional slot motors, the torque calculation is difficult and the accuracy is low. For the torque analysis of fractional slot motors, the virtual displacement method is a more feasible solution, but because it is a discrete operation, the virtual displacement deflection angle is difficult to determine. If the virtual displacement deflection angle is set too large, the high-order harmonic components of the motor cannot be accurately reflected. If the virtual displacement deflection angle is set too small, the calculation accuracy will be low due to sampling noise. Therefore, it is necessary to improve the method of setting the virtual displacement calculation deflection angle. Summary of the invention
[0004] 1. Technical problems to be solved:
[0005] In view of the above technical problems, the present invention provides a method for adjusting the deflection angle of virtual displacement calculation for torque observation of fractional-slot motor. This scheme can not only realize high-frequency torque pulsation, but also avoid errors caused by too small virtual displacement deflection angle, current sampling noise, and data truncation of digital processing chips, thereby effectively improving the torque observation accuracy and reliability of the system.
[0006] 2. Technical solution:
[0007] A method for calculating deflection angle of virtual displacement based on torque observation of fractional slot motor, characterized in that it comprises the following steps:
[0008] Step 1: For motors with fractional slots in both the stator and rotor, the slot effect is ignored and the circumferential magnetic flux distribution obtained by excitation of each coil is approximately considered to be an ideal rectangle. Based on the inductance, the analytical expression of the stator and rotor currents under the conditions of zero rotor speed and stator three-phase winding AC voltage source excitation is derived:
[0009] Step 2: setting a preset number of time sampling points in an equal time step manner within one electrical cycle, and calculating the stator and rotor current sampling matrix within one electrical cycle of theoretical calculation according to the derived analytical expression of the stator and rotor currents;
[0010] Step 3: Analyze the magnetic field energy of the three-phase simplified model motor to obtain the collection of the motor magnetic field energy expressions of each time sampling point calculated theoretically, so as to obtain the virtual displacement torque theoretical calculation value matrix;
[0011] Step 4: Perform a stator AC voltage source excitation experiment on the above motor under the condition that the rotor speed is equal to zero, and measure the motor stator and rotor currents at each time sampling point; set a set of virtual displacement deflection angles θ displace ; For each time sampling point, a set of virtual displacement method experimental calculation torque is obtained through each element in a set of virtual displacement deflection angles, where each element corresponds to a Z, that is, corresponds to an N, that is, each element θ displace =Nπ, N∈{(1 / 100) Z}, where Z is a natural number and 1≤Z≤K, K is the upper limit of the preset power; for different time sampling points, the virtual displacement method experimental calculation torque matrix within an electrical cycle is obtained; the standard mean square error of the virtual displacement torque theoretical calculation value matrix and the virtual displacement method experimental calculation torque is calculated;
[0012] Step 5: Select the smallest value among the standard mean square error calculated in step 4, and use the virtual displacement deflection angle corresponding to the electromagnetic torque calculated by the virtual displacement method as the virtual displacement deflection angle setting value of the torque observation.
[0013] Furthermore, step one specifically includes:
[0014] S11: Ignoring the core magnetic resistance, slot effect, slot leakage inductance and end leakage inductance of the motor, it is approximately assumed that the air gap flux density obtained by the excitation of a single coil current is distributed as a rectangular wave in the circumferential direction of the motor. According to the motor pole slot scheme, the inductance matrix composed of the self-inductance of each phase winding and the mutual inductance between different phase windings is obtained through theoretical calculation, as shown in the following formula:
[0015]
[0016] (1) In the formula, L AA , L BB , L CC are the stator A, B, and C phase self-inductances respectively; Laa , L bb , L cc are the rotor phase a, b, c self-inductances respectively; L AB Equal to L BA , is the mutual inductance of stators A and B; L AC Equal to L CA , is the mutual inductance of stators A and C; L BC Equal to L CB , is the mutual inductance of stators B and C; L aa , L bb , L cc are the rotor phase a, b, c self-inductances respectively; L ab Equal to L ba , is the mutual inductance of rotors a and b; L bc Equal to L cb , is the mutual inductance of rotors b and c; L ac Equal to L ca , is the mutual inductance of rotors a and c; θ m is the mechanical angle between the geometric center axis of the stator A phase winding and the geometric center axis of the rotor a phase winding;
[0017] S12: Stator A, B, C three-phase windings and rotor a, b, c three-phase windings are excited by a DC voltage source, and the stator three-phase resistance R is obtained according to Ohm's law. SA , R SB , R SC And rotor three-phase winding resistance R ra , R rb , R rc ; Wherein, subscript S represents the stator, subscripts A, B, C represent the stator A, B, C phase windings respectively; subscript r represents the rotor, a, b, c represent the rotor a, b, c phase windings respectively;
[0018] S13: Align the geometric center axis of the stator A phase winding with the geometric center axis of the rotor a phase winding, i.e. θ m =0, let the rotor speed equal to 0, under the condition of three-phase symmetrical AC voltage source excitation of the stator three-phase winding, the stator and rotor current time function I is obtained by solving the differential equation group shown in equation (2): sA (t), I sB (t), I sC (t), I ra (t), I rb (t), I rc (t);
[0019]
[0020] (2) In the formula, u SA (t),u SB (t),uSC (t),u ra (t),u rb (t),u rc (t) represent the time function of the stator and rotor AC excitation power supply with the corresponding subscripts.
[0021] Furthermore, step 2 specifically includes: setting M sampling moments in one electrical cycle by using a time equal step method; obtaining each sampling moment t by using the stator and rotor current time function obtained by theoretical calculation. J The stator A, B, C phase currents and the rotor a, b, c phase currents correspond to each column of the theoretical calculation stator and rotor current sampling matrix; where J≤M; the theoretical calculation stator and rotor current sampling matrix is shown in the following formula (3), and the Mth sampling moment in the matrix corresponds to the Mth column of the sampling matrix:
[0022]
[0023] In the above formula, M is set according to engineering needs.
[0024] Furthermore, step three specifically includes:
[0025] S31: Energy analytical expression W(θ m ) is θ m and the function of the stator and rotor three-phase winding currents, and its specific expression is shown in the following formula (4):
[0026]
[0027] S32: The theoretical calculation of the theoretical calculation of the stator and rotor current sampling matrix t J Substitute the sampling values of the corresponding columns at the time into the above formula (4), and then get m The expression of the magnetic field energy of the three-phase motor with W as the independent variable J (θ m ); Since one electrical cycle includes M sampling moments, a set of M three-phase motor magnetic field energy expressions is obtained; {W(θ m )}={W1(θ m ),W2(θ m ),…W M (θ m )};
[0028] S33: Targeting t J W corresponding to the moment J (θ m ), in θ m = 0 to obtain the derivative T(t J )=dW J (θ m) / dθ m , we can get the derivative matrix T(t J )=[T(t1),T(t2),···,T(t M )], which is the virtual displacement torque theoretical calculation value matrix of M sampling moments; according to the above process, it can be seen that the virtual displacement torque theoretical calculation value matrix of the M sampling moments is a row matrix containing only one row of elements.
[0029] Furthermore, step four specifically includes:
[0030] S41: The stator AC voltage source excitation experiment is to align the geometric center axis of the stator A phase winding with the geometric center axis of the rotor a phase winding, set the rotor to a locked state, and then apply a three-phase symmetrical AC voltage source excitation to the stator;
[0031] S42: Obtain the three-phase steady-state current I of the stator and rotor in the experiment sA '(t),I sB '(t),I sC '(t),I ra '(t),I rb '(t),I rc '(t);
[0032] S43: The sampling period and time sampling points are the same as those calculated theoretically in step 2. Then the M columns of the experimental stator and rotor current sampling matrix corresponding to the M sampling moments are as follows:
[0033]
[0034] In the above formula, the superscript “,” represents the experimental sampling data related to the stator AC voltage source excitation experiment;
[0035] S44: Set the virtual displacement deflection angle to: θ displace =Nπ, N∈{(1 / 100) Z}, where Z is a natural number and 1≤Z≤K; K is the upper limit of the preset power; through finite element modeling and numerical calculation, for each N, two inductance matrices as shown in the following equation (6) can be obtained. The inductance matrix expression of the following equation corresponds to θ m They are θ dis1 = +Nπ / 2 and θ dis2 =-Nπ / 2:
[0036]
[0037] (6) In the formula, each N corresponds to a virtual displacement deflection angle θ displace , and the corresponding deflection angle end position θ m =θ dis1and the starting position θ m =θ dis2 , and then get two inductance matrices; for K virtual displacement deflection angles θ displace ,There are a total of 2K numerically solved inductance matrices;
[0038] S45: t obtained from step S43 (5) J The experimental current sampling data of this column at the time and the inductance matrix obtained by S44 formula (6) are used to obtain the experimental result expression of the three-phase motor magnetic field energy as follows:
[0039]
[0040] θ m =θ dis1 and θ m =θ dis2 Substitute into the above formula respectively, and get θ m =θ dis1 ,θ m =θ dis2 When using the experimental t J The two three-phase motor magnetic field energy expressions W'(θ dis1 ) and W'(θ dis2 );
[0041] S46: For each sampling moment in the experiment, K virtual displacement torque experimental results are obtained through K virtual displacement steps. When Z is a specific value, the specific virtual displacement deflection angle is θ displace =Nπ, N∈{(1 / 100) Z}; then the experimental result expression of virtual displacement torque corresponding to a specific virtual displacement step is T ex =[W'(θ dis1 )-W'(θ dis2 )] / θ displace ; Then the virtual displacement method experimental torque matrix in this electrical cycle is shown in formula (8):
[0042]
[0043] S47: Calculate the standard mean square error of each element in the only row of the virtual displacement torque theoretical calculation value matrix at M sampling moments and the elements in each row and column of the virtual displacement method experimental torque matrix to obtain K standard mean square errors, forming a standard mean square error matrix ε=[ε1,ε1,···,ε K ].
[0044] Furthermore, step five is as follows: Analyze and obtain the standard mean square error matrix ε=[ε1,ε1,···,ε K ] is the minimum value of ε G; That is: when Z = G, the standard mean square error reaches the minimum value, corresponding to N = (1 / 100) G , the discrete step size setting value of virtual displacement torque calculation is θ displace =(1 / 100) G π, and θ dis1 =+(1 / 100) G π / 2 and θ dis2 =-(1 / 100) G π / 2 are the end position and start position of the virtual displacement deflection angle calculated by the virtual displacement torque of the real-time control system.
[0045] 3. Beneficial effects:
[0046] (1) The method for calculating the deflection angle of virtual displacement of fractional slot motor torque observation proposed in the present invention is to derive the analytical expression of the magnetic field energy of the simplified model three-phase fractional slot motor, and theoretically calculate the torque observation verification benchmark. Since the torque observation value obtained by theoretical calculation is relatively accurate, and the error and sampling noise of different sampling systems are also very different. Therefore, the torque observation verification benchmark obtained by theoretical calculation is used as a reference for the torque observation value obtained by specific engineering applications, which can avoid excessive torque observation errors caused by current sampling errors and noise interference in engineering.
[0047] (2) The method for adjusting the deflection angle by calculating the virtual displacement of the fractional-slot motor torque observation proposed in the present invention is to establish a virtual displacement torque observation matrix of the real-time control system corresponding to different current sampling times and different virtual displacement deflection angles through experimental means for the torque observation in the experiment. Since the current magnitude at different time points is different, the sampling error and noise interference at each time point are also very random, and the randomness of the impact on the error of the virtual displacement torque calculation result at each moment is very large. Therefore, for different time points within an electrical cycle, the current is obtained, the virtual displacement is calculated, and the virtual displacement method torque observation value at each time point within an electrical cycle is obtained through experimental means, so as to facilitate the deviation analysis of the torque observation verification benchmark value and the virtual displacement method torque observation value obtained by the experiment at different time points.
[0048] (3) The method for calculating the deflection angle by observing the virtual displacement of the fractional-slot motor torque proposed in the present invention calculates the standardized mean square error of the virtual displacement torque observation matrix obtained by experiment and the torque observation verification benchmark calculated by theory to obtain the global optimal virtual displacement deflection angle on the time axis.
[0049] In summary, this method can not only achieve high-frequency torque pulsation, but also avoid errors caused by too small virtual displacement deflection angle, current sampling noise, and data truncation of digital processing chips, effectively improving the system torque observation accuracy and reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 This is the overall flow chart of this method;
[0051] Figure 2 Schematic diagram of the relative positions of the stator three-phase winding and the rotor three-phase winding in this method;
[0052] Figure 3 It is a sampling schematic diagram of the electrical cycle and sampling points in this method;
[0053] Figure 4 This is the flow chart of the calculation of the matrix of virtual displacement torque experimental results in the method;
[0054] Figure 5 This is a flow chart for determining the setting value of the virtual displacement deflection angle in this method;
[0055] Figure 6 An experimental platform constructed for a stator-rotor fractional slot concentrated winding doubly-fed induction motor in a specific embodiment;
[0056] Figure 7 It is a schematic diagram of a finite element model of a stator and rotor fractional slot concentrated winding doubly-fed induction motor in a specific embodiment;
[0057] Figure 8 It is a comparison diagram of torque obtained by virtual displacement method torque calculation and finite element analysis software in a specific embodiment. DETAILED DESCRIPTION
[0058] The present invention will be described in detail below with reference to the accompanying drawings.
[0059] As attached Figure 1 To Attachment Figure 5 As shown, a method for calculating the deflection angle of virtual displacement based on torque observation of a fractional slot motor is characterized in that it includes the following steps:
[0060] Step 1: For motors with fractional slots in both the stator and rotor, the slot effect is ignored and the circumferential magnetic flux distribution obtained by excitation of each coil is approximately considered to be an ideal rectangle. Based on the inductance, the analytical expression of the stator and rotor currents under the conditions of zero rotor speed and stator three-phase winding AC voltage source excitation is derived:
[0061] Step 2: setting a preset number of time sampling points in an equal time step manner within one electrical cycle, and calculating the stator and rotor current sampling matrix within one electrical cycle of theoretical calculation according to the derived analytical expression of the stator and rotor currents;
[0062] Step 3: Analyze the magnetic field energy of the three-phase simplified model motor to obtain the collection of the motor magnetic field energy expressions of each time sampling point calculated theoretically, so as to obtain the virtual displacement torque theoretical calculation value matrix;
[0063] Step 4: Perform a stator AC voltage source excitation experiment on the above motor under the condition that the rotor speed is equal to zero, and measure the motor stator and rotor currents at each time sampling point; set a set of virtual displacement deflection angles θ displace ; For each time sampling point, a set of virtual displacement method experimental calculation torque is obtained through each element in a set of virtual displacement deflection angles, where each element corresponds to a Z, that is, corresponds to an N, that is, each element θ displace =Nπ, N∈{(1 / 100) Z}, where Z is a natural number and 1≤Z≤K, K is the upper limit of the preset power; for different time sampling points, the virtual displacement method experimental calculation torque matrix within an electrical cycle is obtained; the standard mean square error of the virtual displacement torque theoretical calculation value matrix and the virtual displacement method experimental calculation torque is calculated;
[0064] Step 5: Select the smallest value among the standard mean square error calculated in step 4, and use the virtual displacement deflection angle corresponding to the electromagnetic torque calculated by the virtual displacement method as the virtual displacement deflection angle setting value of the torque observation.
[0065] Further, S11: Ignoring the core magnetic resistance, slot effect, slot leakage inductance and end leakage inductance of the motor, it is approximately considered that the air gap flux density obtained by the excitation of a single coil current is distributed as a rectangular wave in the circumferential direction of the motor. According to the motor pole slot scheme, the inductance matrix composed of the self-inductance of each phase winding and the mutual inductance between different phase windings is obtained through theoretical calculation, as shown in the following formula:
[0066]
[0067] (1) In the formula, L AA , L BB , L CC are the stator A, B, and C phase self-inductances respectively; L aa , L bb , L cc are the rotor phase a, b, c self-inductances respectively; L AB Equal to L BA , is the mutual inductance of stators A and B; L AC Equal to L CA , is the mutual inductance of stators A and C; L BC Equal to L CB , is the mutual inductance of stators B and C; L aa , L bb , L cc are the rotor phase a, b, c self-inductances respectively; L ab Equal to L ba , is the mutual inductance of rotors a and b; L bc Equal to L cb , is the mutual inductance of rotors b and c; L ac Equal to L ca, is the mutual inductance of rotors a and c; θ m is the mechanical angle between the geometric center axis of the stator A phase winding and the geometric center axis of the rotor a phase winding;
[0068] As attached Figure 2 As shown in the figure, it is a schematic diagram of the relative positions between the stator three-phase winding and the rotor three-phase winding in this scheme, where the mechanical angle between the geometric center axis of the stator A phase winding and the center axis of the rotor a phase winding is θ m express.
[0069] S12: Stator A, B, C three-phase windings and rotor a, b, c three-phase windings are excited by a DC voltage source, and the stator three-phase resistance R is obtained according to Ohm's law. SA , R SB , R SC And rotor three-phase winding resistance R ra , R rb , R rc ; Wherein, subscript S represents the stator, subscripts A, B, C represent the stator A, B, C phase windings respectively; subscript r represents the rotor, a, b, c represent the rotor a, b, c phase windings respectively;
[0070] S13: Align the geometric center axis of the stator A phase winding with the geometric center axis of the rotor a phase winding, i.e. θ m =0, let the rotor speed equal to 0, under the condition of three-phase symmetrical AC voltage source excitation of the stator three-phase winding, the stator and rotor current time function I is obtained by solving the differential equation group shown in equation (2): sA (t), I sB (t), I sC (t), I ra (t), I rb (t), I rc (t);
[0071]
[0072] (2) In the formula, u SA (t),u SB (t),u SC (t),u ra (t),u rb (t),u rc (t) represent the time function of the stator and rotor AC excitation power supply with the corresponding subscripts.
[0073] Furthermore, step 2 specifically includes: setting M sampling moments in one electrical cycle by using a time equal step method; obtaining each sampling moment t by using the stator and rotor current time function obtained by theoretical calculation. JThe stator A, B, C phase currents and the rotor a, b, c phase currents correspond to each column of the theoretical calculation stator and rotor current sampling matrix; where J≤M; the theoretical calculation stator and rotor current sampling matrix is shown in the following formula (3), and the Mth sampling moment in the matrix corresponds to the Mth column of the sampling matrix:
[0074]
[0075] In the above formula, M is set according to engineering needs.
[0076] As attached Figure 3 The figure shows a sampling schematic diagram of the electrical cycle and sampling points used in the present method. One electrical cycle is provided with a plurality of sampling points, and the time between adjacent sampling points is equal, that is, an equal time step method.
[0077] Furthermore, step three specifically includes:
[0078] S31: Energy analytical expression W(θ m ) is θ m and the function of the stator and rotor three-phase winding currents, and its specific expression is shown in the following formula (4):
[0079]
[0080] S32: The theoretical calculation of the theoretical calculation of the stator and rotor current sampling matrix t J Substitute the sampling values of the corresponding columns at the time into the above formula (4), and then get m The expression of the magnetic field energy of the three-phase motor with W as the independent variable J (θ m ); Since one electrical cycle includes M sampling moments, a set of M three-phase motor magnetic field energy expressions is obtained; {W(θ m )}={W1(θ m ),W2(θ m ),…W M (θ m )};
[0081] S33: Targeting t J W corresponding to the moment J (θ m ), in θ m = 0 to obtain the derivative T(t J )=dW J (θ m ) / dθ m , we can get the derivative matrix T(t J )=[T(t1),T(t2),···,T(t M)], which is the virtual displacement torque theoretical calculation value matrix of M sampling moments; according to the above process, it can be seen that the virtual displacement torque theoretical calculation value matrix of the M sampling moments is a row matrix containing only one row of elements.
[0082] Furthermore, step four specifically includes:
[0083] S41: The stator AC voltage source excitation experiment is to align the geometric center axis of the stator A phase winding with the geometric center axis of the rotor a phase winding, set the rotor to a locked state, and then apply a three-phase symmetrical AC voltage source excitation to the stator;
[0084] S42: Obtain the three-phase steady-state current I of the stator and rotor in the experiment sA '(t),I sB '(t),I sC '(t),I ra '(t),I rb '(t),I rc '(t);
[0085] S43: The sampling period and time sampling points are the same as those calculated theoretically in step 2. Then the M columns of the experimental stator and rotor current sampling matrix corresponding to the M sampling moments are as follows:
[0086]
[0087] In the above formula, the superscript “,” represents the experimental sampling data related to the stator AC voltage source excitation experiment;
[0088] S44: Set the virtual displacement deflection angle to: θ displace =Nπ, N∈{(1 / 100) Z}, where Z is a natural number and 1≤Z≤K; K is the upper limit of the preset power; through finite element modeling and numerical calculation, for each N, two inductance matrices as shown in the following equation (6) can be obtained. The inductance matrix expression of the following equation corresponds to θ m They are θ dis1 = +Nπ / 2 and θ dis2 =-Nπ / 2:
[0089]
[0090] (6) In the formula, each N corresponds to a virtual displacement deflection angle θ displace , and the corresponding deflection angle end position θ m =θ dis1 and the starting position θ m =θ dis2 , and then get two inductance matrices; for K virtual displacement deflection angles θ displace ,There are a total of 2K numerically solved inductance matrices;
[0091] S45: t obtained from step S43 (5) J The experimental current sampling data of this column at the time and the inductance matrix obtained by S44 formula (6) are used to obtain the experimental result expression of the three-phase motor magnetic field energy as follows:
[0092]
[0093] θ m =θ dis1 and θ m =θ dis2 Substitute into the above formula respectively, and get θ m =θ dis1 ,θ m =θ dis2 When using the experimental t J The two three-phase motor magnetic field energy expressions W'(θ dis1 ) and W'(θ dis2 );
[0094] S46: For each sampling moment in the experiment, K virtual displacement torque experimental results are obtained through K virtual displacement steps. When Z is a specific value, the specific virtual displacement deflection angle is θ displace =Nπ, N∈{(1 / 100) Z}; then the experimental result expression of virtual displacement torque corresponding to a specific virtual displacement step is T ex =[W'(θ dis1 )-W'(θ dis2 )] / θ displace ; as attached Figure 4 As shown, the virtual displacement method experimental torque matrix in this electrical cycle is shown in formula (8):
[0095]
[0096] S47: Calculate the standard mean square error of each element in the only row of the virtual displacement torque theoretical calculation value matrix at M sampling moments and the elements in each row and column of the virtual displacement method experimental torque matrix to obtain K standard mean square errors, forming a standard mean square error matrix ε=[ε1,ε1,···,ε K ].
[0097] The specific calculation process of standard mean square error is as follows: Figure 4 The following formula is shown in the figure:
[0098]
[0099] Furthermore, step five is as follows: Analyze and obtain the standard mean square error matrix ε=[ε1,ε1,···,εK ] is the minimum value of ε G ; That is: when Z = G, the standard mean square error reaches the minimum value, corresponding to N = (1 / 100) G , the discrete step size setting value of virtual displacement torque calculation is θ displace =(1 / 100) G π, and θ dis1 =+(1 / 100) G π / 2 and θ dis2 =-(1 / 100) G π / 2 are the end position and start position of the virtual displacement deflection angle calculated by the virtual displacement torque of the real-time control system. Specific embodiment:
[0101] This specific embodiment involves the trial production of a fractional-slot concentrated winding prototype, and the specific parameters are shown in the following table.
[0102]
[0103] The prototype in this specific embodiment is a conventional distributed winding motor with 45 slots and 36 slots in the stator and rotor respectively, and a fractional slot concentrated winding design with 15 slots in the stator and 18 slots in the rotor is realized. No coils are set in the extra slots, such as Figure 7 As shown in the figure, the black part in the outer ring stator indicates that there are coils, and the blank part is the empty slot; the black part in the inner ring rotor indicates that there are coils, and the white part is the empty slot.
[0104] The problems solved by this method in this embodiment involve: harmonic analysis of the open-circuit induced electromotive force of the prototype rotor, measurement of the average torque of the rotor, determination of the relative position of the geometric center axis of the stator A phase winding and the geometric center axis of the rotor a phase winding, measurement of the stator and rotor currents, and calculation of the torque in the virtual displacement method experiment. The experimental platform used in the embodiment is as follows Figure 8 As shown. In the experimental platform, the prime mover is connected to the prototype through a coupling and drives the prototype to rotate. The prime mover is powered by a frequency converter. One side of the frequency converter is connected to a 380V three-phase power supply, and the other side is connected to the prime mover. The frequency converter can set different output frequencies, adjust the speed of the prime mover, and drive the prototype to rotate. The prime mover shaft is set with a photoelectric encoder to measure the speed; the oscilloscope measures the three-phase voltages of the prototype stator A, B, and C and the three-phase induced electromotive force of the rotor a, b, and c through a voltage differential probe. The torque sensor installed on the prototype shaft measures the speed of the prototype and the average electromagnetic torque. The three-phase winding of the prototype rotor is a winding structure, connected to carbon brushes and slip rings. The three-phase winding of the rotor can achieve AC excitation through carbon brushes and slip rings, or three-phase short circuit, and the rotor current can be measured through the lead wires of carbon brushes and slip rings.
[0105] because Figure 6The torque measurement principle of the torque sensor installed on the rotating shaft of the prototype is based on material deformation, and the material deformation has inertia, which makes it impossible to measure high-frequency components, and this torque measurement method is currently the most widely used method. The torque measurement response time is greater than 10ms, so it is impossible to achieve the high-frequency torque pulsation measurement that the present invention focuses on. Therefore, in order to verify the practicality of the content of the present invention, finite element modeling was used for preliminary verification. The obtained finite element model and magnetic density and flux linkage analysis, such as Figure 7 shown.
[0106] Under static conditions, the mechanical position center axis of the stator A phase winding is aligned with the mechanical position center axis of the rotor a phase winding. Under the above specific conditions, the torque calculated by the virtual displacement method is compared with the torque obtained by the finite element analysis software. Figure 8 As shown in the figure, the discrete calculation methods used in the virtual displacement method torque calculation are the forward method with a step size of 0.1 degrees, the backward method with a step size of 0.1 degrees, and the forward + backward method with a step size of 0.2 degrees. Figure 8 The comparison results show that discrete forward, discrete backward, discrete forward + discrete backward, and discrete step length all have a significant impact on the results of torque calculation using the virtual displacement method.
[0107] Through the above experimental analysis, it can be seen that virtual displacement calculation is a very effective technical means for observing high-frequency torque components that cannot be measured by sensors. In addition, in practical engineering applications, the technical method proposed by the present invention plays an important role in improving the torque calculation accuracy based on the virtual displacement method.
[0108] Although the present invention has been disclosed as above in terms of preferred embodiments, they are not intended to limit the present invention. Anyone skilled in the art can make various changes or modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection defined by the claims of this application.
Claims
1. A method for calculating the deflection angle of virtual displacement based on torque observation of a fractional slot motor, characterized in that: The following steps are involved: Step 1: For motors with fractional slots in both the stator and rotor, the slot effect is ignored and the circumferential magnetic flux distribution obtained by excitation of each coil is approximately considered to be an ideal rectangle. Based on the inductance, the analytical expression of the stator and rotor currents under the conditions of zero rotor speed and stator three-phase winding AC voltage source excitation is derived: Step 2: setting a preset number of time sampling points in an equal time step manner within one electrical cycle, and calculating the stator and rotor current sampling matrix within one electrical cycle of theoretical calculation according to the derived analytical expression of the stator and rotor currents; Step 3: Analyze the magnetic field energy of the three-phase simplified model motor to obtain the collection of the motor magnetic field energy expressions of each time sampling point calculated theoretically, so as to obtain the virtual displacement torque theoretical calculation value matrix; Step 4: Perform a stator AC voltage source excitation experiment on the above motor under the condition that the rotor speed is equal to zero, and measure the motor stator and rotor currents at each time sampling point; set a set of virtual displacement deflection angles θ displace ; For each time sampling point, a set of virtual displacement method experimental calculation torque is obtained through each element in a set of virtual displacement deflection angles, where each element corresponds to a Z, that is, corresponds to an N, that is, each element θ displace =Nπ, N∈{(1 / 100) Z }, where Z is a natural number and 1≤Z≤K, K is the upper limit of the preset power; for different time sampling points, the virtual displacement method experimental calculation torque matrix within an electrical cycle is obtained; the standard mean square error of the virtual displacement torque theoretical calculation value matrix and the virtual displacement method experimental calculation torque is calculated; Step 5: Select the smallest value among the standard mean square error calculated in step 4, and use the virtual displacement deflection angle corresponding to the electromagnetic torque calculated by the virtual displacement method as the virtual displacement deflection angle setting value of the torque observation.
2. The method for calculating the deflection angle of virtual displacement based on torque observation of a fractional slot motor according to claim 1, characterized in that: Step 1 specifically includes: S11: Ignoring the core magnetic resistance, slot effect, slot leakage inductance and end leakage inductance of the motor, it is approximately assumed that the air gap flux density obtained by the excitation of a single coil current is distributed as a rectangular wave in the circumferential direction of the motor. According to the motor pole slot scheme, the inductance matrix composed of the self-inductance of each phase winding and the mutual inductance between different phase windings is obtained through theoretical calculation, as shown in the following formula: (1) In the formula, L AA , L BB , L CC are the stator A, B, and C phase self-inductances respectively; L aa , L bb , L cc are the rotor phase a, b, c self-inductances respectively; L AB Equal to L BA , is the mutual inductance of stators A and B; L AC Equal to L CA , is the mutual inductance of stators A and C; L BC Equal to L CB , is the mutual inductance of stators B and C; L aa , L bb , L cc are the rotor phase a, b, c self-inductances respectively; L ab Equal to L ba , is the mutual inductance of rotors a and b; L bc Equal to L cb , is the mutual inductance of rotors b and c; L ac Equal to L ca , is the mutual inductance of rotors a and c; θ m is the mechanical angle between the geometric center axis of the stator A phase winding and the geometric center axis of the rotor a phase winding; S12: Stator A, B, C three-phase windings and rotor a, b, c three-phase windings are excited by a DC voltage source, and the stator three-phase resistance R is obtained according to Ohm's law. SA , R SB , R SC And rotor three-phase winding resistance R ra , R rb , R rc ; Wherein, subscript S represents the stator, subscripts A, B, C represent the stator A, B, C phase windings respectively; subscript r represents the rotor, a, b, c represent the rotor a, b, c phase windings respectively; S13: Align the geometric center axis of the stator A phase winding with the geometric center axis of the rotor a phase winding, i.e. θ m =0, let the rotor speed equal to 0, under the condition of three-phase symmetrical AC voltage source excitation of the stator three-phase winding, the stator and rotor current time function I is obtained by solving the differential equation group shown in equation (2): sA (t), I sB (t), I sC (t), I ra (t), I rb (t), I rc (t); (2) In the formula, u SA (t),u SB (t),u SC (t),u ra (t),u rb (t),u rc (t) represent the time function of the stator and rotor AC excitation power supply with the corresponding subscripts.
3. The method for calculating the deflection angle of virtual displacement based on torque observation of a fractional slot motor according to claim 2, characterized in that: Step 2 specifically includes: setting M sampling moments in one electrical cycle by using a time equal step method; obtaining each sampling moment t by using the stator and rotor current time function obtained by theoretical calculation J The stator A, B, C phase currents and the rotor a, b, c phase currents correspond to each column of the theoretical calculation stator and rotor current sampling matrix; where J≤M; the theoretical calculation stator and rotor current sampling matrix is shown in the following formula (3), and the Mth sampling moment in the matrix corresponds to the Mth column of the sampling matrix: In the above formula, M is set according to engineering needs.
4. The method for calculating the deflection angle by virtual displacement based on torque observation of a fractional slot motor according to claim 3, characterized in that: Step three specifically includes: S31: Energy analytical expression W(θ m ) is θ m and the function of the stator and rotor three-phase winding currents, and its specific expression is shown in the following formula (4): S32: The theoretical calculation of the theoretical calculation of the stator and rotor current sampling matrix t J Substitute the sampling values of the corresponding columns at the time into the above formula (4), and then get m The expression of the magnetic field energy of the three-phase motor with W as the independent variable J (θ m ); Since one electrical cycle includes M sampling moments, a set of M three-phase motor magnetic field energy expressions is obtained; {W(θ m )}={W1(θ m ),W2(θ m ),…W M (θ m )}; S33: Targeting t J W corresponding to the moment J (θ m ), in θ m = 0 to obtain the derivative T(t J )=dW J (θ m ) / dθ m , we can get the derivative matrix T(t J )=[T(t1),T(t2),···,T(t M )], which is the virtual displacement torque theoretical calculation value matrix of M sampling moments; according to the above process, it can be seen that the virtual displacement torque theoretical calculation value matrix of the M sampling moments is a row matrix containing only one row of elements.
5. The method for calculating the deflection angle by virtual displacement based on torque observation of a fractional slot motor according to claim 4, characterized in that: Step 4 specifically includes: S41: The stator AC voltage source excitation experiment is to align the geometric center axis of the stator A phase winding with the geometric center axis of the rotor a phase winding, set the rotor to a locked state, and then apply a three-phase symmetrical AC voltage source excitation to the stator; S42: Obtain the three-phase steady-state current I of the stator and rotor in the experiment sA '(t),I sB '(t),I sC '(t),I ra '(t),I rb '(t),I rc '(t); S43: The sampling period and time sampling points are the same as those calculated theoretically in step 2. Then the M columns of the experimental stator and rotor current sampling matrix corresponding to the M sampling moments are as follows: In the above formula, the superscript "," represents the experimental sampling data related to the stator AC voltage source excitation experiment; S44: Set the virtual displacement deflection angle to: θ displace =Nπ, N∈{(1 / 100) Z }, where Z is a natural number and 1≤Z≤K; K is the upper limit of the preset power; through finite element modeling and numerical calculation, for each N, two inductance matrices as shown in the following equation (6) can be obtained. The inductance matrix expression of the following equation corresponds to θ m They are θ dis1 = +Nπ / 2 and θ dis2 =-Nπ / 2: (6) In the formula, each N corresponds to a virtual displacement deflection angle θ displace , and the corresponding deflection angle end position θ m =θ dis1 and the starting position θ m =θ dis2 , and then get two inductance matrices; for K virtual displacement deflection angles θ displace ,There are a total of 2K numerically solved inductance matrices; S45: t obtained by equation (5) in step S43 J The experimental current sampling data of this column at the time and the inductance matrix obtained by equation (6) in S44, the expression of the three-phase motor magnetic field energy experimental result is as follows: θ m =θ dis1 and θ m =θ dis2 Substitute into the above formula respectively, and get θ m =θ dis1 ,θ m =θ dis2 When using the experimental t J The two three-phase motor magnetic field energy expressions W'(θ dis1 ) and W'(θ dis2 ); S46: For each sampling moment in the experiment, K virtual displacement torque experimental results are obtained through K virtual displacement steps. When Z is a specific value, the specific virtual displacement deflection angle is θ displace =Nπ, N∈{(1 / 100) Z }; then the experimental result expression of virtual displacement torque corresponding to a specific virtual displacement step is T ex =[W'(θ dis1 )-W'(θ dis2 )] / θ displace ; Then the virtual displacement method experimental torque matrix in this electrical cycle is shown in formula (8): S47: Calculate the standard mean square error of each element in the only row of the virtual displacement torque theoretical calculation value matrix at M sampling moments and the elements in each row and column of the virtual displacement method experimental torque matrix to obtain K standard mean square errors, forming a standard mean square error matrix ε=[ε1,ε1,···,ε K ].
6. The method for calculating the deflection angle by virtual displacement based on torque observation of a fractional slot motor according to claim 5, characterized in that: Step 5 is as follows: Analyze and obtain the standard mean square error matrix ε=[ε1,ε1,···,ε K ] is the minimum value of ε G ; That is: when Z = G, the standard mean square error reaches the minimum value, corresponding to N = (1 / 100) G , the discrete step size setting value of virtual displacement torque calculation is θ displace =(1 / 100) G π, and θ dis1 =+(1 / 100) G π / 2 and θ dis2 =-(1 / 100) G π / 2 are the end position and start position of the virtual displacement deflection angle calculated by the virtual displacement torque of the real-time control system.
