CN106788063A - Motor load mechanical impedance it is online from sensing detection method and system - Google Patents

Abstract

The invention discloses a kind of the online from sensing detection method and system of motor load mechanical impedance, the indirect measurement of the mechanical impedance for realizing loading it by the electrical impedance for detecting motor；First, threephase stator electric current is sampled and is transformed to rotor rotating coordinate system, the q shaft current compositions with the detection voltage same frequency of injection are extracted using bandpass filter；Carry out Hilbert conversion to the detection voltage and q shaft currents composition respectively, and resultant voltage and electric current analytic signal；The electrical impedance that complex operation obtains motor is carried out to the analytic signal；The mechanical impedance of the load on motor output shaft is finally calculated using the electrical impedance of motor and the coupled relation of mechanical impedance.In the case of not using additional force/torque, motion sensor, the detection of mechanical impedance loaded to it is realized in itself using motor driver, so that the shortcomings of solving big tradition machinery impedance measurement method complex structure, volume, high cost, sensor load effect.

Description

Online self-sensing detection method and system for mechanical impedance of motor load

Technical Field

The invention relates to the technical field of detection of mechanical impedance, in particular to an online self-sensing detection method and system for mechanical impedance of a motor load.

Background

The mechanical impedance is a parameter describing the frequency dynamic characteristics of a mechanical system, and is the ratio of the simple harmonic excitation force to the simple harmonic motion response when the system is excited to vibrate. The mechanical impedance is represented by lumped parameters mass, damping, stiffness and excitation frequency of the system, so that the system characteristics of mass, damping and stiffness can be extracted through mechanical impedance measurement. The mechanical impedance can also be used to build a dynamic model of the system or component to obtain an analytical expression of the system transfer function. The measurement of mechanical impedance has wide application prospect in the field of robots. In the interaction process of the robot and an unknown environment, the characteristics of the unknown object such as mass, damping, rigidity and the like can be detected through the measurement of mechanical impedance, so that the unknown object is identified and analyzed. In addition, in robot control, the measurement of environmental mechanical impedance can be used to design an adaptive optimal controller to improve the performance of the system and improve the stability and robustness of the system.

However, conventional mechanical impedance measurement methods typically involve multiple excitation and detection devices, including actuators, force sensors, and acceleration sensors. For example, patent CN103344322B uses a vibration table as the excitation generator, a force sensor for measuring the excitation force, and an acceleration sensor for measuring the motion response. Patent CN102204815B discloses a device and method for measuring mechanical impedance of human body, which uses a linear motor as a driver to generate a disturbing signal acting on the human body, uses a force sensor to measure the disturbing force, and uses a grating ruler to measure the motion response signal of the human body. Patent CN103364160A discloses a device and a method for measuring mechanical impedance of heald frame, which uses a force hammer as an excitation generator, and uses a force sensor, an acceleration sensor and a displacement sensor to measure a force application signal and a motion response signal. The above-mentioned conventional mechanical impedance measuring devices generally have the disadvantages of complicated structure, large volume, high cost, etc. due to the plurality of excitation and sensing devices involved. This also makes these methods difficult to use for mechanical impedance detection of small objects, and also difficult to perform real-time online detection of objects in their normal operating state. Furthermore, the use of sensors introduces loading effects, which leads to a reduction in the accuracy of the measurement.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides an online self-sensing detection method and system for mechanical impedance of a motor load, which utilize a motor driver to realize the detection of the mechanical impedance of the load under the condition of not using additional force/moment and motion sensors, thereby overcoming the defects of complex structure, large volume, high cost, sensor load effect and the like of the traditional mechanical impedance measurement method, and realizing the detection of a tiny object and the real-time online detection of the object in a normal working state.

The invention provides an online self-sensing detection method for mechanical impedance of a motor load, which realizes indirect measurement of the mechanical impedance of the load by detecting the electrical impedance of a motor; firstly, sampling three-phase stator current and converting the three-phase stator current into a rotor rotating coordinate system, and extracting q-axis current components with the same frequency as the injected detection voltage by using a band-pass filter; secondly, Hilbert conversion is carried out on the detected voltage and the q-axis current components respectively, and analytic signals of the voltage and the current are synthesized; then, carrying out complex operation on the analytic signal to obtain the electrical impedance of the motor; and finally, calculating the mechanical impedance of the load on the output shaft of the motor by using the coupling relation between the electrical impedance of the motor and the mechanical impedance.

The sampling and transformation of the three-phase stator current to the rotor rotating coordinate system specifically comprises the following steps:

according to the formula (1), sampling a three-phase stator current signal i_a、i_bAnd i_cCurrent signal i converted to rotor rotating coordinate system_dAnd i_q

Wherein p is the number of pole pairs of the motor; theta_mIs the rotor mechanical angle.

The analytic signal of the voltage is as follows:

wherein v is_qiFor detecting voltage signals

The analytic signal of the current is as follows:

wherein i_qiTo and detect the voltage signal v_qiCurrent signal components having the same frequency.

Obtaining the impedance Z of the detected voltage signal frequency according to the analytic signals of the voltage and the current_eq，

Then obtaining the mechanical impedance Z of the output end of the motor according to the formula (18)_mImpedance Z with its input detecting the frequency of the voltage signal_eqThe mechanical impedance of the load on the motor output shaft is calculated as follows:

wherein L is_qAnd the inductance value of the q axis, lambda is the amplitude of magnetic flux induced by the rotor permanent magnet in the stator phase, J is the rotational inertia of the motor rotor, R is the resistance value of the stator winding, and p is the number of pole pairs of the motor.

Through detecting the mechanical impedance values of the load at a plurality of different frequencies, lumped parameter mass c, damping m and rigidity k characteristics of the load are further extracted, and the method is characterized by solving the following equation set:

wherein ω is_iTo detect the frequency, i ═ 1 … n; c is the damping of the load; m is the mass of the load; k is the stiffness of the load.

The method is also applied to a direct current motor, and the indirect measurement of the mechanical impedance of the load of the direct current motor is realized by detecting the electrical impedance of the motor; firstly, injecting a detection voltage signal into a motor stator, sampling the stator current, and extracting a current component i with the same frequency as the injected detection voltage by using a band-pass filter_qi(ii) a Secondly, the detection voltage and the current component i are respectively compared_qiCarrying out Hilbert conversion and synthesizing analysis signals of voltage and current; then, carrying out complex operation on the analytic signal to obtain the electrical impedance of the motor; and finally, calculating the mechanical impedance of the load on the output shaft of the motor by using the coupling relation between the electrical impedance of the motor and the mechanical impedance.

An online self-sensing detection system for mechanical impedance of a motor load comprises a permanent magnet synchronous motor, a current control module, a detection voltage signal injection module, an adder, a current detection module and a load mechanical impedance extraction module, wherein the detection voltage signal injection module injects a detection voltage signal to a q axis of the motor through the adder, the current detection module detects three-phase current of the permanent magnet synchronous motor and inputs the three-phase current to the load mechanical impedance extraction module, and the load mechanical impedance extraction module acts on a load object of the motor according to the excitation generated by the detection voltage signal on a mechanical output end of the motor under the electromechanical coupling action of the motor and reversely transmits the excitation to an electrical input end of the motor, so that indirect measurement of the mechanical impedance of the load is realized.

The load mechanical impedance extraction module comprises a dq conversion module, a band elimination filter, a band-pass filter, a first Hilbert conversion module, a second Hilbert conversion module, a first analytic signal synthesis module, a second analytic signal synthesis module, a complex division module and a load mechanical impedance calculation module; wherein,

the dq conversion module is connected with the current detection module and converts the three-phase stator current sampling signal to a rotor rotation coordinate system current signal i_dAnd i_qI is to_qAs input signals of a band-stop filter and a band-pass filter, i is obtained_qiThe first analytic signal synthesis module is used as input of a first Hilbert transformation module; then the output of the second Hilbert conversion module is used as the input of the second analytic signal synthesis module to obtain an analytic signal of the current;

the detection voltage signal injection module generates a detection voltage signal with a frequency different from that of the control voltage command signal, the detection voltage signal is used as the input of the adder, the first Hilbert conversion module and the first analysis signal synthesis module, and the output of the first Hilbert conversion module is used as the input of the first analysis signal synthesis module to obtain an analysis signal of voltage;

and the analytic signals of the current and the voltage are sequentially calculated by a complex division module and a load mechanical impedance calculation module to obtain the mechanical impedance of the load on the output shaft of the motor.

Compared with the prior art, the invention adopting the technical scheme has the following technical effects:

compared with the prior art, the invention utilizes the motor to generate excitation and avoids using an additional force/moment and motion sensor. The beneficial effects brought by the method comprise: the device has the advantages of simple structure, small volume, low cost and no sensor load effect, is suitable for detecting small objects, and can carry out real-time online detection on the objects in the normal working state. In addition, because the electrical impedance is the inherent property of the motor and does not change along with the change of the voltage amplitude, the method also has the effect of resisting the interference of power supply voltage fluctuation.

Drawings

FIG. 1 is a schematic diagram of a rotor rotating coordinate system of a three-phase permanent magnet synchronous motor;

FIG. 2 is a block diagram illustrating an embodiment of a mechanical impedance self-sensing detection system for a motor load according to the present invention;

FIG. 3 is a waveform diagram of a q-axis control voltage, a q-axis injected detection voltage, and their resultant signals;

FIG. 4 is a waveform diagram of a three-phase stator current sampling signal, a current signal in a rotor rotation coordinate system obtained through dq transformation, a q-axis control current signal extracted through filtering, and an injected detection current response signal;

FIG. 5 is a waveform diagram of a sensed voltage signal injected into the q-axis, a q-axis sensed current response signal having the same frequency as the q-axis, and a Hilbert transform result thereof;

FIG. 6 shows the detected electrical impedance values under 5 load conditions;

FIG. 7 is a graph comparing measured values of load mechanical impedance with actual values of load mechanical impedance;

FIG. 8 is a flowchart illustrating the processing procedure of the present detection system;

symbolic illustration in the figures: 1. a permanent magnet synchronous motor; 2. a current control module; 3. a detection voltage signal injection module; 4. an adder; 5. an inverse dq transformation module; 6. an output voltage processing module; 7. a power conversion module; 8. a current detection module; 9. a load mechanical impedance extraction module; a dq transformation module; 11. a band-stop filter; 12. a band-pass filter; a Hilbert transform module; a Hilbert transform module; 14a, an analytic signal synthesis module; 14b, analyzing a signal synthesis module; 15. a plurality of division modules; 16. and a load mechanical impedance calculation module.

Detailed Description

The invention provides an online self-sensing detection method and system for mechanical impedance of a motor load, which aims to make the purpose, technical scheme and effect of the invention clearer and make the invention further described in detail by referring to the attached drawings and taking examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The embodiment of the invention injects a detection voltage signal into a q axis of a motor under the condition of controlling the d axis current of the motor to be 0, and extracts the mechanical impedance of a load by detecting a three-phase current response signal of the motor and processing, converting, synthesizing and calculating the signal.

Fig. 2 is a block diagram showing the structure of an embodiment of the self-sensing detection of mechanical impedance of a motor load according to the present invention. The motor load mechanical impedance self-sensing detection device of the present embodiment includes: a permanent magnet synchronous motor 1; a current control module 2; a detection voltage signal injection module 3; an adder 4; an inverse dq transformation module 5; an output voltage processing module 6; a power conversion module 7; a current detection module 8; a load mechanical impedance extraction module 9; a dq transformation module 10; a band-stop filter 11; a band-pass filter 12; hilbert transform modules 13a and 13 b; analytic signal synthesis modules 14a and 14 b; a complex division module 15; the load mechanical impedance calculation module 16.

The current control module 2 controls the current command value i according to the d-axis and the q-axis^* _dcAnd i^* _qcCurrent value i corresponding to the actually detected d-axis and q-axis_dcAnd i_qcCalculating the control voltage command signal v to be applied to the d-axis and the q-axis^* _dcAnd v^* _qcSo that the actually detected d-axis and q-axis current values i_dcAnd i_qcAnd current command value i^* _dcAnd i^* _qcAre equal, i_dc＝i^* _dc＝0，i_qc＝i^* _qc。

The detection voltage signal injection module 3 generates a control voltage command signal v^* _qcDetection voltage signals v with different frequencies_qi. As inputs to the adder 4, the Hilbert transform module 13a and the analytic signal synthesis module 14a.

The adder 4 adds the q-axis control voltage command signal v^* _qcAnd detecting the voltage signal v_qiResult v after addition_qIs fed to the inverse dq converter 5. Shown in FIG. 3 as v^* _qc、v_qiAnd v_qExamples of (2). Where v is^* _qcIs a sine sin signal with a frequency of 1Hz, v_qiIs a sine sin signal with a frequency of 100Hz, v_qAre the signals they synthesize.

The inverse dq converter 5 will need to apply the d-axis voltage v to the motor_d(＝v^* _dc) And q-axis voltage v_qAnd transforming to a stator coordinate system and sending a transformation result to the output voltage processing module 6.

The output voltage processing module 6 processes (for example, converts into a space vector pulse modulation signal SVPWM, or linearly amplifies) the stator coordinate system voltage value output from the inverse dq converter 5, and sends the result to the power conversion module 7.

The power conversion module 7 may be a pulse modulation signal controlled inverter, or a linear power amplifier, etc. The output is connected with the motor 1.

The current detection module 8 samples the three-phase stator current of the motor, and may be a hall current sensor, or a shunt resistor plus voltage amplification circuit, etc.

The load mechanical impedance extraction module 9 samples a three-phase stator current sampling signal i output by the current detection module 8_a、i_bAnd i_cProcessing, converting, synthesizing and calculating to extract the mechanical impedance Z of the load_m. It includes: a dq transformation module 10; a band-stop filter 11; a band-pass filter 12; hilbert transform modules 13a and 13 b; analytic signal synthesis modules 14a and 14 b; a complex division module 15; the load mechanical impedance calculation module 16.

The dq conversion module 10 converts the three-phase stator current sampling signal i output from the current detection module 8 according to dq conversion expressed by formula (1)_a、i_bAnd i_cCurrent signal i converted to rotor rotating coordinate system_dAnd i_q. Wherein i_dAs i_dcIs used for the current control module 2, i_qAs inputs to a band-stop filter 11 and a band-pass filter 12.

Wherein p is the number of pole pairs of the motor; theta_mIs the rotor mechanical angle. i.e. i_a、i_b、i_cAnd i_dAnd i_qSee fig. 4.

Band elimination filter 11 to the i output from dq conversion module 10_qFiltering, filtering and detecting the voltage signal v_qiCurrent signal components having the same frequency, filtering result i_qcIs used for the current control module 2. i.e. i_qcSee fig. 4.

With bandpass filters 12 for the output of dq conversion modules 10i_qFiltering to extract and detect voltage signal v_qiCurrent signal components having the same frequency, filtering result i_qiAs inputs to the Hilbert transform module 13b and the analytic signal synthesis module 14b. i.e. i_qiSee fig. 4.

The Hilbert conversion module 13a injects the detection voltage signal v generated by the detection voltage signal injection module 3 into_qiThe Hilbert transform represented by formula (2) is performed to shift the phase of the negative frequency component of the signal by pi/2 and to shift the phase of the positive frequency component of the signal by-pi/2. The Hilbert transform may be implemented by a finite impulse response Filter (FIR), a Fast Fourier Transform (FFT), or the like. Its output HT [ v ]_qi]Is used as an input to the analytic signal synthesis module 14a. v. of_qiAnd HT [ v ]_qi]See fig. 5.

The Hilbert transform module 13b outputs a filtering result i to the band-pass filter 12_qiThe Hilbert transform represented by formula (3) is performed to shift the phase of the negative frequency component of the signal by pi/2 and to shift the phase of the positive frequency component of the signal by-pi/2. The Hilbert transform may be implemented by a finite impulse response Filter (FIR), a Fast Fourier Transform (FFT), or the like. Its output HT [ i_qi]Is used as an input to the analytic signal synthesis module 14b. i.e. i_qiAnd HT [ i_qi]See fig. 5.

The analysis signal synthesis module 14a injects the detection voltage signal into the detection voltage signal v generated by the detection voltage signal injection module 3 according to the formula (4)_qiAnd HT [ v ] output by the Hilbert transform module 13a_qi]Synthesized as an analytic signal V_qi。V_qiWith v_qiAs its real part, it is denoted HT [ v ]_qi]As its imaginary part. V_qiIs used asThe input of the complex division module 15.

The analysis signal synthesis module 14b outputs the filtering result i output by the band-pass filter 12 according to the formula (5)_qiAnd HT [ i ] output by the Hilbert transform module 13b_qi]Synthesized to an analytic signal I_qi。I_qiWith i_qiAs its real part, it is denoted HT [ i ]_qi]As its imaginary part. I is_qiIs used as an input to the complex division module 15.

The complex division module 15 synthesizes the output V of the analysis signal synthesis module 14a according to the formula (6)_qiDivided by the output I of the analytic signal synthesis module 14b_qiObtaining a corresponding detection voltage signal v_qiElectrical impedance of frequency Z_eqAnd input to the load mechanical impedance calculation module 16.

The load mechanical impedance calculation module 16 outputs Z output by the complex division module 15_eqCalculating the mechanical impedance Z of the load according to the functional relationship expressed by the formula (18)_m. The principle is as follows:

the mathematical model of the three-phase permanent magnet synchronous machine in the dq-rotor rotation coordinate system shown in fig. 1 can be represented by equations (7-10)

Wherein v is_d、v_qVoltages of a d axis and a q axis in a rotor coordinate system are respectively; i.e. i_d、i_qCurrents of a d axis and a q axis in a rotor coordinate system respectively; r is the resistance value of the stator winding; l is_d、L_qThe inductance values of the d axis and the q axis respectively; omega_mIs the rotor mechanical angular velocity; p is the number of pole pairs of the motor; λ is the amplitude of the magnetic flux induced in the stator phase by the rotor permanent magnets; t is_eIs electromagnetic torque; t is_LAn external load moment; j is the moment of inertia of the motor rotor.

Fourier transforming equations (7-10) results in a mathematical model of the motor in the frequency domain as represented by equations (11-14):

jωJΩ_m(ω)＝T_e(ω)-T_L(ω) (14)

where ω is the signal frequency; i is_d(ω)、I_q(ω)、V_d(ω)、V_q(ω)、Ω_m(ω)、T_e(omega) are respectively time domain signals i_d、i_q、v_d、v_q、Ω_m、T_eIs shown.

Controlling the current component in the d-axis direction to be 0 by using a control algorithm (such as PID, sliding mode control, adaptive control, fuzzy control and the like), namely:

I_d(ω)＝0 (15)

the mechanical impedance at the output of the motor is defined as:

the electrical impedance of the motor input in the q-axis direction is defined as:

the mechanical impedance Z of the output end of the motor can be obtained by substituting equations (15-17) into equations (11-14) and arranging_mQ-axis impedance Z of its input end_eqThe relationship of (1) is:

wherein Re [ Z ]_eq]And Im [ Z ]_eq]Q-axis impedance Z of input end respectively_eqReal and imaginary parts of (c).

In addition, by detecting the mechanical impedance values of the load at a plurality of different frequencies, the characteristics of lumped parameter mass m, damping c, rigidity k and the like of the load can be further extracted. The method comprises the following steps of solving the following equation set:

wherein ω is_i(i ═ 1 … n) is the detection frequency; c is the damping of the load; m is the mass of the load; k is the stiffness of the load. In a rotating system, c, m and k are rotational damping, rotational inertia and rotational stiffness, respectively.

Finally, the effectiveness of the method and the system is tested, the q-axis resistance of the motor is 0.32 omega, and the inductance is 0.082 × 10^-3H, the moment of inertia of the rotor is 1.4 × 10^-5kg·m²The rated voltage is 12V. The amplitude of the injected detection voltage is 1V, and the frequency is 100 Hz. The following 5 loads were tested:

load 1 moment of inertia 1 × 10^-4kg·m²Damping is 0.01 N.m.s/rad, and rigidity is 10 N.m/rad;

load 2 moment of inertia 1 × 10^-5kg·m²Damping is 0.02 N.m.s/rad, and rigidity is 50 N.m/rad;

load 3 moment of inertia 1 × 10^-4kg·m²Damping is 0.03 N.m.s/rad, and rigidity is 20 N.m/rad;

load 4 moment of inertia 1 × 10^-5kg·m²The damping is 0.04 N.m.s/rad, and the rigidity is 40 N.m/rad;

load 5 moment of inertia 1 × 10^-4kg·m²Damping is 0.05 N.m.s/rad, and rigidity is 30 N.m/rad;

FIG. 6 shows the detected values of electrical impedance Z under the above 5 load conditions_eq. FIG. 7 is a graph of using detected Z_eqA comparison of the calculated measured value of the mechanical impedance of the load with the actual value of the mechanical impedance of the load. It has been found that the detected mechanical impedance of the load effectively reflects its actual value.

The flow chart of the execution of the detection system processing program is shown in fig. 8.

The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (8)

1. An online self-sensing detection method for mechanical impedance of a motor load is characterized in that under the condition that the current of a d axis of a motor is controlled to be 0, a detection voltage signal is injected into a q axis of the motor, and indirect measurement of the mechanical impedance of the load is realized by detecting the electrical impedance of the motor; firstly, sampling three-phase stator current and converting the three-phase stator current into a rotor rotating coordinate system, and extracting q-axis current components with the same frequency as the injected detection voltage by using a band-pass filter; secondly, Hilbert conversion is carried out on the detected voltage and the q-axis current components respectively, and analytic signals of the voltage and the current are synthesized; then, carrying out complex operation on the analytic signal to obtain the electrical impedance of the motor; and finally, calculating the mechanical impedance of the load on the output shaft of the motor by using the coupling relation between the electrical impedance of the motor and the mechanical impedance.

2. The method for on-line self-sensing detection of mechanical impedance of motor load according to claim 1, wherein the sampling and transformation of three-phase stator current to rotor rotating coordinate system is specifically:

according to the formula (1), sampling a three-phase stator current signal i_a、i_bAnd i_cCurrent signal i converted to rotor rotating coordinate system_dAnd i_q

$\left[\begin{array}{c}{i}_d\\ i_q\end{array}\right]=\frac{2}{3}\left[\begin{array}{ccc}\cos \left({\mathrm{p\θ}}_m\right)& \cos ({\mathrm{p\θ}}_m-\frac{2\mathrm{\π}}{3})& \cos \left({\mathrm{p\θ}}_m+\frac{2\mathrm{\π}}{3}\right)\\ -\sin \left({\mathrm{p\θ}}_m\right)& -\sin ({\mathrm{p\θ}}_m-\frac{2\mathrm{\π}}{3})& -\sin ({\mathrm{p\θ}}_m+\frac{2\mathrm{\π}}{3})\end{array}\right]\left[\begin{array}{c}{i}_a\\ i_b\\ i_c\end{array}\right]---\left(1\right)$

Wherein p is the number of pole pairs of the motor; theta_mIs the rotor mechanical angle.

3. The on-line self-sensing detection method for the mechanical impedance of the motor load according to claim 1, wherein the analytic signal of the voltage is as follows:

wherein v is_qiFor detecting voltage signals

The analytic signal of the current is as follows:

wherein i_qiTo and detect the voltage signal v_qiCurrent signal components having the same frequency.

4. The method as claimed in claim 1, wherein the voltage and current are analyzed according to the self-sensing methodElectrical impedance Z for detecting voltage signal frequency by signal acquisition_eq，

$Z_{eq}=\frac{V_{qi}\left(t\right)}{I_{qi}\left(t\right)}---\left(6\right)$

Then obtaining the mechanical impedance Z of the output end of the motor according to the formula (18)_mImpedance Z with its input detecting the frequency of the voltage signal_eqThe mechanical impedance of the load on the motor output shaft is calculated as follows:

$\begin{array}{c}{Z}_m=f\left(Z_{eq}\right)\\ =\frac{\frac{3}{2}{\mathrm{\λ}}^2{p}^2-{\mathrm{\ω}}^2{\mathrm{JL}}_q+j\mathrm{\ω}JR-{\mathrm{j\ωJZ}}_{eq}}{Z_{eq}-R-{\mathrm{j\ωL}}_q}\\ =\frac{\frac{3}{2}{\mathrm{\λ}}^2{p}^2-{\mathrm{\ω}}^2{\mathrm{JL}}_q+{\mathrm{\ω}}^{2}J\mathrm{Im}\[Z_{eq}\]+j\mathrm{\ω}\{JR-J\mathrm{Re}\[Z_{eq}\]\}}{\mathrm{Re}\[Z_{eq}\]-R+j\mathrm{\ω}\mathrm{Im}\[Z_{eq}\]-{\mathrm{j\ωL}}_q}\end{array}---\left(18\right)$

wherein L is_qAnd the inductance value of the q axis, lambda is the amplitude of magnetic flux induced by the rotor permanent magnet in the stator phase, J is the rotational inertia of the motor rotor, R is the resistance value of the stator winding, and p is the number of pole pairs of the motor.

5. The on-line self-sensing detection method for the mechanical impedance of the motor load according to claim 1, characterized in that lumped parameter mass c, damping m and stiffness k characteristics of the load are further extracted by detecting the mechanical impedance values of the load at a plurality of different frequencies, and the method is to solve the following equation sets:

$\left\{\begin{array}{c}{Z}_m\left({\mathrm{\ω}}_1\right)\\ Z_m\left({\mathrm{\ω}}_2\right)\\ .\\ .\\ .\\ Z_m\left({\mathrm{\ω}}_n\right)\end{array}\right\}=\left[\begin{array}{ccc}1& {\mathrm{j\ω}}_1& -\frac{j}{{\mathrm{\ω}}_1}\\ 1& {\mathrm{j\ω}}_2& -\frac{j}{{\mathrm{\ω}}_2}\\ .& .& .\\ .& .& .\\ .& .& .\\ 1& {\mathrm{j\ω}}_n& -\frac{j}{{\mathrm{\ω}}_n}\end{array}\right]\left\{\begin{array}{c}c\\ m\\ k\end{array}\right\}---\left(19\right)$

wherein ω is_iTo detect the frequency, i ═ 1 … n; c is the damping of the load; m is the mass of the load; k is the stiffness of the load.

6. The system is characterized by comprising a permanent magnet synchronous motor, a current control module, a detection voltage signal injection module, an adder, a current detection module and a load mechanical impedance extraction module, wherein the detection voltage signal injection module injects a detection voltage signal to a q axis of the motor through the adder, the current detection module detects three-phase current of the permanent magnet synchronous motor and inputs the three-phase current to the load mechanical impedance extraction module, and the load mechanical impedance extraction module applies excitation generated by the detection voltage signal on a mechanical output end of the motor to a load object of the motor according to the electromechanical coupling action of the motor and reversely transmits the excitation to an electrical input end of the motor to realize indirect measurement of the mechanical impedance of the load.

7. The on-line self-sensing detection system of the mechanical impedance of the motor load according to claim 6, wherein the load mechanical impedance extraction module comprises a dq transformation module, a band elimination filter, a band pass filter, a first and a second Hilbert transformation modules, a first and a second analytic signal synthesis modules, a complex division module and a load mechanical impedance calculation module; the dq conversion module is connected with the current detection module and converts the three-phase stator current sampling signal to a rotor rotation coordinate system current signal i_dAnd i_qI is to_qAs input signals of a band-stop filter and a band-pass filter, i is obtained_qiThe first analytic signal synthesis module is used as input of a first Hilbert transformation module; then the output of the second Hilbert conversion module is used as the input of the second analytic signal synthesis module to obtain an analytic signal of the current;

the detection voltage signal injection module generates a detection voltage signal with a frequency different from that of the control voltage command signal, the detection voltage signal is used as the input of the adder, the first Hilbert conversion module and the first analysis signal synthesis module, and the output of the first Hilbert conversion module is used as the input of the first analysis signal synthesis module to obtain an analysis signal of voltage;

and the analytic signals of the current and the voltage are sequentially calculated by a complex division module and a load mechanical impedance calculation module to obtain the mechanical impedance of the load on the output shaft of the motor.

8. The online self-sensing detection method of the mechanical impedance of a motor load is characterized in that the method is applied to a direct current motor, and the indirect measurement of the mechanical impedance of the load is realized by detecting the electrical impedance of the motor; firstly, injecting a detection voltage signal into a motor stator, sampling the stator current, and extracting a current component i with the same frequency as the injected detection voltage by using a band-pass filter_qi(ii) a Secondly, the detection voltage and the current component i are respectively compared_qiPerforming Hilbert conversion, and synthesizing voltage and currentThe analytic signal of (1); then, carrying out complex operation on the analytic signal to obtain the electrical impedance of the motor; and finally, calculating the mechanical impedance of the load on the output shaft of the motor by using the coupling relation between the electrical impedance of the motor and the mechanical impedance.