Unitized permanent-magnet in-wheel motor and collaborative control system and method thereof

Abstract

A unitized permanent-magnet in-wheel motor comprises identical motor units 1 evenly distributed along a circumference of a radial section. Each of the motor units comprises an outer rotor 2, inner sta

Description

TECHNICAL FIELD The present disclosure belongs to the field of permanent-magnet motors, and in particular to a permanent-magnet in-wheel motor with high efficiency and multi-mode operation performance for electric vehicles, ship propulsion, electric tractors, and other fields. BACKGROUND Permanent-magnet in-wheel motors have promising applications in direct drive fields such as electric vehicles, ship propulsion, and electric tractors due to their advantages of high torque density and high power density. However, due to the constant and hard-to-adjust permanent magnetic field and the difficult deep flux weakening, permanent-magnet in-wheel motors have a small speed range and low efficiency in the high-speed range, making them hard to apply to electric vehicles and other fields that require multi-mode operation. Chinese patent application 202210042232.8 proposes a salient-pole hybrid excitation motor, which adopts hybrid excitation by a permanent magnet and an excitation current. Due to the adjustable excitation current, the difficulty of weakening the excited field of the motor is reduced, thereby achieving multi-mode operation of the motor. However, the introduction of the excitation current increases the copper loss of the motor and reduces the overall operating efficiency. Chinese patent application 201410768272.6 proposes a hybrid permanent-magnet memory motor with a magnetism gathering stator. The motor adopts hybrid excitation by a soft magnetic material with low coercivity and a rare-earth permanent magnet. The magnetomotive force of the soft magnetic material changes with the pulsation of the winding current, and the intensity of the excited field changes accordingly, thereby reducing the difficulty of flux weakening and achieving multi-mode operation of the motor. However, the introduction of the soft magnetic material and the pulsating winding increases the volume and weight of the motor and reduces the power density of the motor. In addition, the introduction of the pulsating winding increases the loss of the motor and reduces the operating efficiency of the motor. Therefore, it is an urgent issue to achieve efficient multi-mode operation of permanentmagnet in-wheel motors. SUMMARY Aiming to solve the problems in the multi-mode operation of existing permanent-magnet in wheel motors, the present disclosure proposes a more-poles fewer-slots unitized permanentmagnet in-wheel motor and a multi-unit collaborative control system and method thereof, to improve the operating efficiency and expand the high-efficiency operation range while satisfying the need of the permanent-magnet in-wheel motor for multi-mode operation. In a technical solution of the present disclosure, the more-poles fewer-slots unitized permanent-magnet in-wheel motor includes N identical motor units evenly distributed along a circumference of a radial section, where each of the motor units includes 1 / N outer rotors, 1 / N inner stators, and 1 / N centralized windings; the inner stators are coaxially nested inside the outer rotors and wound with the centralized windings; the centralized windings in each of the motor units are three-phase symmetrical and identically distributed; the outer rotors include a rotor core; 2a permanent magnet groups are evenly distributed along a circumference of the rotor core; each of the permanent magnet groups includes a first rectangular permanent magnet, a second rectangular permanent magnet, and an arc-shaped permanent magnet; the first rectangular permanent magnet and the second rectangular permanent magnet are identically structured and each provided with a rectangular radial section; the first rectangular permanent magnet and the second rectangular permanent magnet form a V-shaped arrangement, with an opening facing an air gap and with inner and outer oblique directions formed by rectangular length directions, outside the arc-shaped permanent magnet; the first rectangular permanent magnet and the second rectangular permanent magnet are symmetrical about a centerline of the arc-shaped permanent magnet in a diametral direction; the first rectangular permanent magnet has a magnetization direction perpendicular to a length direction of the first rectangular permanent magnet, the second rectangular permanent magnet has a magnetization direction perpendicular to a length direction of the second rectangular permanent magnet, and the arc-shaped permanent magnet has a magnetization direction the same as a direction of the centerline of the arc-shaped permanent magnet; in a same one of the permanent magnet groups, the magnetization directions of the first rectangular permanent magnet, the second rectangular permanent magnet, and the arc-shaped permanent magnet simultaneously point towards or away from the air gap, while the magnetization directions of each two adjacent ones of the permanent magnet groups are opposite; and a number Pr of rotor pole pairs, a number Ns of stator slots, a number m of motor phases, a slot-pitch angle t, and the number N ot the motor units simultaneously satisfy: r s, r , s N P ^5=1.5 r = ^*27r N=eN , , ct = d*2n, c , and ^=2 / ., where i. a, b, c, d, and e are positive integers. Further, the 1 / N outer rotors are each axially divided into M identical rotor segments, 20 mm<lef / M<12Q mm, where / e / denotes an axial length of the motor; and the M rotor segments are arranged in a way that the rotor segments rotate one mechanical misalignment angle in sequence along a same rotation direction. Further, each arc-shaped permanent magnet includes a radial section surrounded by an outer long side, an inner long side, and two short sides, of the arc-shaped permanent magnet; an arc center of the outer long side and the inner long side of the arc-shaped permanent magnet is the same as a center of the outer rotor; the short side of the arc-shaped permanent magnet is in a direction the same as a diametral direction of the outer rotor; the inner long side of the arc-shaped permanent magnet forms a sine curve ^max s'n^)' where e , ancjylmaY is an amplitude; and when 0i=3ti / 2, a point / i(3rr / 2) is located on an inner surface of the outer rotor. Further, the first rectangular permanent magnet and the second rectangular permanent magnet each are provided with an inner magnetic barrier at an end close to the air gap, and provided with an outer magnetic barrier at an end away from the air gap; and two ends of each arc-shaped permanent magnet in a tangential direction each are provided with a virtual slot that becomes a part of the air gap. In a technical solution of the present disclosure, the collaborative control system of the more-poles fewer-slots unitized permanent-magnet in-wheel motor includes a battery, two control modules, and N winding electronic switches, where the winding electronic switches each are configured to control on / off of the centralized windings in a respective one of the motor units; each of the two control modules includes a power electronic switch, a digital signal processor (DSP) controller, and an inverter in series; input terminals of the two power electronic switches are respectively connected to an output terminal of the battery; an output terminal of each inverter is connected to N / 2 of the winding electronic switches; and an output terminal of each of the centralized windings is connected to the battery through a rectifier. In a technical solution of the present disclosure, a control method of the collaborative control system includes: closing the two power electronic switches and the A’winding electronic switches; simulating with an abscissa denoting a speed of the motor and an ordinate denoting a torque output by the motor to acquire an outer characteristic curve g of the motor; disconnecting one of the power electronic switches, and simulating to acquire an outer characteristic curve f of the motor; taking a highest speed corresponding to a highest torque on the outer characteristic curve / as a critical speed nb', and determining a range in which the speed of the motor at an operation point satisfies np<nb as a constant-torque range; dividing the constant-torque range into a first range and a second range, where the first range is a range in which the torque at the operation point satisfies Tp<Tb, and the second range is a range in which the torque at the operation point torque satisfies Tp>Tb, Tb being a critical torque that is 7V-2 times a peak torque Tx of one of the motor units; and a maximum torque at the operation point in the second range does not exceed a corresponding torque on the outer characteristic curve g; and closing, when the operation point is in the first range, the two power electronic switches and at least L * j of winding electronic switches of the NI2 of the motor units connected to each of the control modules, Tpi being the torque at the operation point in the first range, and outputting, by the two inverters, currents of a same amplitude but different phases; and closing, when the operation point is in the second range, the two power electronic switches and the N winding electronic switches. In a technical solution of the present disclosure, the control method of the collaborative control system further includes: dividing a constant-power range in which the speed of the motor at the operation point satisfies n^nb into third to eighth ranges, where a range in which an efficiency at the operation point satisfies rjp>Tib is defined as a fourth range, being a boundary efficiency of the motor when a single control module is operating; the abscissa, the outer characteristic curve g, and a straight line j passing through a highest-speed point E of the fourth range and perpendicular to the abscissa enclose the eighth range; a range in which the torque is less than the torque at a lowest-speed point D of the fourth range and the speed is less than the speed at the point D is defined as a range S31; the abscissa, a straight line A: passing through the point / ) and perpendicular to the abscissa, a lower boundary connected by the points D and E of the fourth range, and the straight line j enclose a range S32; a union of S31 and S32 forms a third range; a range in which the speed is the same as the speed in the fourth range and the torque is twice the torque in the fourth range is defined as a sixth range; the sixth range includes a highest-speed point G on the straight line j and a lowest-speed point F on the straight line k: a range in which the torque is less than the torque at the point F but higher than the torque at the point D and the speed is less than the speed at the point F is defined as a range S51; the straight line k the straight line j, an upper boundary connected by the points D and E of the fourth range, and a lower boundary connected by the points F and G of the sixth range enclose a range S52; a union of S51 and S52 forms a fifth range; and a remaining range in the constant-power range forms a seventh range; and closing, when the operation point is in the third range, only one of the two power electronic switches and all the N winding electronic switches, and enabling uprated operation of the motor by increasing the torque without changing the speed of the motor; closing, when the operation point is in the fourth range, only one of the two power electronic switches and all the N winding electronic switches; closing, when the operation point is in the fifth range, the two power electronic switches and the Nwinding electronic switches, and enabling uprated operation; closing, when the operation point is in the sixth range, the two power electronic switches and the N winding electronic switches; closing, when the operation point is in the seventh range, the two power electronic switches and the N winding electronic switches, and outputting, by the two inverters, different currents; and closing, when the operation point is in the eighth range, the two power electronic switches and the N winding electronic switches. Further, the control method includes: simulating to calculate a torque waveform output by the motor and transition currents output by the two inverters when the of the winding electronic switches are closed; performing Fourier decomposition on the torque waveform, calculating a harmonic order r of a main harmonic component, and calculating a current misalignment angle ¢ = - I ^■r , where three-phase currents output by the two inverters have an amplitude of =l.Q51maxo, Imaxo being an amplitude of the transition currents; a phase of the three-phase current output by a first one of the inverters is ^ / 2 ahead of a phase of the transition currents; and a phase of the three-phase current output by a second one of the inverters is ¢12 behind the phase of the transition currents. Further, the enabling the uprated operation in the third range includes: determining a transition point P3 \nP3,TP3) with a same speed as an operation point P3(nP3,TP3) in the third range, PPg~Pp3'rp Pg~ PP3 rp P3' lp3 forming a set Sp3 with points satisfying ^P3 , selecting a point that belongs to the third or fourth range and has a torque greater than or equal to Tp3 from the set Sp3 as an uprated operation point simulating to acquire three-phase currents output by the inverters when the motor is operating at the uprated operation point H3Q1H3,Ths), and feeding back excess energy to the battery, where the uprated operation in the fifth range is the same as the uprated operation in the third range; np3, Tp3, and denote the speed, the torque, and the efficiency at the operation point P3(nP3,TP3), respectively; TP3’ and tjP3- denote the torque and an efficiency at the transition point, respectively; and i]pg denotes a power generation efficiency. Further, the control method further includes: determining, in the seventh range, two transition points with the same speed as the operation point in the seventh range and with a sum of the torques at the two transition points equal to the torque at the operation point in the seventh range; calculating a sum of total system power consumptions at the two transition points; and if a sum of 5 total system power consumptions of two operation points is equal to a minimum value of the sum of the total system power consumptions at the two transition points, controlling the two operation points by the two control modules respectively. With the above technical solutions, the present disclosure has the following advantages. 1. The present disclosure adopts a multi-unit design concept for the structural design of the unitized motor. The non-overlapping unitized design in space makes all the motor units independent of each other. Each motor unit is operated and controlled independently, greatly improving the operation and control freedom of the motor. The different combinations and working modes of the motor units enable the multi-mode operation of the permanent-magnet inwheel motor. 2. The present disclosure adopts a more-poles fewer-slots structure, abandoning the high requirements of traditional unitized motors for the pole-slot ratio, providing a new pole-slot ratio for unitized motors and increasing the number of rotor pole pairs. In this way, the permanentmagnet in-wheel motor has the performance characteristics of low speed and high torque, satisfying the performance requirements of in-wheel motors. 3. In the present disclosure, the rotor adopts a segmented oblique pole structure in the axial direction, which facilitates adjustment in the torque harmonic phase distribution of the rotor, such that the torque harmonic phase of the first rotor segment and the torque harmonic phase of the second rotor segment are mutually compensated. The design eliminates the highest-amplitude harmonic of the output torque, significantly reducing the torque ripple of the motor and improving the torque quality of the motor. 4. In the present disclosure, the long side of the arc-shaped permanent magnet close to the air gap adopts a sine curve design. The design facilitates adjustment in the saliency ratio of the motor and improves the flux weakening capability of the motor. The design also improves the sine performance of the magnetomotive force of the permanent magnet, reduces the harmonic complexity of the permanent magnetic field, and thus reduces the core loss of the motor. Therefore, the present disclosure improves the operating efficiency of the motor, reduces the torque ripple of the motor, and improves the comprehensive output performance of the motor. 5. In the present disclosure, the end of the rectangular permanent magnet close to the air gap is provided with the leakage magnetic circuit. Thus, with the change of the q-axis current, the permanent magnetic field of the motor achieves the design that leakage occurs in the event of low-load operation rather than high-load operation. That is, when the motor is overloaded, the permanent magnet flux is the effective flux to improve the torque output capability of the motor and achieve the high-load operation of the motor. When the motor is operating at high speed, the q-axis current decreases and the permanent magnetic field weakens to broaden the speed range of the motor and achieve high-speed operation of the motor. 6. In the multi-unit collaborative control system proposed by the present disclosure, independent winding electronic switches achieve independent control of the motor units, improving the control freedom and fault tolerance of the motor. The mutually independent power electronic switches decouple the two control modules of the motor, improving the control freedom of the motor, providing a hardware foundation for achieving high-performance operation of the motor, and improving the operational reliability of the motor in the low-load range. 7. In the present disclosure, when the motor is operating in the first or low-speed load range, the two inverters output currents of the same frequency and amplitude but different phases, and the motor units of the two modules are operating at the same operation point. At this point, the main harmonics of the torques output by the two controllers have a phase difference of 180°, and they compensate with each other to reduce the torque ripple of the motor. 8. In the present disclosure, when the motor is operating in the third or high-speed low-load range, only one of a first power electronic switch and a second power electronic switch is closed, all the winding electronic switches are closed, and the controlled motor units enable uprated operation. At this point, the operating efficiency of the motor increases, and excess energy is fed back to the battery through the rectifier, thereby improving the system efficiency of the motor. 9. In the present disclosure, when the motor is operating in the fifth or high-speed load range, all the power electronic switches and winding electronic switches are closed, and the motor units enable uprated operation. At this point, the operating efficiency of the motor increases, and excess energy is fed back to the battery through the rectifier, thereby improving the efficiency of the motor and control system. 10. In the present disclosure, when the motor is operating in the seventh or high-speed overload range, all the power electronic switches and winding electronic switches are closed, and the motor units of the two modules are operating at different operation points. The output torques are algebraically superposed, and the losses are algebraically superposed. In the collaborative operating mode, since all the motor units are operating in the high-efficiency range, the total loss of the motor is reduced, thereby improving the operating efficiency of the motor. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows a radial structure and unitized decomposition of a more-poles fewer-slots unitized permanent-magnet in-wheel motor according to the present disclosure. FIG. 2 shows a detail of a motor unit 1 shown in FIG. 1. FIG. 3 shows an axial structure of two rotor segments of the more-poles fewer-slots unitized permanent-magnet in-wheel motor according to the present disclosure. FIG. 4 is a radial view of the rotor segments shown in FIG. 3. FIG. 5 is a flowchart of calculating a mechanical misalignment angle a shown in FIG. 4. FIG. 6 shows torques output by the two rotor segments shown in FIG. 3. FIG. 7 shows a resultant torque of the output torques of the two rotor segments shown in FIG. 6. FIG. 8 shows a detail of the rotor structure shown in FIG. 2 and a schematic diagram of magnetization of permanent magnets. FIG. 9 shows a detail and geometric dimensions of a permanent magnet group shown in FIG. 8. FIG. 10 shows a detail and geometric dimensions of an arc-shaped permanent magnet shown in FIG. 9. FIG. 11 shows a detail and geometric dimensions of a first rectangular permanent magnet, an inner magnetic barrier and an outer magnetic barrier shown in FIG. 8. FIG 12 shows a detail and geometric dimensions of three permanent magnet groups shown in FIG. 8. FIG. 13 shows a detail and formation of a virtual slot of the rotor shown in FIG. 12. FIG. 14 shows a detail of a stator structure shown in FIG. 2. FIG. 15 shows an operating magnetic circuit of the more-poles fewer-slots unitized permanent-magnet in-wheel motor in low-load operation according to the present disclosure. FIG. 16 shows an operating magnetic circuit of the more-poles fewer-slots unitized permanent-magnet in-wheel motor in high-load operation according to the present disclosure. FIG. 17 is a block diagram of a multi-unit collaborative control system of the more-poles fewer-slots unitized permanent-magnet in-wheel motor in low-load operation according to the present disclosure. FIG. 18 shows a constant-torque range and a constant-power range of the multi-unit collaborative control system divided based on a critical speed in operation. FIG. 19 shows two sub-ranges of the constant-torque range shown in FIG. 18. FIG. 20 shows eight sub-ranges of the constant-power range shown in FIG. 18. Reference Numerals in the figures: 1. motor unit; 2. outer rotor; 3. inner stator; 4. centralized winding; 5. rotating shaft; 2.1. first rotor segment; 2.2. second rotor segment; 2.3. rotor core; 2.4. permanent magnet group; 2.5. inner magnetic barrier; 2.6. outer magnetic barrier; 2.7. virtual slot; 2.4.1. first rectangular permanent magnet; 2.4.2. second rectangular permanent magnet; 2.4.3. arc shaped permanent magnet; 2.4.3.1. outer long side; 2.4.3.2. inner long side; 2.4.3.3. short side; 2.5.1. inner magnetic barrier first side; 2.5.2. inner magnetic barrier second side; 2.5.3. inner magnetic barrier third side; 2.5.4. inner magnetic barrier fourth side; 2.5.5. inner magnetic barrier fifth side; 2.6.1. outer magnetic barrier first side; 2.6.2. outer magnetic barrier second side; 2.6.3. outer magnetic barrier third side; 2.6.4. outer magnetic barrier fourth side; 2.6.5. outer magnetic barrier fifth side; 3.1. magnetic ring; 3.2. salient pole; and 3.3. pole shoe. DETAILED DESCRIPTION OF THE EMBODIMENTS As shown in FIGS. 1 and 2, in the present disclosure, a more-poles fewer-slots unitized permanent-magnet in-wheel motor includes N identical motor units 1 evenly distributed along a circumference of a radial section. To ensure that the N motor units 1 are independently controlled by two control modules, the number of the motor units 1, namely N, satisfies: N=2i, where i is a positive integer. A center angle of each motor unit 1 on the radial section is Pn=2tiIN. Each motor unit 1 includes MN outer rotors 2, MN inner stators 3, MN centralized windings 4, and MN rotating shafts 5 distributed along the circumference of the radial section. Therefore, the permanent-magnet in-wheel motor composed of the N motor units 1 includes an outer rotor 2, an inner stator 3, a centralized winding 4, and a rotating shaft 5. The inner stator 3 is coaxial with the outer rotor 2 and nested inside the outer rotor 2. A center of the inner stator 3 is configured to hold the rotating shaft 5. The inner stator 3 is wound with the centralized winding 4. An air gap is formed between an inner wall of the outer rotor 2 and an outer wall of the inner stator 3. A thickness of the air gap is related to a power level of the motor, a material of permanent magnets, and a machining and assembly process of the outer rotor 2 and inner stator 3. The outer rotor 2 and the inner stator 3 are laminated with 0.35 mm thick silicon steel sheets, with a stacking coefficient of 0.95. Referring to FIGS. 3 and 4, the MN outer rotors 2 of each motor unit 1 are axially divided into Midentical segments, forming AT rotor segments, which are sequentially a first rotor segment, a second rotor segment, ..., and an M-th rotor segment. In order to reduce a torque ripple while reducing the machining difficulty of the outer rotor 2, M satisfies: 20 mm< / e / A / <120 mm, where lef denotes an axial length of the motor. The M rotor segments are arranged in such a way that they rotate one mechanical misalignment angle a in sequence along a same rotation direction and each two adjacent rotor segments differ by one mechanical misalignment angle a. In FIG. 3, takingM= 2 as an example, only two rotor segments are shown, namely a first rotor segment 2.1 and a second rotor segment 2.2. As shown in FIG. 4, the first rotor segment 2.1 and the second rotor segment 2.2 are staggered by one mechanical misalignment angle a. As shown in FIG. 3, the mechanical misalignment angle a refers to a counterclockwise rotation angle a of the first rotor segment 2.1 relative to the second rotor segment 2.2. To reduce the torque ripple of the unitized permanent-magnet in-wheel motor, as shown in FIG. 5, the mechanical misalignment angle a is determined as follows. Step 1. An initial mechanical misalignment angle «o is assigned as 0. Step 2. Simulation is performed through finite element software to acquire a motor output torque T(f) at the initial mechanical misalignment angle oco, and a motor torque ripple at the initial mechanical misalignment angle «o is calculated as an initial torque ripple TriPo. The initial torque ripple Tnpo is calculated by first calculating an average value of the output torque T(f), calculating a difference between maximum and minimum values of the output torque T(f), and calculating a percentage of the difference in the average value of the output torque T(f) as the initial torque ripple Tnpo. Step 3. Fast Fourier decomposition is performed on the motor output torque T(i), and a harmonic order k of a highest-amplitude harmonic component is calculated. Through the fast Fourier decomposition, the motor output torque T(t) is divided into a direct current component To, T costkwt + 0 ) the highest-amplitude harmonic component k k' , and a remaining harmonic YTscos(swt + 0s) component ^=1 , where, Tk denotes an amplitude of the highest-amplitude harmonic component; 3k denotes a phase; and 5 denotes an order of the remaining harmonic component, with an amplitude of Ts and a phase of 3S. Therefore, the fast Fourier decomposition of the motor output torque T(f) is expressed as follows: T (t) = To + Tk cos (kwt + 0k) + Ts cos (swt + 0s) " s= 1 k where, t denotes a time, and w denotes a rotational speed of the motor output torque T(t). Step 4. Based on the harmonic order k of the highest-amplitude harmonic component calculated in the step 3, a transitional mechanical misalignment angle ' is calculated, where Pr denotes a number of rotor pole pairs, and AT denotes the number of the rotor segments. Step 5. Simulation is performed through finite element software to acquire the motor output torque at the transitional mechanical misalignment angle on, and the motor torque ripple at the transitional mechanical misalignment angle is calculated as a transitional torque ripple TnP}. The calculation method of the transitional torque ripple TriP\ is the same as the calculation method of the initial torque ripple Tupo in the step 2. Step 6. The transitional torque ripple TriP\ is compared with the initial torque ripple Tripo in the step 2 to determine whether the transitional mechanical misalignment angle «i is able to effectively reduce the torque ripple. If the transitional torque ripple TriP\ is less than the initial torque ripple Tnpo, then the transitional mechanical misalignment angle al is able to effectively reduce the torque ripple, and the step 9 is executed. On the contrary, if the transitional torque ripple TrtPi is greater than or equal to the initial torque ripple Tripo, then the transitional mechanical misalignment angle «i is not able to effectively reduce the torque ripple, and the step 7 is executed. Step 7. The transitional mechanical misalignment angle ai is assigned to the initial mechanical misalignment angle ao. Step 8. Based on the initial mechanical misalignment angle «o assigned in the step 7, simulation is performed through finite element software to acquire the motor output torque 7( / ) after the mechanical misalignment angle is assigned to the initial mechanical misalignment angle ao. The steps 3 to 6 are repeatedly executed. If the transitional torque ripple Tripi in the step 6 is less than the initial torque ripple Tripo, then the transitional mechanical misalignment angle is able to effectively reduce the torque ripple, and the step 9 is executed. Step 9. The transitional mechanical misalignment angle «i is assigned to the mechanical misalignment angle a and output. At this point, an output torque T2.i(t) generated by the first rotor segment 2.1 is a sum of a —T, —Tkcos(kwt + 0k^ direct current component , a highest-amplitude harmonic component 7V7 and a remaining harmonic component 71. (t) = —Tn +—T, cos(kwt + 0. ) +—V T cos(swt + 0 ‘ M M k V k) M 5 V s^k The phases Ok and 0s of the second rotor segment 2.2 change relative to the first rotor segment 2.1. An output torque 7; ?( / ) generated by the second rotor segment 2.2 is a sum of a direct current — To —7^ cos | kwt + 0k + ^~ j component M , a highest-amplitude harmonic component M 1 M ' , and a 1 x—* y ( ~ 2571 ] — >Tcos\ swt + 0+-- -I y- SI S -» r-t I remaining harmonic component M S=1 ' : A Tk cos kwt + 0k+— + s^k By analogy, the phases Ok and 0s change sequentially. Among the M rotor segments, for a j-th rotor segment 2j,j being a positive integer and j<M, an output torque 72. / (0 generated by the j-th rotor segment 2j is a sum of a direct current component M °, a highest-amplitude harmonic 1 / —T, cos kwt+ 0,+( i -1)-- component 11 m j and a remaining harmonic component 1 ( zi . / . i --- / 1 cos swt + u 4“ I / — 1)---- s L k >Mk). T (A = —Tn+—T,cos\ kwt + 0.+(j-1)— +—V T cos\ SkVt+0 +(7-1)-' .V / A / L 7 / 1( / 71 I "U Mk) s^k Therefore, the phases of the k harmonics in the output torques of the M rotor segments differ 2£ by in sequence. That is, when the output torques of all the rotor segments are synthesized, the k harmonics form compensation, and the amplitude of the synthesized harmonics is 0. In other words, the output torque T’(f) of a magnetomotive force of the motor is a sum of a direct current component j=i and a remaining harmonic component 2sn s^k where, k denotes the harmonic order of the highest-amplitude harmonic component. Therefore, as the k harmonics disappear, the torque ripple is reduced. Taking M=2 as an example, FIG. 6 shows the effect after the mechanical misalignment angle a is introduced into the rotor. The abscissa indicates a rotor position, in electrical angle; and the ordinate indicates a cogging torque, in Nm. In FIG. 6, the black line indicates the cogging torque of the first rotor segment 2.1, and the gray line indicates the cogging torque of the second rotor segment 2.2. When the rotor position changes within the electrical angle of 0-360°, the cogging torques of the first rotor segment 2.1 and the second rotor segment 2.2 oscillate evenly within [-6.4,6.1]. When the cogging torque of the first rotor segment 2.1 is the maximum, the cogging torque of the second rotor segment 2.2 is the minimum, and vice versa. Therefore, the output torque of the first rotor segment 2.1 and the output torque of the second rotor segment 2.2 achieve peakvalley compensation, thereby eliminating the highest-amplitude harmonic component, reducing the torque ripple, and improving the torque quality of the motor. As shown in FIG. 7, the abscissa indicates the rotor position, in electrical angle; and the ordinate indicates the cogging torque, in Nm. In FIG. 7, the solid line indicates an actual rotor torque curve that combines the output torque of the first rotor segment 2.1 and the output torque of the second rotor segment 2.2. When the rotor position changes within the electrical angle of 0-360°, the rotor torque oscillates evenly within [-1.1,1], Compared to the torque curve in FIG. 6, in FIG. 7, the peak value of the torque curve is significantly reduced. Therefore, in the present disclosure, the mechanical misalignment angle a can significantly reduce the torque ripple of the motor and improve the torque quality. As shown in FIG. 8, the outer rotor 2 of the permanent-magnet in-wheel motor includes a rotor core 2.3, 2a permanent magnet groups 2.4, 4a inner magnetic barriers 2.5, 4a outer magnetic barriers 2.6, and 4a virtual slots 2.7, where a is a positive integer. The 2a permanent magnet groups 2.4 are evenly distributed along the circumference of the entire rotor core 2.3. As shown in FIG. 9, each permanent magnet group 2.4 includes a first rectangular permanent magnet 2.4.1, a second rectangular permanent magnet 2.4.2, and an arc-shaped permanent magnet 2.4.3. 2a arc-shaped permanent magnets 2.4.3 are provided inside an inner surface of the outer rotor 2, and the 2a arc-shaped permanent magnets 2.4.3 are evenly arranged along the circumference. A centerline of each arc-shaped permanent magnet 2.4.3 on the radial section is in a direction the same as a diametral direction, and a centerline of the permanent magnet group 2.4 coincides with the centerline of arc-shaped permanent magnet 2.4.3. The first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 have the same structure. They are provided with rectangular radial sections, and are both located outside the arcshaped permanent magnet 2.4.3. In a same permanent magnet group 2.4, the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 are symmetrically arranged about the centerline of the permanent magnet group 2.4, and the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 are also symmetrically arranged about the centerline of the arc-shaped permanent magnet 2.4.3. In the same permanent magnet group 2.4, the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 form a V-shaped arrangement with an opening facing the air gap and inner and outer oblique directions formed by rectangular length directions. The arc-shaped permanent magnet 2.4.3 is provided in a middle of the V-shaped opening formed by the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2, and these three permanent magnets are not in contact with each other. As shown in FIG. 8, in the same permanent magnet group 2.4, the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 each are magnetized in a direction that is perpendicular to a length direction thereof and the same as a width direction thereof. The arc-shaped permanent magnet 2.4.3 is magnetized in a direction the same as the centerline direction thereof, that is, pointing towards or away from a center of a circle. In the same permanent magnet group 2.4, the magnetization directions of the first rectangular permanent magnet 2.4.1, the second rectangular permanent magnet 2.4.2, and the arc-shaped permanent magnet 2.4.3 simultaneously point towards or away from the air gap, while the magnetization directions of each two adjacent permanent magnet groups 2.4 are opposite. Ends of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 close to and away from the air gap each are provided with a magnetic barrier, that is, inner and outer ends of the rectangular permanent magnet each are provided with a magnetic barrier. Specifically, the inner ends of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 each are provided with an inner magnetic barrier 2.5, and there are a total of 4a inner magnetic barriers 2.5. The outer ends of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 each are provided with an outer magnetic barrier 2.6, and there are a total of 4a outer magnetic barriers 2.6. Two ends of each arcshaped permanent magnet 2.4 in a tangential direction on the inner surface of the outer rotor 2 each are provided with a virtual slot 2.7, and there are a total of 4a virtual slots 2.7. The virtual slot 2.7 are connected to the air gap and become an integral part of the air gap. As shown in FIG. 9, in order to balance the flux weakening capability and peak torque output capability of the permanent-magnet in-wheel motor, a V-shaped angle ppm formed by the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 satisfies: 40°<^pm<65°. In addition, in order to balance the strength, machining difficulty, and stress distribution of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2, a length and a width hpm of each of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 satisfy: 2<wpm / hpm<4. As shown in FIG. 10, each arc-shaped permanent magnet 2.4.3 is surrounded by an outer long side 2.4.3.1, an inner long side 2.4.3.2, and two short sides 2.4.3.3 that are symmetrical about the center of the arc-shaped permanent magnet 2.4.3. An arc center of the outer long side 2.4.3.1 is the same as the center of the outer rotor 2. A center angle of the inner long side 2.4.3.2 is the same as that of the outer long side 2.4.3.1. The short side 2.4.3.3 is in a direction the same as the diametral direction of the outer rotor 2 and is located on a radius of the outer rotor 2. The inner long side 2.4.3.2 is a half-cycle sine curve, with an independent variable within [7t,27t]. That is, a sine curve function of the inner long side 2.4.3.2 is: fi (A) = fma. sin ), ^ g [^, 2^-] where fimca denotes an amplitude of the sine curve, which is determined by a specific 14 performance requirement of the motor. When B\=3ti / 2, a corresponding point / i(3tt / 2) on the sine curve is located exactly on the inner surface of the outer rotor 2. In this way, the arc-shaped permanent magnet 2.4.3 is integrally located inside the inner surface of the outer rotor 2, and when Bi=3ti / 2, the point / i(3ti / 2) coincides with the inner surface of the outer rotor 2. The sine curve design of the inner long side 2.4.3.2 changes the magnetomotive force of the arc-shaped permanent magnet 2.4.3. The magnetomotive force of the arc-shaped permanent magnet 2.4.3 changes from an original square wave to a superposition of a rectangular wave and a sine wave. This changes the harmonic distribution of the magnetomotive force of the arc-shaped permanent magnet 2.4.3, improves the sine performance of the magnetomotive force of the arcshaped permanent magnet 2.4.3, increases the fundamental wave amplitude of the magnetomotive force, and thus improves the torque output capability of the motor. In addition, the design of this sine curve shape also reduces the fundamental wave amplitude of the magnetomotive force of the permanent magnet, thereby reducing the core loss of the motor. As shown in FIG. 10, in order to enhance the permanent magnetic field of the motor while reducing the reluctance of a main magnetic circuit to increase the peak torque of the motor, a minimum width hpmm of the arc-shaped permanent magnet in the diametral direction satisfies: 6<hpjhpmin<8. In addition, in order to reduce the harmonic of the permanent magnetic field to reduce the iron loss of the motor while reducing the average air gap width of the motor to reduce the reluctance of the main magnetic circuit, the minimum width hpmin of the arc-shaped permanent magnet 2.4.3 and a maximum width hpmax of the arc-shaped permanent magnet 2.4.3 satisfy: 1. S^hpmaxlhpmin^2 ■ Referring to FIGS. 11 and 12, the inner magnetic barriers 2.5 provided at the ends of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 close to the air gap have the same structure and are symmetrically distributed about the centerline of the permanent magnet group 2.4. Taking the inner magnetic barrier 2.5 at the end of the first rectangular permanent magnet 2.4.1 as an example, the radial section of the inner magnetic barrier 2.5 isapentagon, surrounded by an inner magnetic barrier first side 2.5.1, an inner magnetic barrier second side 2.5.2, an inner magnetic barrier third side 2.5.3, an inner magnetic barrier fourth side 2.5.4, and an inner magnetic barrier fifth side 2.5.5. The inner magnetic barrier first side 2.5.1 is an extended side of the long side of the first rectangular permanent magnet 2.4.1 close to the air gap side, and the inner magnetic barrier second side 2.5.2 is an arc side coaxial with the outer rotor 2 of the motor. In order to construct a leakage magnetic circuit of the permanent magnetic field, expand the motor speed range, and improve the efficiency of the motor in a high-speed range, while controlling the leakage flux of the motor to improve the torque output capability of the motor in a low-speed low-load range, a distance hbi between the inner magnetic barrier second side 2.5.2 and the inner surface of the outer rotor 2 satisfies: 0.75<Am / / ?pm<0.9. The inner magnetic barrier third side 2.5.3 is located a radius of the outer rotor 2. The inner magnetic barrier fourth side 2.5.4 is parallel to the inner magnetic barrier first side 2.5.1 and located outside the inner magnetic barrier first side 2.5.1, i.e. away from the air gap. In order to reduce the uncontrollable magnetic leakage of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 while ensuring installation reliability, a distance hbi between the inner magnetic barrier first side 2.5.1 and the inner magnetic barrier fourth side 2.5.4 is smaller than the width of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2, and satisfies: 0^hpm<hb2<0.9hpm. The inner magnetic barrier fifth side 2.5.5 coincides with the short side of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 close to the air gap. The outer magnetic barriers 2.6 are provided at the ends of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 away from the air gap. The two outer magnetic barriers 2.6 have the same structure and are symmetrically distributed about the centerline of the permanent magnet group 2.4. Taking the outer magnetic barrier 2.6 at the end of the first rectangular permanent magnet 2.4.1 as an example, the radial section of the outer magnetic barrier 2.6 is a pentagon, surrounded by an outer magnetic barrier first side 2.6.1, an outer magnetic barrier second side 2.6.2, an outer magnetic barrier third side 2.6.3, an outer magnetic barrier fourth side 2.6.4, and an outer magnetic barrier fifth side 2.6.5. The outer magnetic barrier first side 2.6.1 is an extended side of the long side of the first rectangular permanent magnet 2.4.1 close to the air gap side, the outer magnetic barrier second side 2.6.2 is located on the radius of the outer rotor 2, and the outer magnetic barrier third side 2.6.3 is an arc side coaxial with the outer rotor 2. In order to reduce the uncontrollable magnetic leakage of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 while reducing a g-axis magnetic reluctance to maximize the torque, a distance hb3 between the outer magnetic barrier third side 2.6.3 and an outer surface of the outer rotor 2 satisfies: Q2(Rro-Rri)<hb3<035(Rm-Rr^ where Rm denotes an outer diameter of the outer rotor 2, and Rn denotes an inner diameter of the outer rotor 2. The outer magnetic barrier fourth side 2.6.4 is parallel to the outer magnetic barrier first side 2.6.1 and located outside the outer magnetic barrier first side 2.6.1, i.e. away from the air gap. In order to reduce the magnetic leakage of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 while ensuring installation reliability, a distance hb4 between the outer magnetic barrier first side 2.6.1 and the outer magnetic barrier fourth side 2.6.4 is smaller than the width of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2, and satisfies: Q8hpm<hb4=hb\<Q.9hpm. The outer magnetic barrier fifth side 2.6.5 coincides with the short side of the first rectangular permanent magnet 2.4.1 and the second rectangular permanent magnet 2.4.2 away from the air gap. As shown in FIG. 13, in order to improve the sine performance of the air gap flux density and reduce the iron loss and torque ripple of the motor, two virtual slots 2.7 symmetrical about the centerline of the arc-shaped permanent magnet 2.4.3 are arranged at two ends of each arc-shaped permanent magnet 2.4.3. Each virtual slot 2.7 forms a sine curve. On the radial section, the virtual slot 2.7 intersects with an end of the inner long side 2.4.3.2 of the arc-shaped permanent magnet 2.4.3 at a point A, which is also an intersection point of the inner long side 2.4.3.2 and the short side 2.4.3.3. The inner magnetic barrier second side 2.5.2 intersects with the inner magnetic barrier third side 2.5.3 at a point B. A radius passing through the point B intersects with the inner surface of the outer rotor 2 at a point C. A sine curve from the point A to the point C forms the virtual slot 2.7. The sine curve of the virtual slot 2.7 is provided with an independent variable 6*2 within [7t / 2,7t], and a sine curve function of the virtual slot 2.7 is: / 2 (A)= sin(e2), t where, fimax denotes an amplitude of the function, which is determined by the specific performance requirement of the motor. When 6*2=71 / 2, a point^2(71 / 2) on the sine curve coincides with the point A. When 02=71, the point72(71) on the sine curve coincides with the point C. The virtual slot 2.7 can change the magnetic conductivity of the outer rotor 2 and the air gap magnetomotive force of the permanent magnetic field. The magnetic conductivity changes from an original square wave to the superposition of a square wave and a sine wave. Meanwhile, pole reduction is enabled for the magnetomotive force of the permanent magnet group 2.4, which improves the sine performance of the magnetomotive force of the arc-shaped permanent magnet 2.4.3, thereby increasing the fundamental amplitude of the magnetomotive force and enhancing the torque output capability of the motor. In addition, the design of the virtual slot 2.7 reduces the fundamental amplitude of the magnetomotive force, thereby reducing the core loss of the motor. As shown in FIG. 14, the inner stator 3 of the permanent-magnet in-wheel motor includes a magnetic ring 3.1, B salient poles 3.2, and IB pole shoes 3.3, whereB=mj, m denotes a number of motor phases, and j is a positive integer. An outer end of each salient pole 3.2 extends towards two sides of the tangential direction to form two pole shoes 3.3. The salient poles 3.2 are rectangular in the radial section and evenly distributed along a circumference of an outer surface of the magnetic ring 3.1. The outer end of the salient pole 3.2 forms two identical pole shoes 3.3 at the two sides of the tangential direction. A radial width of the magnetic ring 3.1, a radial length and tangential width of the salient pole 3.2, and a tangential width of the pole shoe 3.3 are determined by the motor power. In the more-poles fewer-slots unitized permanent-magnet in-wheel motor of the present disclosure, a number Pr of rotor poles, a number Ns of stator slots, the number m of the motor phases, and the number N of the motor units 1 satisfy the following conditions. (1) More-poles fewer-slots design. The number Pr of the rotor pole pairs is greater than the number Ns of the stator slots, and a slot-pitch angle t is equal to a product of and a quotient of the number Pr of the rotor pole pairs divided by the number Ns of the stator slots. (2) Rotor unitization. The rotor is divided into N parts, where the number Pr of the rotor pole pairs is an integer multiple of the number N of the motor units 1. (3) Stator unitization. The stator is divided into A’ parts, and the number Ns of the stator slots is an integer multiple of the number m of the motor phases and an integer multiple of the number N of the motor units 1. (4) Control module unitization. In order to improve the control freedom of the unitized motor and improve the overall operating efficiency and performance of the motor, the more-poles fewer-slots unitized permanent-magnet in-wheel motor in the present disclosure is provided with two control modules, so the number TV of the motor units 1 is an integer multiple of 2. (5) Centralized winding. The winding is a centralized winding 4, with a winding span of 1. (6) Winding unitization. On the basis of stator unitization, the windings in each motor unit 1 are three-phase symmetrical and identically distributed, that is, the following equations are established simultaneously through positive integers i, a, b, c, d, and e: Pr> Ns Pr = Na Ns = mNb N = 2i < ^Pr Pr *0 r = —— * 2n Ns ct = d * 2n N s _ C In the more-poles fewer-slots unitized permanent-magnet in-wheel motor of the present disclosure, a condition for low-load operation is that the output torque T of the motor satisfies: T<Q.6Trated, where Trated denotes a rated output torque of the motor. A condition for high-load operation is that the output torque T of the motor satisfies: T>Q.6Trated. A condition for low-speed operation is that a motor speed n satisfies: n<nrated, where nrated denotes a rated speed of the motor. A condition for high-speed operation is that the motor speed n satisfies: n>nrated. Referring to FIG. 15, when the more-poles fewer-slots unitized permanent-magnet in-wheel motor in the present disclosure is in low-load operation, there are two magnetic circuits in the motor, that is, a main magnetic circuit I and a leakage magnetic circuit II which are in parallel. A flux path of the main magnetic circuit I starts from the first rectangular permanent magnet 2.4.1 in a first permanent magnet group 2.4. It then passes through the rotor core 2.3, a first arc-shaped permanent magnet 2.4.3, the air gap, the inner stator 3, the air gap, a second arc-shaped permanent magnet 2.4.3 adjacent to the first arc-shaped permanent magnet 2.4.3, the rotor core 2.3, the second rectangular permanent magnet 2.4.2 in a second permanent magnet group 2.4 adjacent to the first permanent magnet group 2.4, and the rotor core 2.3 in sequence. Finally, the flux path of the main magnetic circuit I returns to the first rectangular permanent magnet 2.4.1 to form a closed magnetic circuit. There is a magnetic leakage circuit II in the motor that is different from a traditional magnetic circuit. A flux path of the leakage magnetic circuit II starts from the first rectangular permanent magnet 2.4.1 in the first permanent magnet group 2.4. It then passes through the rotor core 2.3, the second arc-shaped permanent magnet 2.4.3 adjacent to the first arc-shaped permanent magnet 2.4.3, the rotor core 2.3, the second rectangular permanent magnet 2.4.2 in the second permanent magnet group 2.4 adjacent to the first permanent magnet group 2.4, and the rotor core 2.3. Finally, the flux path of the leakage magnetic circuit II returns to the first rectangular permanent magnet 2.4.1 to form a closed magnetic circuit in the outer rotor 2. Therefore, the main magnetic circuit I is connected in parallel with the leakage magnetic circuit II. Referring to FIG. 16, when the more-poles fewer-slots unitized permanent-magnet in-wheel motor in the present disclosure is in high-load operation, there are two magnetic circuits in the motor, that is, a main magnetic circuit I and a q-axis magnetic circuit III which operate in parallel. The main magnetic circuit I is the same as the main magnetic circuit I (FIG. 15) involved in low-load operation. A flux path of the q-axis magnetic circuit III starts from the inner stator 3. It then passes through the air gap, the arc-shaped permanent magnet 2.4.3, the rotor core 2.3, an adjacent arc-shaped permanent magnet 2.4.3, and the air gap in sequence. Finally, the flux path of the q-axis magnetic circuit III returns to the inner stator 3 to form a closed magnetic circuit. Therefore, the main magnetic circuit I is in parallel with the q-axis magnetic circuit III. According to FIGS. 15 and 16, the leakage magnetic circuit II and the q-axis magnetic circuit III overlap in some sections, and the reluctance of the overlapping section is high, making it easy to saturate. When the more-poles fewer-slots unitized permanent-magnet in-wheel motor in the present disclosure is in low-speed low-load operation, the speed and torque of the motor are relatively low. Therefore, if a current amplitude / output by the centralized winding 4 is small, then a q-axis current component iq is small. That is to say, the magnetic flux of the q-axis magnetic circuit III is weak, and the overlapping section of the leakage magnetic circuit II and the q-axis magnetic circuit III is saturated with the magnetic flux of the leakage magnetic circuit II. When the motor is in high-speed low-load operation, due to the high torque of the motor, the output current amplitude I and current angle of the centralized winding 4 are large, and the q-axis current component iq is small. The magnetic circuit distribution in high-speed low-load operation is the same as that in low-speed low-load operation. In addition, due to the weak magnetic flux of the q-axis magnetic circuit III and the strong magnetic flux of the leakage magnetic circuit II, the excited field of the motor is weakened. Therefore, compared to a traditional motor, the motor in the present disclosure has lower copper consumption in the high-speed range. In addition, when the motor of the present disclosure is in high-load operation, as the q-axis current component iq increases, the magnetic flux of the q-axis magnetic circuit III increases, and the magnetic flux of the leakage magnetic circuit II gradually weakens until it disappears. At this point, all the magnetic flux of the permanent magnet group 2.4 forms an effective magnetic flux through the main magnetic circuit I, thereby improving the torque output capability of the motor. Referring to FIG. 17, the more-poles fewer-slots unitized permanent-magnet in-wheel motor in the present disclosure is controlled by a collaborative control system. The collaborative control system includes a battery, two control modules, and N winding electronic switches. A winding electronic switch is connected to the centralized winding 4 in the motor unit 1 to control the on / off of the centralized winding 4 in the motor unit 1. Each control module includes a power electronic switch, a digital signal processor (DSP) controller, and an inverter in series. Input terminals of the two power electronic switches are respectively connected to output terminals of the battery. An output terminal of each inverter is connected to N / 2 winding electronic switches. That is, the first control module includes a first power electronic switch, a first DSP controller, and a first inverter in series. The second control module includes a second power electronic switch, a second DSP controller, and a second inverter in series. The power electronic switches in the two control modules are independent of each other. The two control modules are independent of each other, with the same structure, and each controlling N / 2 motor units 1. The design reduces the coupling between the two modules and improves the control freedom and quality of the more-poles fewer-slots unitized permanent-magnet in-wheel motor. The N winding electronic switches are divided into two independent groups. Each group of winding electronic switches control A72 motor units 1. In other words, the N winding electronic switches are divided into two groups connected to the two control modules. For example, a first winding electronic switch is connected to a first motor unit 1, a second winding electronic switch is connected to a second motor unit 1, an (A72)-th winding electronic switch is connected to an (A / 2)-th motor unit 1, an (A72+l)-th winding electronic switch is connected to an (A72+l)-th motor unit 1, and an A-th winding electronic switch is connected to an .V-th motor unit 1. The input terminals of the NH winding electronic switches from the first winding electronic switch to the (A72)-th winding electronic switch are connected to the first inverter in the first control module. The input terminals of the N / 2 winding electronic switch from the (A72-1 )-th winding electronic switch to the .V-th winding electronic switch are connected to the second inverter in the second control module. In this way, the on / off of each motor unit 1 is independently controlled by the corresponding winding electronic switches. The total armature field strength of the motor can be changed at multiple levels, making it easy to adjust the magnetic field of the motor and achieve multi-mode operation of the motor. The independent control of the motor units 1 further improves the operating freedom of the more-poles fewer-slots unitized permanent-magnet in-wheel motor, providing a hardware foundation for improving the efficiency of the motor and the collaborative control system. In addition, the output terminals of the centralized windings 4 of all the motor units 1 are connected to the battery through a feedback module. The main component of the feedback module is a rectifier. When the motor is in uprated operation, if the energy generated by the motor is higher than the energy required for in-wheel drive, the remaining energy is recovered into the battery through the feedback module. The collaborative control system adopts the following control strategy to control the more-poles fewer-slots unitized hub permanent magnet motor proposed in the present disclosure. Step 1. Simulation is performed through finite element software to acquire an outer characteristic curve of the motor when the control module is operating. The two power electronic switches and the N winding electronic switches are closed. Simulation is performed through finite element software to acquire an outer characteristic curve g of the motor, as shown in FIG. 18. All the N winding electronic switches are closed. Only one of the first power electronic switch and the second power electronic switch is closed, that is, one of the power electronic switches is disconnected, and only one control module is operating. Simulation is performed through finite element software to acquire an outer characteristic curve / of the motor. In the figure, the abscissa indicates the motor speed n, in rpm; and the ordinate indicates the output torque T of the motor, in Nm. Step 2. Based on the acquired outer characteristic curve, a constant-torque range and a constant-power range of the motor are divided. T’ Step 2.1. A critical speed rib is calculated. A highest torque max is derived from the outer characteristic curve f. The critical speed »6 is a highest speed corresponding to a highest torque T max, that is: T = nb =A / ax[»;] ? where, n, denotes a set of speeds at an operation point when the motor outputs a peak torque. Step 2.2. The constant-torque range and the constant-power range are divided based on the critical speed Hb. A point P(np,Tp) is set as any operation point of the motor, where np denotes a speed at the point, and Tp denotes a torque at the point. If the speed np at the point satisfies: np<nb, then the point belongs to the constant-torque range, that is, the speed of the motor at the operation point in the constant-torque range satisfies: np<nb. If the speed np at this point satisfies: np>nb, the point belongs to the constant-power range. As shown in FIG. 18, the dotted line h is a boundary between the constant-torque range and the constant-power range, and the speed at the boundary satisfies: Hp nb. Step 3. Based on the critical torque, the constant-torque range is divided into a first or low-speed load range and a second or low-speed overload range. Step 3.1. Simulation is performed through finite element software to calculate a peak torque Tx of one motor unit 1. When the first power electronic switch and the second power electronic switch are closed, that is, when the first control module and the second control module are working, there is only one winding electronic switch closed. At this point, the peak torque of the motor is acquired through simulation using the finite element software. This torque is the peak torque Tx of one motor unit 1. The critical torque Tb is calculated based on the peak torque Tx of one motor unit I. The critical torque Tb is N-2 times the peak torque Tx of one motor unit 1: r^N-iyr, Step 3.3. Referring to FIG. 19, the abscissa indicates the motor speed in rpm; and the ordinate indicates the output torque of the motor, in Nm. Based on the critical torque Tb, two ranges are determined, namely the first or low-speed load range and the second or low-speed overload range. If the operation point P(nP,Tp) satisfies: Tp<Tb and np<nb, then the operation point P(np,Tp) belongs to the first or low-speed load range. That is, in the constant-torque range, if the torque Tp of the operation point P(np,Tp) is equal to or less than the critical torque Tb, then the operation point S P(np,Tp) belongs to the first or low-speed load range '. On the contrary, if the point P(jh„TP) S satisfies: Tp>Tb and np<nb, then the pointP belongs to the second or low-speed overload range 2. The maximum value of the torque Tp of the operation point P(np,Tp) in the second or low-speed g overload range 2 does not exceed the torque on the outer characteristic curve g. The division result of the constant-torque range is as follows: \={p(npJp)\Q<rip<nbS^ T I 0 <n T">t\ \np’ p) Iu - np -nbAp-^*b) In FIG. 19, the horizontal line j is a boundary between the first or low-speed load range and the second or low-speed overload range, and its function is expressed as follows: The two ranges of the constant-torque range are marked differently in FIG. 19. The grid range represents the first or low-speed load range, and the forward slash range represents the second low-speed low-load range. Step 4. Based on the division result of the constant-torque range in the step 3, a control method is determined for the first or low-speed load range. A main control principle for the first or low-speed load range is a principle of minimizing the torque ripple. The first power electronic switch and the second power electronic switch are closed, and the first inverter and the second inverter output currents of the same amplitude but different phases. The phases of the output torques of the two control modules are adjusted to achieve valley-peak overlap, thereby eliminating harmonics of higher amplitudes. For any operation pointin the first or low-speed load range, where npi denotes the speed at the point and Tpi denotes the torque at the point, the control method is implemented as follows. Step 4.1. A number K (l<K<N / 2) of winding electronic switches connected to each control module that need to be closed is calculated based on an output torque Tp\ required by the motor. The output torque required by the motor at the operation point P\{np^Tpi) is Tp\, so the required torque output for a single control module is TP\I2. In addition, the peak torque of a single motor unit 1 calculated in the step 3.1 is Tx. Therefore, among the N / 2 motor units 1 connected to each control module, a minimum number of the closed winding electronic switches isK=(Tpi / 2Tx). To ensure the torque output capability of the motor, the number K of the closed winding electronic K = switches in each control module is rounded according to a "plus one" method, T 1 pi 2Tx Step 4.2. Based on the number K of the closed winding electronic switches calculated in the step 4.1, simulation is performed through finite element software to calculate an output torque TAf) and transition currents output by the two inverters when the number of the closed winding electronic switches is K. The first power electronic switch and the second power electronic switch are closed. Arbitrarily K switches from the first winding electronic switch to the (A / 2)-th winding electronic switch are closed, and arbitrarily K switches from the (A / 2+l)-th winding electronic switch to the A-th winding electronic switch are closed. At this point, the output currents of the two inverters are exactly the same. Simulation is performed through finite element software to calculate the output torque / 2( / ) of the motor and the transition currents (amplitude Imaxo and phase Oo) output by the two inverters, and the corresponding sine functions are as follows: I sin\ —n Pt + 0n ho p r 2 nPt — ^- + ft 0sin\ Pt + maxi) I 3Q p r 3 2 a — IT + ca 3 0 n 30 'n sin \—nPt- — n + 0, moxv I 30 P r 3 I ( 7T 2 (,s i n \ —n Pt + — 7r + 0, max® I 30 P r 3 C where, ^1.4( / ), Ab(0> and Ac( / ) denote the transition currents output by the first inverter; ^24( / ), ^25( / ), and 2c( / ) denote the transition currents output by the second inverter; np denotes the speed at the operation point P(np,Tpy, and Pr denotes the number of rotor pole pairs. Step 4.3. Based on the result acquired in the step 4.2, fast Fourier decomposition is performed to acquire the output torque / 2( / ), and a harmonic order r of a main harmonic component is calculated. The output torque / 2( / ) is divided into a direct current component / 20, a main harmonic t / , n\ \ Tcos(^\t + 0v\ Tcos\rwJ+0\ , "7 component r • 2 r! , and a remaining harmonic component V=1 1 ., . T cos(rwJ+0} . through the fast Founer decomposition. The mam harmonic component ' “ r / has an amplitude of Tr and a phase of 0r. The remaining harmonic component V=1 has an order of v, an amplitude of Tv, and a phase of (P Therefore, the fast Fourier decomposition of the output torque Tz(t) is expressed as follows: = T,o + Tr cos (rw2t + Or)+ cos (yw2t + Ov) I V=1 k R. v + r ? where, wi denotes the speed at the output torque / 2( / ) Step 4.4. A current misalignment angle is calculated based on the harmonic order r of the T cos (rw t + 6 ) main harmonic component r ' 2 '. The current misalignment angle / ? refers to an angle difference between the current phases output by the two inverters. Specifically, it is the quotient / ? = — of n / 2 and the harmonic order r of the main harmonic component, Step 4.5. Based on the transition currents 1.2( / ), ^1,9( / ), ■( / ), ^22( / ), ^23( / ), and (I) output by the two inverters in the step 4.2 and the current misalignment angle 0 calculated in the step 4.4, the three-phase (ABC) currents output by the two inverters are determined. In order to ensure the torque output capability of the motor, the current amplitude ^ma is increased by 5% on the basis of the amplitude Imaxo of the transition current, =1.05 Imaxo. The phases of the three-phase (ABC) currents output by the first inverter are / / / 2 ahead of the transition currents AXO, Ab(0, and Ac(<)> and the phases of the three-phase (ABC) currents output by the second inverter are 0 / 2 behind the transition currents AXX A«( / ), and Ad / ). 1 B hA (t) = 1.05 / sin —n Prt + 0(,+ — r+i x / max 1 q p r v M') = 105+ ( 2 B sin — n Pt — 77 + 0, 4— K ^30 P >3 O 2 / 1C.(O=L()5+ ( 7t 2 0 sin\ —n P t + — 7t + 0,+— (30 p 3 2 / ,,( / ) = 1.057 sin\ — n Pt + 0n | 2A x / max y 3Q p r 0 2 J [ TT ) < / .,( / ) = !.05 / sin — n Pt^-v + 0, 2B \ J max I gg p r 3 ( TF S Pr (t) = 1.051 sin\ —n Pt a—& + 0n 2C y / max y 30 p r 3 where, I\a, Iib, and he denote the three-phase (ABC) currents output by the first inverter; Pa, Izb, and / 2c denote the three-phase (ABC) currents output by the second inverter; np denotes the speed at the operation point P(np,Tpy, and Pr denotes the number of the rotor pole pairs. The three-phase currents Pa, Pb, he, Pa, Pb, and he are stored in the first DSP controller and the second DSP controller respectively, and the first DSP controller and the second DSP controller control the first inverter and the second inverter to output the three-phase currents. Step 5. Based on the division result of the constant-torque range in the step 3, a control method is determined for the second or low-speed overload range. Due to the high demand of the second or low-speed overload range for the motor torque, all the motor units 1 work together. That is, for any operation point Piippijpi) in the second or low-speed overload range, all the power electronic switches and all the winding electronic switches are closed. Then, simulation is performed through the finite element software to calculate the three-phase (ABC) currents output by the two inverters when the motor operates at the point 7,,2). Step 6. This step is concurrent with the step 3, based on the result acquired in the step 2, a detailed division of the constant-power range is performed. Considering the complex operating conditions in the constant-power range, the constant-torque range is divided into six sub-ranges, including third to eighth ranges. The third range serves as a high-speed low-load range, and is denoted as S3. The fourth range serves as a high-speed high-efficiency range, and is denoted as S4. The fifth range serves as a high-speed load range, and is denoted as S5. The sixth range serves as a double-efficiency range, and is denoted as Sg. The seventh range serves as a high-speed overload range, and is denoted as S7. The eighth range serves as a high-speed flux weakening range, and is denoted as Ss. The maximum value of the torque at the operation point V(np,Tp) in the constantpower range does not exceed the corresponding torque on the outer characteristic curve g. Step 6.1. The fourth or high-speed high-efficiency range is delimitated. Step 6.1.1. Simulation is performed to calculate an efficiency map when a single control module is operating. Only one of the first power electronic switch and the second power electronic switch is closed, and all the winding electronic switches are closed. On this premise, the efficiency map of the motor operation is acquired through simulation using the finite element software, and a maximum efficiency Tjmax of the motor when a single control module is operating is determined based on the efficiency map. Step 6.1.2. A boundary efficiency tjb is calculated based on the maximum efficiency tjmax of the motor when a single control module is operating. The boundary efficiency r / b is 95% of the n = 0 95 / 7 maximum efficiency of the motor when a single control module is running, ,b ' ,max. Step 6.1.3. The fourth or high-speed high-efficiency range is delimitated based on the boundary efficiency tjb. The fourth or high-speed high-efficiency range is a part of the constantpower range surrounded by the boundary efficiency tp,. In other words, if the efficiency rjp at any operation point P(np,Tp) in the constant-power range satisfies: tiP>Tib, then the point P(np,Tp) belongs to the fourth or high-speed high-efficiency range. That is, 4 * and After the fourth or high-speed high-efficiency range is delimitated, a highest-speed point E and a lowest-speed point D of the fourth or high-speed high-efficiency range are determined, as shown in FIG. 20. Step 6.2. Based on the result acquired in the step 6.1, the eighth or high-speed flux weakening range is delimitated. Step 6.2.1: The highest-speed point E in the fourth or high-speed high-efficiency range is not unique. A lowest-torque point is selected from all the highest-speed points as a required highest-speed point E, and a straight line j perpendicular to the abscissa and passing through the required highest-speed point E is acquired. Step 6.2.2. The eighth or high-speed flux weakening range is delimitated. The abscissa, the curve g, and the straight line j enclose the eighth or high-speed flux weakening range. The speed in the eighth or high-speed flux weakening range is higher than the speed riE at the required highest-speed point E. Step 6.3. Based on the results acquired in the steps 6.1 and 6.2, the third or high-speed low-load range is delimitated. Step 6.3.1. Based on the result acquired in the step 6.1, the lowest-speed point D(ud,Td) in the fourth or high-speed high-efficiency range is not unique, where up and Td respectively denote the speed and torque at the lowest-speed point D(hd,Td). A lowest-torque point D(hd,Td) is selected from all the lowest-speed points as the required lowest-speed point D, and a straight line k perpendicular to the abscissa and passing through the required lowest-speed point D is acquired. Step 6.3.2. Based on the result acquired in the step 6.3.1, in the third or high-speed low-load range, a range where the speed is less than that at the point D is denoted as S31. In the constantpower range, a range where the torque is less than the torque Td at the point D and the speed is less than the speed no at the point D is determined as the third or the high-speed low-load range. That is, if the point P{np,Tp) satisfies: np<nD and Tp<Td, then the point P(np,Tp) belongs to the ^31 = >p (nv >T») I n,<n<-n:. ST'T<T,\ third or high-speed low-load range. In other words, 1 ■ 7 E Step 6.3.3. In the third or high-speed low-load range, a range where the speed is greater than the speed at the point D(hd,Td) and less than the speed at the point £( / ¾ Ze) is denoted as S32. S32 is enclosed by the abscissa, the straight line k, a lower boundary (a connecting line between the points E and Z)) of the fourth or high-speed high-efficiency range, and the straight line j. Step 6.3.4. Based on the steps 6.3.2 and 6.3.3, the third or high-speed low-load range is delimitated. The third or high-speed low-load range is a union of S31 and S32, that is, 3 “ 31 32. It is worth noting that the size of S31 and S32 that form the third or high-speed low-load range changes with motor characteristics, and for some motors, there may be an empty set. Step 6.4. This step is concurrent with the step 6.3. Based on the fourth or high-speed high-efficiency range delimitated in the step 6.1, the sixth or double-efficiency range is delimitated. The sixth or double-efficiency range is a double expansion range of the fourth or high-speed high-efficiency range. The expansion method is to keep the speed constant and expand the torque to twice the original value. That is to say, the speed in the sixth or double-efficiency range is the same as that in the fourth or high-speed high-efficiency range, with a highest-speed point G on the straight line j and a lowest-speed point F on the straight line k. For each operation point Pe(np6,Tp6) in the sixth or double-efficiency range, its torque is twice that at the operation point with the same speed nps in the fourth or high-speed high-efficiency range. Any point Pelm,!™) in the sixth or double-efficiency range is a double expansion point of a point in the fourth or high-speed high-efficiency range. When a point P(np,Tp) belongs to the sixth or double-efficiency range, the point , ( T \ ? wp’~k~ g54 P(n ,T s n' P satisfies: z ' and ' p’ p' 4, and “ denotes a point in the fourth or high-speed high-efficiency range, that is: e s^P(np,TP) ¢: > Step 6.5. Based on the results acquired in the steps 6.1 and 6.4, the fifth or high-speed load range is delimitated. Step 6.5.1. A range where the torque is less than the torque Tf at the point F and higher than the torque at the point D and the speed is less than the speed np at the point F is determined as the fifth or high-speed load range, denoted as S51. If the point P(np,Tp) satisfies: np<nF and Td<Tp<Tf, then the point P(np,Tp) belongs to the fifth or high-speed load range, that is: ^51 = I nb<n<nF,TD <Tp<Ff} Step 6.5.2. In the fifth or high-speed load range, A range where the speed is greater than that at the point F(kf,Tf) and less than that at the point G(nG,To) is denoted as S52. S52 is enclosed by the straight line k, the straight line j, an upper boundary of the fourth or high-speed high-efficiency range, and a lower boundary of the sixth or double-efficiency range. Step 6.5.3. Based on the steps 6.5.1 and 6.5.2, the fifth or high-speed load range is delimitated. S = S US' The fifth or high-speed load range is a union of S51 and S52, that is, 5 51 52. It is worth noting that the size of S51 and S52 that form the fifth or high-speed load range changes with motor characteristics, and for some motors, one or both of S51 and S52 may be empty sets. Step 6.6. Based on the results acquired in the steps 6.1 to 6.5, the seventh or high-speed overload range is delimitated. The seventh range is a range of the constant-power range that does not belong to the third, fourth, fifth, sixth, and eighth ranges. That is, the remaining range of the constant-power range forms the seventh range. If a point P(np,Tp) satisfies: , then the point P(nP,TP) belongs to the seventh or high-speed overload range, that is: As shown in FIG. 20, the six sub-ranges of the constant-power range are marked differently. The back slash range represents the third or high-speed low-load range, the dark gray range represents the fourth or high-speed high-efficiency range, the diagonal grid range represents the fifth or high-speed load range, the white range indicates the sixth or double-efficiency range, the dotted range represents the seventh or high-speed overload range, and the light gray range represents the eighth or high-speed flux weakening range. Step 7. Based on the division result of the constant-power range in the step 6, a control method is determined for the third or high-speed low-load range. The torque in the third or high-speed low-load range is lower than that in the fifth, sixth, and seventh ranges, and the speed in the third or high-speed low-load range is lower than that in the eighth range. Therefore, for any point P3(nP3,TP3) in the third or high-speed low-load range, where nP3 and TP3 respectively denote the speed and torque at any point in the third or high-speed low-load range, only one of the first power electronic switch and the second power electronic switch is closed, and all the winding electronic switches are closed. In addition, the main control principle for the third or high-speed low-load range is a principle of maximizing the system efficiency, and the motor adopts uprated operation, which means that the output torque is increased without changing the motor speed. The motor operates at an uprated operation point H3(hh3,Th3), and excess energy is fed back to the battery, thereby improving the system efficiency of the motor. For any point P3(nP3,TP3) in the third or highspeed low-load range, the control method includes the following steps. Step 7.1. The total system power consumption W3 at any point P3(nP3,in the third or highspeed low-load range is calculated. When the motor is operating at the point P3(nP3,TP3\ its efficiency is rjP3, then the total system power consumption W3 at the point P3(nP3,TP3) is 2u times a quotient of a product of the speed and the torque divided by a product of 60 and the efficiency, that 2?™ T pf -is, 3 Step 7.2. This step is concurrent with the step 7.1. The total system power consumption PF3 ’ at a transition point P3 '(np^Tp-y) is calculated. The transition point P3 \nP3,TP3-) is any point with the same speed as the point P3(nP3,TP3). When the motor operates at the transition point P3 ’(nP3',Tp3:), its efficiency is i]P3- and power generation efficiency is r / pg, then the total system power consumption at the transition point P3 \nP3 -,TP3 ) is the motor operation consumption minus yy =----------------------— 60 / 7 77 the efficiency fed back to the battery, that is, py pg Step 7.3. Based on the steps 7.1 and 7.2, a mathematical model is established for uprated operation. The mathematical model for uprated operation is the total system power consumption W3'’ at the transition point P3 ’ minus the total system power consumption W3 at the point P3, that is: f(Tp3.) = W3'^W3 Inn 3T v 2nn3(T3,-T3\ 2nn 3T 3 ______p3 p3 _____P^ \ P^_____P^ / _____p3 p3 ^P3- rlP3 60f1P3 Step 7.4. Based on the step 7.3, a set Sp3 is determined. Let the mathematical model for lP3' "P3 uprated operation be greater than 0, then the equation ^Pi' is yielded from 2nn -T 3, 2nn 3 (T y - T 3) Inn 3T, p3 p3 Pj \ pj Pj J p3 p3 the equation ^^py^ = ^py ^pg >q Therefore, Sp3 is a set of all ^Ipg-PP^T Pg~Vp3 -,- 1P3' *P3 transition points P- \nP3,Tp3 ) that satisfy ^py ^P3 It should be noted that for some points P3(nP3,TP3), Sp3 may be an empty set, which mainly depends on the power and specific performance of the motor. When Sp3 is an empty set, the motor does not adopt uprated operation, but direct operation. Step 7.5. Based on the step 7.4, when Sp3 may be a non-empty set and the motor adopts uprated operation, the uprated operation point / / 3(^3,Tm) in 5p3 is further determined. In order to maximize the system efficiency of the motor within the normal operating range of the motor units 1, the uprated operation point / / 3(^3, Pp) needs to belong to Sp3 and belong to the third or high-speed low-load range or the fourth or high-speed high-efficiency range. In addition, the torque Th3 at the uprated operation pointH3 needs to be greater than or equal to Tp3. That is to say, the uprated operation point / / 3(^3, Zra) needs to satisfy: 77 in T V n 3 y?H3’JH3 ) S^H3 (nH3,TH3)GS4 TH3 >tP3 It should be noted that if the point / / 3(^3,7^3) satisfying the condition is not unique in Sp3, among the points that satisfy the condition, a point with a maximum value of the mathematical model ' p3’' for uprated operation is taken. Step 7.6. Simulation is performed through finite element software to acquire the three-phase (ABC) currents output by the inverters when the motor operates at the point / / 3(^3, Th). Step 8. This step is concurrent with the step 7. Based on the division result acquired in the step 6, a control method is determined for the fourth or high-speed high-efficiency range. The torque in the fourth or high-speed high-efficiency range is lower than that in the fifth, sixth, and seventh ranges, and the speed in the fourth or high-speed high-efficiency range is lower than that in the eighth range. Therefore, for any point P^np4,TPP) in the fourth or high-speed high-efficiency range, only one of the first power electronic switch and the second power electronic switch is closed, and all the winding electronic switches are closed. In addition, the main control principle of the fourth or high-speed high-efficiency range is a principle of maximizing the system efficiency. Since the efficiency of the operation point in the fourth or high-speed high-efficiency range is higher than that in the third, fifth, seventh, and eighth ranges, the operation point P^np^Tp-x) operates normally. Simulation is performed through finite element software to acquire the three-phase (ABC) currents output by the inverters when the motor operates at ^4(^4,7p4). Step 9. This step is concurrent with the step 8. Based on the division result acquired in the step 6, a control method is determined for the fifth or high-speed load range. The torque in the fifth or high-speed low range is lower than that in the third and fourth ranges. Therefore, for any point Ps(nP5,Tp5) in the fifth or high-speed load range, all the power electronic switches and winding electronic switches are closed. In addition, the main control principle for the fifth or high-speed load range is a principle of maximizing the system efficiency, and the motor adopts uprated operation, which means that the output torque is increased without changing the motor speed. The motor operates at an uprated operation point and excess energy is fed back to the battery, thereby improving the system efficiency of the motor. For any point P4np<,Tpp in the fifth or high-speed load range, the control method includes the following steps. Step 9.1. The total system power consumption W5 at the point PPnpp.Tps) is calculated. When the motor is operating at the point P5(nP5,TP5), its efficiency is tjP5, then the total system power consumption Ws at the point ^5(^5, / ^5) is 2% times a quotient of a product of the speed nP5 and the Itih T w _ ^"pP P5 5 6O / 7 torque / „5 divided by a product of 60 and the efficiency, that is, p5 Step 9.2. This step is concurrent with the step 9.1. The total system power consumption Ws’ at a transition point P5 \nps,TPs) is calculated. The transition point P5 ’(nps,Tps) is any point with the same speed as the point PsPhn, Tpi). When the motor operates at the transition point P5 ’(np$, Tps), its efficiency is T / ps, then the total system power consumption at the transition point P5 ’(nP5,TP5) is the motor operation consumption minus the efficiency fed back to the battery, that is, 60 / ^, Step 9.3. Based on the steps 9.1 and 9.2, a mathematical model is established for uprated operation. The mathematical model for uprated operation is the total system power consumption IPs’ at the transition point P5 ’ minus the total system power consumption Ws at the point P5, that is: 2anp5(Tp5,-Tp5) 17mp5Tp5 6Qr!P5' vPg Step 9.4. Based on the step 9.3, a set Sps is determined. Let the mathematical model Ppg^py r 'pg~Pp5 r , \ lp5 py' for uprated operation be greater than 0, then the equation ^py is yielded. The transition point P5 ’(nPs,TPs) satisfies the condition. Therefore, Sps is a set of all transition points P5 \ivppJP) that satisfy the condition. It should be noted that for some points Ps(nps,,TP5), Sps may be an empty set, which mainly depends on the power and specific performance of the motor. When Sps is an empty set, in the fifth or high-speed load range, the motor does not adopt uprated operation, but direct operation. Step 9.5. When Sps is a non-empty set and the motor adopts uprated operation, the uprated operation point Hsinns, Ths) in Sps is further determined. In order to maximize the system efficiency of the motor within the normal operating range of the motor units, the uprated operation point Hs(nH5,THs) needs to belong to Sps and belong to the fifth or high-speed load range or the sixth or double-efficiency range. In addition, Ths needs to be greater than or equal to Tps. That is to say, the uprated operation point / / 5( / / / / 5,7 / / 5) needs to satisfy: It should be noted that if the point Hs(hhs,Ths) satisfying the condition is not unique in Sps, among the points that satisfy the condition, a point with a maximum value of the mathematical model ' py ' for uprated operation is taken. Step 9.6. Based on the result acquired in the step 9.5, simulation is performed through finite element software to acquire the three-phase (ABC) currents output by the inverters when the motor operates at the point HA.m-pApA- Step 10. This step is concurrent with the step 9. Based on the division result acquired in the step 6, a control method is determined for the sixth or double-efficiency range. The torque in the sixth or double-efficiency range is lower than that in the third, fourth, and fifth ranges. Therefore, for any operation point PAnP(,, Tp&) in the sixth or double-efficiency range, all the power electronic switches and winding electronic switches are closed. In addition, the main control principle of the sixth or double-efficiency range is a principle of maximizing the system efficiency. Since the efficiency of the operation point in the sixth or double-efficiency range is higher than that in the third, fifth, seventh, and eighth ranges, the operation point PAnPp, TP&) operates normally. Simulation is performed through finite element software to acquire the three-phase (ABC) currents output by the inverters when the motor operates at Petypt, T^). The outputs of the inverters are controlled by the DSP controllers. Step 11. This step is concurrent with the step 10. Based on the division result acquired in the step 6, a control method is determined for the seventh or high-speed overload range. The torque in the seventh or high-speed overload range is lower than that in the third, fourth, and fifth ranges. Therefore, for any point P7(nP7,Tp7) in the seventh or high-speed overload range, all the power electronic switches and winding electronic switches are closed. In addition, the main control principle of the seventh or high-speed overload range is a principle of maximizing the system efficiency. Therefore, the control strategy for the seventh range involves the coordination of the two control modules. The output currents of the first inverter and the second inverter are different, and the N / 2 motor units 1 work at two operation points. The first control module operates the motor at the point P7i(nP7i,TP7i), and the second control module operates the motor at the point P72(nP72,TP72). For any point P^nP7,TP7) in the seventh or high-speed overload range, the control method includes the following steps. Step 11.1. A collaborative operation power function is established. There are two different points in the seventh or high-speed overload range, namely a first transition point P71 (nP7r,TP7v) and a second transition point ^72 (^72 ,^72 ). The speed at these two transition points is equal to the speed at the operation point P7(nP7, Tpt) in the seventh range, and the sum of the torques at these two transition points is Tp7, that is, w „ r = n = "p71 ,lp72 rlP7 T ’+T ' = T p71 ^72 P7 . A set of transition points that satisfy the condition is denoted as Spt. A total system power consumption W71 at the first transition point P7i(nP7v,TP7v) is 2% times a quotient of a product of the speed nP7i and the torque TP7v divided by a product of 60 and the w_ ’ = 1^1121111211 1 6 On ' efficiency, that is, ': . Similarly, the total system power consumption W7T at the second transition point P72 ’ ( / / / ,72’, / ),72-) is 2% times a quotient of a product of the speed and the w , = ^"" / >72 1p72 '72 ' torque divided by a product of 60 and the efficiency, that is, ^72 The collaborative operation power function is a sum of the total system power consumption FFzr at the first transition point P71 (nP7v,TP7r) and the total system power consumption W72’ at the second transition point P72 (^72^72^, that is: W, (7,,, ’.Tr,, ') = »'„ '+W„' _ p7\ + p!2 ^p72 pl\ p72 IV (T r T Step 11.2. Based on the collaborative operation power function 7 \ ’ ?72 / determined in the step 11.1, two operation points P7i(nP7i,TP7i) and ^72( / ^72, / ),72) are determined. The total system power consumption at the first operation point P?i( / / p7i, / ),71) is denoted as W 71, and the total system power consumption at the second operation point ^72( / ^72, / ^72) is W denoted as 72. Therefore, the collaborative operation power function of the two operation points Tjy 1 7' \ TXT' T / 7 P7i(nP7i,TP7i) and P72(nP72,TP72) is 7' 7’71’ p72 / = 71 + 72 . When the two operation points IF (t ,t 3 satisfy the condition that their collaborative operation power function 7\ / ’71’ ^72 / is equal to JF (7 j) a minimum value of the collaborative operation power function 7' ''7I ' ''7: 7 of the two transition points P7i(nP7v,TP7v) and P72(nP72-,TP72:), the motor operates at these two operation points P?i( / / p7i, / ),71) and ^72( / ^72, / ),72). The operation points P71 ( / / .,71, / p7i) and ^72( / ^72, / ),72) are respectively controlled by the first control module and the second control module. Step 11.3. Based on the point ^71( / / ,71, / ),71) determined in the step 11.2, the output current of the first inverter is determined. At this point, all the winding electronic switches are closed, and only the first power electronic switch is closed. Simulation is performed through finite element software to acquire the output current of the first inverter. Meanwhile, the output current of the second inverter is determined. At this point, all the winding electronic switches are closed, and only the second power electronic switch is closed. Simulation is performed through finite element software to acquire the output current of the second inverter. Step 12. This step is concurrent with the step 11. Based on the division result acquired in the step 6, a control strategy is determined for the eighth or high-speed flux weakening range. The speed in the eighth or high-speed flux weakening range is higher than that in the other ranges. Therefore, for any point Pztnp^Tp*) in the eighth or high-speed flux weakening range, all the power electronic switches are closed and all the winding electronic switches are closed. Simulation is performed through finite element software to acquire the output current of the inverter when the motor operates 2&P%(np%,Tp%). In summary, according to the characteristics of each range, the motor includes 8 operating modes. 1. When the motor operates in the first or low-speed load range, both the first power electronic switch and the second power electronic switch are closed. Arbitrarily K switches from the first winding electronic switch to the (A / 2)-th power electronic switch are closed. Arbitrarily K switches from the (A72+l)-th winding electronic switch to the 7V-th power electronic switch are closed. The two inverters output currents of the same frequency and amplitude but different phases. The motor units 1 in the two parts operate at the same operation point. Through the torque valley compensation design, the torque ripple of the motor in this range is significantly reduced and the torque quality is improved. 2. When the motor operates in the second or low-speed overload range, all the power electronic switches and winding electronic switches are closed. The two inverters output currents of the same frequency, amplitude, and phase. The motor units 1 in the two parts work at the same operation point, and the torques output by the motor units 1 are algebraically superposed. The design effectively improves the peak torque of the motor and enhances the torque output capability of the motor. 3. When the motor operates in the third or high-speed low-load range, only one of the first power electronic switch and the second power electronic switch is closed, and all the winding electronic switches are closed. The motor units 1 adopt uprated operation, effectively improving the working efficiency and overall operating efficiency of the motor. 4. When the motor operates in the fourth or high-speed high-efficiency range, only one of the first power electronic switch and the second power electronic switch is closed, and all the winding electronic switches are closed. The A72 motor units operate normally, improving the operating efficiency of the motor. 5. When the motor is running in the fifth or high-speed load range, all the power electronic switches and winding electronic switches are closed. All the motor units 1 adopt uprated operation, effectively improving the working efficiency and system operation efficiency of the motor. 6. When the motor is running in the sixth or double-efficiency range, all the power electronic switches and winding electronic switches are closed. All the motor units 1 operate normally, and the motor operates efficiently. 7. When the motor is running in the seventh or high-speed overload range, all the power electronic switches and winding electronic switches are closed. The two inverters output currents. The motor units 1 in the two parts operate at different operation points, with their torques algebraically superposed, greatly improving the operating efficiency of the motor. 8. When the motor operates in the eighth or high-speed flux weakening range, all the power electronic switches and winding electronic switches are closed. All the motor units 1 operate normally, and the motor operates at a high speed. The series of detailed descriptions listed above are only specific illustration of feasible implementations of the present disclosure, rather than limiting the claimed scope of the present disclosure. All equivalent manners or changes made without departing from the technical spirit of the present disclosure should be included in the claimed scope of the present disclosure.

Claims

What is claimed is:

1. A more-poles fewer-slots unitized permanent-magnet in-wheel motor, characterized by comprising ^identical motor units (1) evenly distributed along a circumference of a radial section, wherein each of the motor units (1) comprises 1 / N outer rotors (2), MN inner stators (3), and MN centralized windings (4); the inner stators (3) are coaxially nested inside the outer rotors (2), and wound with the centralized windings (4); the centralized windings (4) in each of the motor units (1) are three-phase symmetrical and identically distributed; the outer rotors (2) comprise a rotor core (2.3); 2a permanent magnet groups (2.4) are evenly distributed along a circumference of the rotor core (2.3); each of the permanent magnet groups (2.4) comprises a first rectangular permanent magnet (2.4.1), a second rectangular permanent magnet (2.4.2), and an arc-shaped permanent magnet (2.4.3); the first rectangular permanent magnet (2.4.1) and the second rectangular permanent magnet (2.4.2) are identically structured and each provided with a rectangular radial section; the first rectangular permanent magnet and the second rectangular permanent magnet form a V-shaped arrangement, with an opening facing an air gap and with inner and outer oblique directions formed by rectangular length directions, outside the arc-shaped permanent magnet (2.4.3); the first rectangular permanent magnet and the second rectangular permanent magnet are symmetrical about a centerline of the arc-shaped permanent magnet (2.4.3) in a diametral direction; the first rectangular permanent magnet (2.4.1) has a magnetization direction perpendicular to a length direction of the first rectangular permanent magnet (2.4.1), the second rectangular permanent magnet (2.4.2) has a magnetization direction perpendicular to a length direction of the second rectangular permanent magnet (2.4.2), and the arc-shaped permanent magnet (2.4.3) has a magnetization direction the same as a direction of the centerline of the arcshaped permanent magnet (2.4.3); in a same one of the permanent magnet groups (2.4), the magnetization directions of the first rectangular permanent magnet (2.4.1), the second rectangular permanent magnet (2.4.2), and the arc-shaped permanent magnet (2.4.3) simultaneously point towards or away from the air gap, while the magnetization directions of each two adjacent ones of the permanent magnet groups (2.4) are opposite; and a number Pr of rotor pole pairs, a number Ns of stator slots, a number m of motor phases, a slot-pitch angle r, and the number N of the motorunits simultaneously satisfy:__^-<1.5 TPr =Na Ns= mNb 2Pr^ = eNct = d*2jt c _ and N=2i , wherein i, a, b, c, d, and e are positive integers.

2. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim1, characterized in that the 1 / N outer rotors (2) are each axially divided into M identical rotor segments, 20 mm< / e / A / <120 mm, wherein / -denotes an axial length of the motor; and the M rotor segments are arranged in a way that the rotor segments rotate one mechanical misalignment angle in sequence along a same rotation direction.

3. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 2, characterized in that the mechanical misalignment angle is determined by:step 1) assigning an initial mechanical misalignment angle as 0;step 2) simulating a torque waveform output by the motor, and calculating an initial torque ripple;step 3) performing Fourier decomposition on the torque waveform, calculating a harmonic order k of a highest-amplitude harmonic component, calculating a transitional mechanical71a, =-----1 Mfcpmisalignment angle r , simulating a torque waveform at the transitional mechanicalmisalignment angle «i, and calculating a transitional torque ripple; andstep 4) comparing the transitional torque ripple with the initial torque ripple; determining that the transitional mechanical misalignment angle «i as the mechanical misalignment angle of the rotor segments if the transitional torque ripple is less than the initial torque ripple; and if not, assigning the transitional mechanical misalignment angle «i to the initial mechanical misalignment angle, and repeating the steps 2) and 3).

4. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 1, characterized in that each arc-shaped permanent magnet (2.4.3) comprises a radial section surrounded by an outer long side (2.4.3.1), an inner long side (2.4.3.2), and two short sides (2.4.3.3), of the arc-shaped permanent magnet; an arc center of the outer long side (2.4.3.1) and the inner long side (2.4.3.2) of the arc-shaped permanent magnet is the same as a center of the outer rotor (2); the short side (2.4.3.3) of the arc-shaped permanent magnet is in a direction the same as a diametral direction of the outer rotor (2); the inner long side (2.4.3.2) of the arc-shaped permanent magnet forms a sine curve fimax therein £ f and fimax is anamplitude; and when 0i=3ti / 2, a point / i(3tt / 2) is located on an inner surface of the outer rotor (2).

5. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 4, characterized in that the first rectangular permanent magnet (2.4.1) and the second rectangular permanent magnet (2.4.2) each are provided with an inner magnetic barrier (2.5) at an end close to the air gap and provided with an outer magnetic barrier (2.6) at an end away from the air gap; and two ends of each arc-shaped permanent magnet (2.4.3) in a tangential direction each are provided with a virtual slot (2.7) that becomes a part of the air gap.

386. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 5, characterized in that the inner magnetic barrier (2.5) comprises a pentagonal radial section surrounded by a first side, a second side, a third side, a fourth side, and a fifth side; the first side is an extended side of a long side of the first rectangular permanent magnet (2.4.1) or the second rectangular permanent magnet (2.4.2) close to the air gap; the second side is an arc side coaxial with the outer rotor (2); the third side is located on a radius of the outer rotor (2); the fourth side is parallel with the first side and located outside the first side; a distance between the first side and the fourth side is less than a width of the first rectangular permanent magnet (2.4.1) or the second rectangular permanent magnet (2.4.2); and the fifth side coincides with a short side of the first rectangular permanent magnet (2.4.1) or the second rectangular permanent magnet (2.4.2) close to the air gap.

7. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 5, characterized in that the outer magnetic barrier (2.6) comprises a pentagonal radial section surrounded by a first side, a second side, a third side, a fourth side, and a fifth side; the first side is an extended side of a long side of the first rectangular permanent magnet (2.4.1) or the second rectangular permanent magnet (2.4.2) close to the air gap; the second side is located on a radius of the outer rotor (2); the third side is an arc side coaxial with the outer rotor (2); the fourth side is parallel with the first side and located outside the first side; a distance between the first side and the fourth side is less than a width of the first rectangular permanent magnet (2.4.1) or the second rectangular permanent magnet (2.4.2); and the fifth side coincides with a short side of the first rectangular permanent magnet (2.4.1) or the second rectangular permanent magnet (2.4.2) away from the air gap.

8. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim . . 1 • 1 1 / ) = / sinl0-,) ,6, characterized in that the virtual slot (2.7) forms a sine curve v 7 mx \ on the radial section, wherein 02=[7t / 2,7t], and / mav is an amplitude; when 02=7t / 2, the inner long side (2.4.3.2) and the short side (2.4.3.3) intersect at a point / 2(71 / 2); and when ()2=71, a radius passing through an intersection point of the second side and the third side of the inner magnetic barrier (2.5) intersects with the inner surface of the outer rotor (2) at a point. / 2(71).

9. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 1, characterized in that the first rectangular permanent magnet (2.4.1) and the second rectangular permanent magnet (2.4.2) form a V-shaped angle ppm satisfying 4O°<0pm<65 and have a length wpm and a width hpm satisfying 2<wpmlhpm<A.

10. The more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 1, characterized in that a minimum width hpmin and a maximum width hpmax of the arc-shaped 39permanent magnet (2.4.3) in a diametral direction satisfy 1.5<hpmax / hpmm<2.

11. A collaborative control system of the more-poles fewer-slots unitized permanent-magnet in-wheel motor according to claim 1, characterized by comprising a battery, two control modules, and N winding electronic switches, wherein the winding electronic switches each are configured to control on / off of the centralized windings (4) in a respective one of the motor units (1); each of the two control modules comprises a power electronic switch, a digital signal processor (DSP) controller, and an inverter in series; input terminals of the two power electronic switches are respectively connected to an output terminal of the battery; an output terminal of each inverter is connected to A72 of the winding electronic switches; and an output terminal of each of the centralized windings (4) is connected to the battery through a rectifier.

12. A control method of the collaborative control system according to claim 11, characterized by comprising the following steps:step 1) closing the two power electronic switches and the N winding electronic switches; simulating with an abscissa denoting a speed of the motor and an ordinate denoting a torque output by the motor to acquire an outer characteristic curve g of the motor; disconnecting one of the power electronic switches, and simulating to acquire an outer characteristic curve f of the motor; taking a highest speed corresponding to a highest torque on the outer characteristic curve / as a critical speed nb; and determining a range in which the speed of the motor at an operation point satisfies np<nb as a constant-torque range;step 2) dividing the constant-torque range into a first range and a second range, wherein the first range is a range in which the torque at the operation point satisfies Tp<Tb, and the second range is a range in which the torque at the operation point satisfies Tp>Tb, Tb being a critical torque that is N-2 times a peak torque Tx of one of the motor units (1); and a maximum torque at the operation point in the second range does not exceed a corresponding torque on the outer characteristic curve g; andstep 3) closing, when the operation point is in the first range, the two power electronic switchesand at least L J of the winding electronic switches of the N / 2 of the motor units (1) connected to each of the control modules, Tpi being the torque at the operation point in the first range, and outputting, by the two inverters, currents of a same amplitude but different phases; and closing, when the operation point is in the second range, the two power electronic switches and the N winding electronic switches.

13. The control method according to claim 12, characterized in that the step 3) further comprises: simulating to calculate a torque waveform output by the motor and transition currentsoutput by the two inverters when the of the winding electronic switches are closed; performing Fourier decomposition on the torque waveform, calculating a harmonic order r of a main harmonic component, and calculating a current misalignment angle , wherein three- / phase currents output by the two inverters have an amplitude of ^=1.05 / ^0, being an amplitude of the transition currents; a phase of the three-phase current output by a first one of the inverters is ^ / 2 ahead of a phase of the transition currents; and a phase of the three-phase current output by a second one of the inverters is ^ / 2 behind the phase of the transition currents.

14. A control method of the collaborative control system according to claim 11, characterized by comprising the following steps:step (I): closing the two power electronic switches and the N winding electronic switches; simulating with an abscissa denoting a speed of the motor and an ordinate denoting a torque output by the motor to acquire an outer characteristic curve g of the motor; disconnecting one of the power electronic switches, and simulating to acquire an outer characteristic curve f of the motor; taking a highest speed corresponding to a highest torque on the outer characteristic curve / as a critical speed nb; and determining a range in which the speed of the motor at an operation point satisfies np>nb as a constant-power range;step (II): dividing the constant-power range into third to eighth ranges, wherein a range in which an efficiency at the operation point satisfies h is defined as a fourth range, ip being a boundary efficiency of the motor when a single control module is operating; the abscissa, the outer characteristic curve g, and a straight line / passing through a highest-speed point E of the fourth range and perpendicular to the abscissa enclose the eighth range; a range in which the torque is less than the torque at a lowest-speed point D of the fourth range and the speed is less than the speed at the point!) is defined as a range S31; the abscissa, a straight line k passing through the point D and perpendicular to the abscissa, a lower boundary connected by the points D and E of the fourth range, and the straight line j enclose a range S32; a union of S31 and S32 forms a third range; a range in which the speed is the same as the speed in the fourth range and the torque is twice the torque in the fourth range is defined as a sixth range; the sixth range comprises a highest-speed point G on the straight line j and a lowest-speed point F on the straight line k; a range in which the torque is less than the torque at the point F but higher than the torque at the point D and the speed is less than the speed at the point F is defined as a range S51; the straight line k, the straight line j, an upper boundary connected by the points D and E of the fourth range, and a lower boundary connected by the points F and G of the sixth range enclose a range S52; a union of S51 and S52 forms a fifth range; and a remaining range in the constant-power range forms a seventh 41range; andstep (III): closing, when the operation point is in the third range, only one of the two power electronic switches and all the N winding electronic switches, and enabling uprated operation of the motor by increasing the torque without changing the speed of the motor; closing, when the operation point is in the fourth range, only one of the two power electronic switches and all the N winding electronic switches; closing, when the operation point is in the fifth range, the two power electronic switches and the TV winding electronic switches, and enabling uprated operation; closing, when the operation point is in the sixth range, the two power electronic switches and the N winding electronic switches; closing, when the operation point is in the seventh range, the two power electronic switches and the N winding electronic switches, and outputting, by the two inverters, different currents; and closing, when the operation point is in the eighth range, the two power electronic switches and the N winding electronic switches.

15. The control method according to claim 14, characterized in that in the step (III), the enabling the uprated operation in the third range comprises: determining a transition point P3 ’(nP3,TP3) with a same speed as an operation point ^3(^3,¾) in the third range, forming a set1 py pySp3 with points satisfying ^py ^P3 , selecting a point that belongs to thethird or fourth range and has a torque greater than or equal to Tp3 from the set Sp3 as an uprated operation point Hi(tih3, Th3\ simulating to acquire three-phase currents output by the inverters when the motor is operating at the uprated operation point / / 3(^3, Thi), and feeding back excess energy to the battery, wherein the uprated operation in the fifth range is the same as the uprated operation in the third range; np3, Tp3, and t]P3 denote the speed, the torque, and the efficiency at the operation point ^3( / ^3,^3), respectively; TP3’ and i]P3- denote the torque and an efficiency at the transition point, respectively; and Tjpg denotes a power generation efficiency.

16. The control method according to claim 14, characterized in that the step (III) further comprises: determining, in the seventh range, two transition points with the same speed as the operation point in the seventh range and with a sum of the torques at the two transition points equal to the torque at the operation point in the seventh range; calculating a sum of total system power consumptions at the two transition points; and if a sum of total system power consumptions of two operation points is equal to a minimum value of the sum of the total system power consumptions at the two transition points, controlling the two operation points by the two control modules respectively.

17. The control method according to claim 16, characterized by further comprising: closing the N winding electronic switches and only a first one of the power electronic switches, simulatingto acquire a current output by a first one of the inverters, and controlling a first operation point of the two operation points; and closing the N winding electronic switches and only a second one of the power electronic switches, simulating to acquire a current output by a second one of the inverters, and controlling a second operation point of the two operation points.