CN114744939B - An online compensation method and device for the dead zone voltage of the current loop of a permanent magnet synchronous motor - Google Patents

Abstract

The invention belongs to the technical field of dead zone voltage compensation, and discloses an online compensation method and equipment for dead zone voltage of a current loop of a permanent magnet synchronous motor, wherein the equipment comprises a q-axis current PI controller, a d-axis current PI controller, a q-axis voltage estimation module, a d-axis voltage estimation module, a q-axis error voltage fractional order PI controller and a d-axis voltage fractional order PI controller; the q-axis current PI controller is used for receiving the q-axis reference current value and the q-axis current and calculating a q-axis reference voltage; the d-axis current PI controller is used for receiving the d-axis reference current value and the d-axis current value and calculating d-axis reference voltage; the input of the q-axis voltage estimation module is q-axis current, and the output of the q-axis voltage estimation module is q-axis estimated voltage; the input of the q-axis error voltage fractional PI controller is the difference between the q-axis reference voltage and the q-axis estimated voltage, and the output is the q-axis compensation voltage. The invention greatly improves the response performance of the current loop of the permanent magnet synchronous motor in low current response.

Description

An online compensation method and device for the dead zone voltage of the current loop of a permanent magnet synchronous motor

Technical Field

The present invention belongs to the technical field related to dead zone voltage compensation, and more specifically, relates to an online compensation method and device for the dead zone voltage of a permanent magnet synchronous motor current loop.

Background technique

Permanent magnet synchronous motors have many advantages, making them widely used. The inverter has a dead zone effect because its upper and lower bridge arms must avoid direct conduction. The dead zone effect causes current distortion and torque fluctuations, especially at low speeds, where the impact is particularly obvious and severe.

With the widespread application of servo systems in industrial robots, intelligent robots and high-end CNC machine tools, these equipment require the servo systems to have very high low-speed performance. The performance of the servo system at low speed greatly affects the positioning accuracy and response performance of various high-end equipment.

In order to obtain a phase current waveform with better sinusoidality, the dead zone effect is generally eliminated by dead zone compensation. Since the interference voltage caused by the dead zone is closely related to the polarity of the phase current, the wrong compensation caused by the wrong judgment of the phase current polarity will cause greater distortion of the current. Therefore, the key to dead zone compensation lies in the detection of the phase current zero crossing point. For example, patents 201010566483.3 and 201010268341.9 directly judge the phase current polarity based on the detection value obtained by A/D conversion, which is easily affected by sampling noise and has a high probability of wrong compensation. The traditional method judges the phase current polarity based on the position of the output voltage vector, avoiding the adverse effects of current sampling noise. At the same time, it proposes a strategy of compensating only one phase output voltage in a specific current region. However, the calculation method of "the angle between the voltage vector and the induced electromotive force" is only applicable to the steady-state process, and is not applicable to servo systems with frequent fluctuations in speed and current.

Dead zone compensation algorithms are currently classified into three categories: 1. Compensation algorithm based on dead zone voltage model: it is necessary to directly or indirectly obtain the current polarity, which is easy to cause miscompensation. 2. Algorithm based on dead zone time compensation: because the dead zone time reduces the action time of the upper bridge arm or the lower bridge arm, the bridge arm action time is directly recalculated, which also requires direct or indirect acquisition of current polarity and causes miscompensation. 3. Compensation algorithm based on observer: mainly based on model observation method, using feedforward or feedback to compensate dead zone voltage, such as patent 201110440513 uses a voltage closed-loop control compensation method, but requires a sensor to detect line voltage, which increases hardware cost.

Summary of the invention

In view of the above defects or improvement needs of the prior art, the present invention provides an online compensation method and device for the dead-zone voltage of the current loop of a permanent magnet synchronous motor. The device can monitor the d-axis inverter error voltage and the q-axis inverter error voltage in real time, and compensate them in real time to the permanent magnet synchronous motor current loop control system through a fractional-order PI controller. The fractional-order PI controller has the characteristics of fast response speed and good robustness, and can make the dead-zone error voltage converge quickly, thereby greatly improving the response performance of the permanent magnet synchronous motor current loop at low current response.

To achieve the above-mentioned object, according to one aspect of the present invention, there is provided an online compensation device for the dead zone voltage of the current loop of a permanent magnet synchronous motor, the device comprising a current control device and a fractional-order dead zone voltage compensation device, the current control device comprising a q-axis current PI controller and a d-axis current PI controller; the dead zone voltage compensation device comprising a q-axis voltage estimation module, a d-axis voltage estimation module, a q-axis error voltage fractional-order PI controller and a d-axis voltage fractional-order PI controller;

The q-axis current PI controller is used to receive the q-axis reference current value Iqref and the q-axis current Iq, and calculate the q-axis reference voltage Vq_ref according to the received q-axis reference current value Iqref and the q-axis current Iq; the d-axis current PI controller is used to receive the d-axis reference current value Idref and the d-axis current value Id, and calculate the d-axis reference voltage Vd_ref according to the received d-axis reference current value Idref and the d-axis current value Id; the input of the q-axis voltage estimation module is the q-axis current Iq, and its output is the q-axis estimated voltage Vq; the input of the q-axis error voltage fractional-order PI controller is the q from the q-axis current PI controller The difference between the q-axis reference voltage Vq_ref and the q-axis estimated voltage Vq from the q-axis voltage estimation module is output as the q-axis compensation voltage ΔVq; the q-axis compensation voltage ΔVq is added to the q-axis reference voltage Vq_ref to obtain the q-axis actual command voltage Vq* for the inverter, and the d-axis compensation voltage ΔVd is added to the d-axis reference voltage to obtain the d-axis actual command voltage Vd* for the inverter. After the q-axis actual command voltage Vq* and the d-axis actual command voltage Vd* are transformed by IPARK, they are applied to the inverter through the SVPWM algorithm, and are output to the permanent magnet synchronous motor through the inverter to eliminate the influence of the inverter dead zone voltage error.

Furthermore, the q-axis current PI controller and the d-axis current PI controller are both common PI controllers, which are used to generate the q-axis reference voltage Vq_ref and the d-axis reference voltage Vd_ref, and the specific formulas thereof are:

_GPI ＝kp+ki/s

err_q＝iq _ref -iq

Vq_ref＝err_q*G _PI

err_d=id _ref -id

Vd_ref＝err_d*G _PI

Wherein, G _PI is the transfer function of the PI controller, kp is the proportional coefficient, ki is the integral coefficient, s is the Laplace operator, iq _ref is the q-axis reference current value, iq is the q-axis current value, id _ref is the d-axis reference current value, id is the d-axis current value, Vq_ref is the q-axis reference voltage, and Vd_ref is the d-axis reference voltage.

Furthermore, the q-axis voltage estimation module calculates the q-axis voltage according to the q-axis voltage model of the permanent magnet synchronous motor, and the calculation formula is:

Where: V _q (k-1) represents the q-axis voltage of the k-1th period, i _q (k-1) represents the q-axis current of the k-1th period, i _q (k-2) represents the q-axis current of the k-2th period, _id (k-1) represents the d-axis current of the k-1th period, _we (k-1) represents the electrical angular velocity of the k-1th period, Rs _is the motor resistance, _Lq is the q-axis inductance, _Ld is the d-axis inductance, _λf is the back electromotive force coefficient, and _Ts is the discrete period.

Furthermore, the calculation formula of the d-axis voltage estimation module is:

Wherein, _Vd (k-1) represents the d-axis voltage of the k-1th period, _id (k-1) represents the d-axis current of the k-1th period, _id (k-2) represents the d-axis current of the k-2th period, _iq (k-1) represents the q-axis current of the k-1th period, _we (k-1) represents the electrical angular velocity of the k-1th period, Rs _is the motor resistance, _Lq is the q-axis inductance, _Ld is the d-axis inductance, and _Ts is the discrete period.

Furthermore, the transfer functions of the q-axis error voltage fractional-order PI controller and the d-axis error voltage fractional-order PI controller are:

G _FOPI = kp + ki/s ^α

Where G _FOPI is the transfer function of the FOPI controller, kp is the proportional coefficient, ki is the integral coefficient, s is the Laplace operator, and α is the fractional order.

Furthermore, the estimated value of the q-axis error voltage in the k-th cycle is approximately equal to the q-axis error voltage in the k-1-th cycle, and the q-axis compensation voltage ΔVq can be calculated through the q-axis error voltage fractional-order PI controller:

err_Vq(k)≈err_Vq(k-1)=Vq_ref(k-1) _-Vq (k-1)

ΔVq(k)＝err_Vq(k)*G _FOPI

Wherein, err_Vq(k) is the q-axis error voltage of the k-th cycle, err_Vq(k-1) is the q-axis error voltage of the k-1-th cycle, Vq_ref(k-1) is the q-axis reference voltage of the k-1-th cycle; Vq(k-1) is the q-axis voltage of the k-1-th cycle, ΔVq(k) is the q-axis compensation voltage of the k-th cycle, and G _FOPI is the transfer function of the FOPI controller.

Furthermore, the estimated value of the d-axis error voltage in the kth cycle is approximately equal to the d-axis error voltage in the k-1th cycle, and the d-axis compensation voltage ΔVd can be calculated through the d-axis error voltage fractional-order PI controller:

err_Vd(k)≈err_Vd(k-1)＝Vd_ref(k-1)-Vd(k-1)

ΔVd(k)＝err_Vd(k)*G _FOPI

Where err_Vd(k) is the d-axis error voltage of the k-th cycle, err_Vd(k-1) is the d-axis error voltage of the k-1-th cycle, Vd_ref(k-1) is the d-axis reference voltage of the k-1-th cycle, Vd(k-1) is the d-axis voltage of the k-1-th cycle, ΔVd(k) is the d-axis compensation voltage of the k-th cycle, and G _FOPI is the transfer function of the FOPI controller.

Furthermore, the actual command voltage of the q-axis is:

Vq ^* (k)=Vq_ref(k)+ΔVq(k);

The actual command voltage of the d-axis is:

Vd ^* (k)=Vd_ref(k)+ΔVd(k).

According to another aspect of the present invention, a method for online compensation of a permanent magnet synchronous motor current loop dead zone voltage is provided, wherein the method uses the above-mentioned online compensation device for the permanent magnet synchronous motor current loop dead zone voltage to perform voltage compensation on the permanent magnet synchronous motor.

In general, compared with the prior art, the above technical solution conceived by the present invention, the online compensation method and device for the dead zone voltage of the permanent magnet synchronous motor current loop provided by the present invention mainly have the following beneficial effects:

1. The fractional-order PI controller has the characteristics of fast response speed and good robustness, which can make the dead zone error voltage converge quickly, thereby greatly improving the response performance of the permanent magnet synchronous motor current loop at low current response.

2. Adopt online dead zone voltage compensation and calculate the error voltage in real time. The compensation is accurate and fast. There is no need to detect the current polarity, thus avoiding the error compensation caused by polarity detection.

3. The device utilizes the fast response speed and good robustness of the fractional-order controller, and still has good compensation control performance under non-steady-state conditions such as servo systems where the speed and current are frequently fluctuating.

4. Compared with the dead zone compensation strategy using integer-order PI and not using PI, the q-axis current response of the present invention has a faster response speed and smaller torque pulsation.

BRIEF DESCRIPTION OF THE DRAWINGS

1 is a schematic structural diagram of an online compensation device for a permanent magnet synchronous motor current loop dead zone voltage provided by the present invention;

FIG2 is a schematic diagram of the overall structure of a conventional current loop PI control of a permanent magnet synchronous motor;

3 is a schematic diagram of the structure of the q/d axis error voltage fractional-order PI controller used in the present invention;

FIG4 is a comparison diagram of q-axis current step response using the present invention and using a conventional current loop PI control;

FIG5 is a comparison diagram of d-axis current response using the present invention and using a conventional current loop PI control;

(a) and (b) in FIG6 are comparison diagrams of the three-phase current response of the present invention and that of the common current loop PI control, respectively.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

Please refer to Figures 1 and 3. The present invention provides an online compensation device for the dead zone voltage of the current loop of a permanent magnet synchronous motor, the device comprising a current control device and a fractional-order dead zone voltage compensation device, the current control device comprising a q-axis current PI controller and a d-axis current PI controller. The dead zone voltage compensation device comprises a q-axis voltage estimation module, a d-axis voltage estimation module, a q-axis error voltage fractional-order PI controller and a d-axis voltage fractional-order PI controller.

The current control device is mainly used to generate a d-axis reference voltage and a q-axis reference voltage and control the current of the d-axis and q-axis. The q-axis current PI controller is used to receive the q-axis reference current value Iqref and the q-axis current Iq, and calculate the q-axis reference voltage Vq_ref based on the received q-axis reference current value Iqref and the q-axis current Iq. The d-axis current PI controller is used to receive the d-axis reference current value Idref and the d-axis current value Id, and calculate the d-axis reference voltage Vd_ref based on the received d-axis reference current value Idref and the d-axis current value Id.

The fractional-order dead-zone voltage compensation device is mainly used to calculate the d-axis error voltage and q-axis error voltage caused by the inverter, and to compensate for the dead-zone error voltage by means of the fast response and robustness of the fractional-order PI controller. The input of the q-axis voltage estimation module is the q-axis current Iq, and its output is the q-axis estimated voltage Vq. The input of the q-axis error voltage fractional-order PI controller is the difference between the q-axis reference voltage Vq_ref from the q-axis current PI controller and the q-axis estimated voltage Vq from the q-axis voltage estimation module, and its output is the q-axis compensation voltage ΔVq.

The q-axis compensation voltage ΔVq is added to the q-axis reference voltage Vq_ref to obtain the q-axis actual command voltage Vq* for the inverter, and the d-axis compensation voltage ΔVd is added to the d-axis reference voltage to obtain the d-axis actual command voltage Vd* for the inverter. The q-axis actual command voltage Vq* and the d-axis actual command voltage Vd* are transformed by IPARK, applied to the inverter through the SVPWM algorithm, and output to the permanent magnet synchronous motor through the inverter, which can effectively eliminate the adverse effects of current distortion, torque pulsation and slow response speed caused by the dead zone voltage error of the inverter.

The q-axis current PI controller and the d-axis current PI controller are both common PI controllers, which are used to generate the q-axis reference voltage Vq_ref and the d-axis reference voltage Vd_ref. The specific transfer function formula is:

_GPI ＝kp+ki/s

err_q＝iq _ref -iq

Vq_ref＝err_q*G _PI

err_d=id _ref -id

Vd_ref＝err_d*G _PI

Wherein, G _PI is the transfer function of the PI controller, kp is the proportional coefficient, ki is the integral coefficient, s is the Laplace operator, iq _ref is the q-axis reference current value, iq is the q-axis current value, id _ref is the d-axis reference current value, id is the d-axis current value, Vq_ref is the q-axis reference voltage, and Vd_ref is the d-axis reference voltage.

The q-axis voltage estimation module calculates the q-axis voltage according to the q-axis voltage model of the permanent magnet synchronous motor. The calculation formula is:

Where: V _q (k-1) represents the q-axis voltage of the k-1th period, i _q (k-1) represents the q-axis current of the k-1th period, i _q (k-2) represents the q-axis current of the k-2th period, _id (k-1) represents the d-axis current of the k-1th period, _we (k-1) represents the electrical angular velocity of the k-1th period, Rs _is the motor resistance, _Lq is the q-axis inductance, _Ld is the d-axis inductance, _λf is the back electromotive force coefficient, and _Ts is the discrete period.

The calculation formula of the d-axis voltage estimation module is:

Wherein, _Vd (k-1) represents the d-axis voltage of the k-1th period, _id (k-1) represents the d-axis current of the k-1th period, _id (k-2) represents the d-axis current of the k-2th period, _iq (k-1) represents the q-axis current of the k-1th period, _we (k-1) represents the electrical angular velocity of the k-1th period, Rs _is the motor resistance, _Lq is the q-axis inductance, _Ld is the d-axis inductance, and _Ts is the discrete period.

The transfer functions of the q-axis error voltage fractional-order PI controller and the d-axis error voltage fractional-order PI controller are:

G _FOPI = kp + ki/s ^α

Where G _FOPI is the transfer function of the FOPI controller, kp is the proportional coefficient, ki is the integral coefficient, s is the Laplace operator, and α is the fractional order.

The estimated value of the q-axis error voltage in the kth cycle is approximately equal to the q-axis error voltage in the k-1th cycle. The q-axis compensation voltage ΔVq can be calculated through the q-axis error voltage fractional-order PI controller:

err_Vq(k)≈err_Vq(k-1)=Vq_ref(k-1) _-Vq (k-1)

ΔVq(k)＝err_Vq(k)*G _FOPI

Wherein, err_Vq(k) is the q-axis error voltage of the k-th cycle, err_Vq(k-1) is the q-axis error voltage of the k-1-th cycle, Vq_ref(k-1) is the q-axis reference voltage of the k-1-th cycle; Vq(k-1) is the q-axis voltage of the k-1-th cycle, ΔVq(k) is the q-axis compensation voltage of the k-th cycle, and G _FOPI is the transfer function of the FOPI controller.

The estimated value of the d-axis error voltage in the kth cycle is approximately equal to the d-axis error voltage in the k-1th cycle. The d-axis compensation voltage ΔVd can be calculated through the d-axis error voltage fractional-order PI controller:

err_Vd(k)≈err_Vd(k-1)＝Vd_ref(k-1)-Vd(k-1)

ΔVd(k)＝err_Vd(k)*G _FOPI

Where err_Vd(k) is the d-axis error voltage of the k-th cycle, err_Vd(k-1) is the d-axis error voltage of the k-1-th cycle, Vd_ref(k-1) is the d-axis reference voltage of the k-1-th cycle, Vd(k-1) is the d-axis voltage of the k-1-th cycle, ΔVd(k) is the d-axis compensation voltage of the k-th cycle, and G _FOPI is the transfer function of the FOPI controller.

The actual command voltage of the q-axis is:

Vq ^* (k)＝Vq_ref(k)+ΔVq(k)

The actual command voltage of the d-axis is:

Vd ^* (k)=Vd_ref(k)+ΔVd(k).

The present invention is further described in detail with a specific embodiment as follows: First, a simulation model of the dead zone compensation of the permanent magnet synchronous motor servo system is established in matlab/simulink, and the motor model parameters are R=0.3872 ohm, L=4.37mH, J=0.02727kg.m^2, B=5.027N.m.s, p=5, and the dead zone time is 2.2us; the SVPWM driving mode with Idref=0 is used to perform the Iq current step response experiment of the current closed loop, and given Iqref =1.536A, where the parameters Kp and Ki of the q-axis current PI controller and the d-axis current PI controller are adjusted by canceling the electromagnetic link of the permanent magnet synchronous motor and the given cross-over frequency. The given cross-over frequency is 5000Hz, and the adjusted Kp=1.5008 and Ki=221.3782; the parameters kp, ki, and α of the q-axis error voltage fractional-order PI controller and the d-axis error voltage fractional-order PI controller are adjusted using the particle swarm optimization algorithm, and the adjusted fitness index adopts the q-axis current response ITAE standard as follows:

The parameters of the q-axis error voltage fractional-order PI controller tuned by the particle swarm algorithm are kp=1.6938, ki=503.6825, α=-0.7226. At the same time, the parameters of the d-axis error voltage fractional-order PI controller tuned by the particle swarm algorithm are kp=2.1861, ki=503.68, α=-0.6514.

In order to compare the dead zone compensation effect of this embodiment, the q-axis current step response of the ordinary current loop PI control shown in Figure 2 is used for comparative study. The ordinary current loop PI control described in Figure 2 also adopts Iqref＝1.536A, Idref＝0A, and SVPWM driving mode.

As shown in Figure 4, the results of comparing the q-axis current response of the fractional-order PI online compensation of the present invention and the uncompensated ordinary current loop PI control show that the response speed of the fractional-order PI online compensation of the present invention is much faster than that of the uncompensated ordinary current loop PI control. From the current pulsation situation, the q-axis current pulsation of the uncompensated ordinary current loop PI control is very severe. The fractional-order PI online compensation method plays a very good role in suppressing the torque pulsation, and basically there is only a slight pulsation.

As shown in FIG5 , the results of comparing the d-axis current responses of the fractional-order PI online compensation and the uncompensated ordinary current loop PI control of the present invention show that, from the perspective of the d-axis current fluctuation, the d-axis current fluctuation of the uncompensated ordinary current loop PI control is very severe, and the fractional-order PI online compensation method is much stronger than the uncompensated ordinary current loop PI control in controlling Id=0, which proves that the fractional-order PI online compensation method has a good effect on suppressing d-axis current fluctuation.

As shown in Figure 6, the results of comparing the three-phase current responses of the fractional-order PI online compensation method of the present invention and the uncompensated ordinary current loop PI control are obtained. Since the dead zone effect mainly causes current distortion and zero-point current clamping, it is easy to observe that the zero-point current clamping time of the fractional-order PI online compensation method is much shorter than the zero-point current clamping time of the uncompensated ordinary current loop PI control; it can be seen that the uncompensated ordinary current loop PI control has undergone a large distortion relative to the ideal three-phase sine wave in the steady state, while the fractional-order PI online compensation method has only undergone a small distortion relative to the ideal three-phase sine wave in the steady state, which proves that the fractional-order PI online compensation method has a good effect on eliminating the zero-point current clamping time and current distortion.

The present invention also provides an online compensation method for the dead zone voltage of the current loop of a permanent magnet synchronous motor. The method uses the online compensation device for the dead zone voltage of the current loop of the permanent magnet synchronous motor as described above to perform voltage compensation on the permanent magnet synchronous motor.

It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims (9)

1. An on-line compensation device for dead zone voltage of a current loop of a permanent magnet synchronous motor is characterized in that:

The device comprises a current control device and a fractional order dead zone voltage compensation device, wherein the current control device comprises a q-axis current PI controller and a d-axis current PI controller; the dead zone voltage compensation device comprises a q-axis voltage estimation module, a d-axis voltage estimation module, a q-axis error voltage fractional order PI controller and a d-axis voltage fractional order PI controller;

the q-axis current PI controller is used for receiving the q-axis reference current value Iqref and the q-axis current Iq and calculating a q-axis reference voltage Vq_ref according to the received q-axis reference current value Iqref and the q-axis current Iq; the d-axis current PI controller is used for receiving the d-axis reference current value Idref and the d-axis current value Id and calculating d-axis reference voltage Vd_ref according to the received d-axis reference current value Idref and d-axis current value Id; the input of the q-axis voltage estimation module is q-axis current Iq, and the output of the q-axis voltage estimation module is q-axis estimated voltage Vq; the input of the q-axis error voltage fractional PI controller is the difference between the q-axis reference voltage vq_ref from the q-axis current PI controller and the q-axis estimated voltage Vq from the q-axis voltage estimation module, and the output is the q-axis compensation voltage Δvq; the q-axis compensation voltage delta Vq is added to the q-axis reference voltage vq_ref to obtain the q-axis actual command voltage Vq of the inverter, the d-axis compensation voltage delta Vd is added to the d-axis reference voltage to obtain the d-axis actual command voltage Vd of the inverter, the q-axis actual command voltage Vq and the d-axis actual command voltage Vd are converted by IPARK and then are applied to the inverter through an SVPWM algorithm, and the voltage is output to the permanent magnet synchronous motor through the inverter to eliminate the influence caused by dead zone voltage errors of the inverter.

2. The on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor as claimed in claim 1, wherein: the q-axis current PI controller and the d-axis current PI controller are common PI controllers and are used for generating a q-axis reference voltage Vq_ref and a d-axis reference voltage Vd_ref, and the calculation formulas are as follows:

G_PI＝kp+ki/s

err_q＝iq_ref-iq

Vq_ref＝err_q*G_PI

err_d＝id_ref-id

Vd_ref＝err_d*G_PI

Wherein G _PI is a transfer function of the PI controller, kp is a proportional coefficient, ki is an integral coefficient, s is a laplace operator, iq _ref is a q-axis reference current value, iq is a q-axis current value, id _ref is a d-axis reference current value, id is a d-axis current value, vq_ref is a q-axis reference voltage, and vd_ref is a d-axis reference voltage.

3. The on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor as claimed in claim 1, wherein: the q-axis voltage estimation module calculates q-axis voltage according to a permanent magnet synchronous motor q-axis voltage model, and the calculation formula is as follows:

Wherein: v _q (k-1) represents the q-axis voltage of the kth-1 period, i _q (k-1) represents the q-axis current of the kth-1 period, i _q (k-2) represents the q-axis current of the kth-2 period, i _d (k-1) represents the d-axis current of the kth-1 period, w _e (k-1) represents the electrical angular velocity of the kth-1 period, R _s is the motor resistance, L _q is the q-axis inductance, L _d is the d-axis inductance, lambda _f is the back electromotive force coefficient, and T _s is the discrete period.

4. The on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor as claimed in claim 1, wherein: the calculation formula of the d-axis voltage estimation module is as follows:

Where V _d (k-1) represents the d-axis voltage of the (k-1) th cycle, i _d (k-1) represents the d-axis current of the (k-1) th cycle, i _d (k-2) represents the d-axis current of the (k-2) th cycle, i _q (k-1) represents the q-axis current of the (k-1) th cycle, w _e (k-1) represents the electrical angular velocity of the (k-1) th cycle, R _s is the motor resistance, L _q is the q-axis inductance, L _d is the d-axis inductance, and T _s is the discrete cycle.

5. An on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor as claimed in any one of claims 1 to 4, wherein: the transfer functions of the q-axis error voltage fractional order PI controller and the d-axis error voltage fractional order PI controller are:

G_FOPI＝kp+ki/s^α

Where G _FOPI is the transfer function of the FOPI controller, kp is the scaling factor, ki is the integration factor, s is the Laplacian, and α is the fractional order.

6. The on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor according to claim 5, wherein: the q-axis error voltage estimated value of the kth period is approximately equal to the q-axis error voltage of the kth-1 period, and the q-axis compensation voltage delta Vq can be calculated through the q-axis error voltage fractional order PI controller:

err_Vq(k)≈err_Vq(k-1)＝Vq_ref(k-1)-V_q(k-1)

ΔVq(k)＝err_Vq(k)*G_FOPI

Wherein err_vq (k) is the q-axis error voltage of the kth period, err_vq (k-1) is the q-axis error voltage of the kth-1 period, and vq_ref (k-1) is the q-axis reference voltage of the kth-1 period; vq (k-1) is the q-axis voltage of the kth cycle, Δvq (k) is the q-axis compensation voltage of the kth cycle, and G _FOPI is the transfer function of the FOPI controller.

7. The on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor according to claim 6, wherein: the d-axis error voltage estimated value of the kth period is approximately equal to the d-axis error voltage of the kth-1 period, and the d-axis compensation voltage delta Vd can be calculated through the d-axis error voltage fractional order PI controller:

err_Vd(k)≈err_Vd(k-1)＝Vd_ref(k-1)-Vd(k-1)

ΔVd(k)＝err_Vd(k)*G_FOPI

Where err_Vd (k) is the d-axis error voltage of the kth period, err_Vd (k-1) is the d-axis error voltage of the kth-1 period, vd_ref (k-1) is the d-axis reference voltage of the kth-1 period, vd (k-1) is the d-axis voltage of the kth-1 period, ΔVd (k) is the d-axis compensation voltage of the kth period, and G _FOPI is the transfer function of the FOPI controller.

8. The on-line compensation device for dead zone voltage of current loop of permanent magnet synchronous motor according to claim 7, wherein: the q-axis actual command voltage is:

Vq^*(k)＝Vq_ref(k)+ΔVq(k)；

the d-axis actual command voltage is:

Vd^*(k)＝Vd_ref(k)+ΔVd(k)。

9. an online compensation method for dead zone voltage of a current loop of a permanent magnet synchronous motor is characterized by comprising the following steps of: the method adopts the online compensation equipment of the dead zone voltage of the current loop of the permanent magnet synchronous motor as claimed in any one of claims 1 to 8 to carry out voltage compensation on the permanent magnet synchronous motor.