Mechanical parameter identification method of permanent magnet synchronous motor based on extended disturbance observer
By constructing extended state equations and speed estimation equations through an extended disturbance observer, and designing a moment of inertia trigger, synchronous identification of load torque and moment of inertia of permanent magnet synchronous motors was achieved. This solved the problems of insufficient accuracy and long time consumption in the existing technology, and improved the identification accuracy and robustness.
Patent Information
- Application Number
- CN202411120237.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-15
AI Technical Summary
Existing technologies struggle to achieve synchronous identification of load torque and moment of inertia of permanent magnet synchronous motors within a single observer architecture. Furthermore, the identification accuracy is insufficient and the time consumption is long, especially when the motor is in steady state, the moment of inertia identification results fluctuate significantly.
By employing an extended disturbance observer-based approach, an extended state equation and a rotational speed estimation equation are constructed. The difference between the rotational speed estimation equation and the extended state equation is used to design a moment of inertia trigger, thereby achieving synchronous observation of load torque and moment of inertia.
It improves the accuracy and speed of parameter identification, suppresses the coupling error between load torque and rotational inertia, enhances the robustness of the observer, and ensures the identification accuracy in steady state.
Smart Images

Figure CN118646301B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of permanent magnet synchronous motor parameter identification, and in particular relates to a permanent magnet synchronous motor mechanical parameter identification method based on an extended disturbance observer. Background Art
[0002] Permanent magnet synchronous motors have the characteristics of high power density, large torque-to-inertia ratio and high efficiency, and have been widely used in many industrial fields such as industrial robots, electric vehicles and CNC machine tools. The moment of inertia and load torque in the permanent magnet synchronous motor drive system are key information for motor control. They cannot be directly measured and usually change with different tasks. In industrial applications, the inertia of the permanent magnet synchronous motor drive system changes exponentially on the basis of the nominal inertia as the load changes. Especially in direct drive systems, this will lead to large overshoot and long settling time; although inertia mismatch is sometimes beneficial to anti-disturbance characteristics, it will have an adverse effect on the tracking performance when the reference quantity changes continuously. With the growing demand for high-precision and high-performance control in the industrial field, it is particularly important to obtain the load torque and moment of inertia information in the permanent magnet synchronous motor drive system in real time. Therefore, it is of great significance to quickly and accurately identify the load torque and moment of inertia of the permanent magnet synchronous motor to improve the control performance of the permanent magnet synchronous motor.
[0003] At present, in the field of online parameter identification of permanent magnet synchronous motor systems, although most studies such as the literature [C. Yang, B. Song, Y. Xie, S. Lu, and X. Tang, “Stable Simultaneous Inertia and Disturbance Torque Identification for SPMSM Drive Systems Subject to Mismatched Rotor Flux Linkage,” IEEE J. Emerg. Sel. Top. Power Electron., pp.1–1, 2021, doi: 10.1109 / JESTPE. 2021. 3111122] have been able to effectively identify load torque and moment of inertia respectively through dual observers, they have not yet been able to achieve synchronous identification of the two within a single observer architecture. Only a few studies, such as [L. Wang, Z. Tang, P. Zhang, X. Liu, D. Wang, and X. Li, “DoubleExtended Sliding-Mode Observer-Based Synchronous Estimation of Total Inertiaand Load Torque for PMSM-driven Spindle-Tool Systems,” IEEE Trans. Ind.Inform., vol. 19, no. 7, pp. 8496–8507, Jul. 2023, doi: 10.1109 / TII. 2022.3218858], have attempted to simultaneously identify these two parameters online using a single observer. However, these methods generally face the problems of insufficient parameter identification accuracy and long time consumption. In particular, when the motor is running in a steady state, the identification results of the moment of inertia often show large fluctuations, which has become a key factor restricting its widespread application and urgently needs further optimization and improvement. Summary of the invention
[0004] In view of the above, the present invention provides a method for mechanical parameter identification of a permanent magnet synchronous motor based on an extended disturbance observer. The method can simultaneously identify load torque and moment of inertia, improve the accuracy of parameter identification and the robustness of the observer, and has stronger applicability and practicality for complex industrial equipment.
[0005] A method for identifying mechanical parameters of a permanent magnet synchronous motor based on an extended disturbance observer comprises the following steps:
[0006] (1) Preliminary acquisition of the mechanical and electrical parameters of the permanent magnet synchronous motor, and the actual mechanical position angle of the permanent magnet synchronous motor through the encoder and current sensor q m and three-phase stator current i a , i b and i c , and then calculate the mechanical angular velocity of the permanent magnet synchronous motor oh m and the d-axis stator current i d and q-axis stator current i q ;
[0007] (2) Load torque of permanent magnet synchronous motor T L and the inverse moment of inertia v As the extended state variable, the extended state equation of the permanent magnet synchronous motor is constructed;
[0008] (3) Mechanical angular velocity oh m , Load torque T L and the inverse moment of inertia v As the observation object, the speed estimation equation of the extended disturbance observer is constructed;
[0009] (4) using the speed estimation equation and the extended state equation to perform a subtraction to obtain a relational expression between a load torque estimation error and a reciprocal moment of inertia estimation error;
[0010] (5) Obtain the extended perturbation observation model by transformation according to the relationship in step (4);
[0011] (6) Using the extended disturbance observation model to realize load torque T L and moment of inertia J Make observations.
[0012] Furthermore, the mechanical parameters include the number of motor pole pairs p , initial value of moment of inertia J (0), viscous friction coefficient B And the Coulomb friction coefficient C , the electrical parameters include the d-axis stator inductance L d , q-axis stator inductance L q , stator resistance R s and magnet links ψf .
[0013] Furthermore, in step (1), the mechanical angular velocity is calculated by the following formula: oh m :
[0014]
[0015] in: t Indicates time.
[0016] Furthermore, in step (1), the d-axis stator current is calculated by the following formula: i d and q-axis stator current i q :
[0017] ;
[0018]
[0019] in: M abc / dq is the transformation matrix from the abc three-phase coordinate system to the dq two-phase coordinate system, oh e is the electromagnetic angular velocity of the permanent magnet synchronous motor and oh e = p m , p is the number of motor pole pairs, t Indicates time.
[0020] Furthermore, the extended state equation in step (2) is expressed as follows:
[0021] ;
[0022]
[0023] in: T f and T e are the friction torque and electromagnetic torque of the permanent magnet synchronous motor respectively, T e =1.5 pψ f i q , v =1 / J , sgn( ) represents the sign function, , , They are ohm , T L , v The first derivative of p , ψ f , B , C They are the motor pole pair number, magnetic flux, viscous friction coefficient and Coulomb friction coefficient of the permanent magnet synchronous motor respectively.
[0024] Furthermore, the speed estimation equation in step (3) is expressed as follows:
[0025] ;
[0026]
[0027] in: , , They are oh m , v , T L The observed value of for The first derivative of k is the speed estimation model gain ( k <0), e 1 indicates mechanical angular velocity estimation error.
[0028] Furthermore, the specific implementation method of step (4) is as follows: first, the speed estimation equation is subtracted from the extended state equation to obtain the estimated error relationship of the first-order derivative of the mechanical angular velocity as follows:
[0029]
[0030] Then, according to the above estimation error relationship, the relationship between the load torque estimation error and the inverse moment of inertia estimation error is converted as follows:
[0031]
[0032] in: for e The first derivative of 1, e 2 is the load torque estimation error, that is , e 3 The estimation error of the inverse moment of inertia is .
[0033] Furthermore, the expression of the extended disturbance observation model is as follows:
[0034]
[0035] in: k t and k v are the given load torque gain and moment of inertia gain ( k t <0, k v <0), and They are and The first derivative of .
[0036] Preferably, in order to suppress the influence of the coupling error between the load torque and the moment of inertia on the estimation accuracy thereof and improve the robustness of the extended disturbance observer, the present invention uses the dynamic change information of the motor to design the moment of inertia trigger, so as to ensure that the moment of inertia identification works when the permanent magnet synchronous motor is in a dynamic change and remains stationary when the permanent magnet synchronous motor is in a steady state; therefore, the optimization expression of the extended disturbance observation model is as follows:
[0037] ;
[0038] ;
[0039]
[0040] in: q 1 is the moment of inertia trigger, r For a given trigger proportional gain, d For a given trigger threshold, l m As the trigger factor, is the given mechanical angular velocity reference value.
[0041] Furthermore, the load torque is realized in step (6). T L and moment of inertia J The specific expression for observation is as follows:
[0042]
[0043] in: is the moment of inertia J The observed value of T L (0) is the initial value of the load torque and T L (0)=0, v (0)=1 / J (0), J (0) is the initial value of the moment of inertia.
[0044] Compared with the prior art, the present invention has the following beneficial effects:
[0045] 1. The extended disturbance observer proposed in the present invention can realize the parameter identification of load torque and moment of inertia in one observer, which effectively improves the identification accuracy and identification speed of the parameters.
[0046] 2. In the extended disturbance observer proposed in the present invention, the introduction of the moment of inertia trigger effectively suppresses the coupling error between the load torque and the moment of inertia, ensures the identification accuracy of the moment of inertia in the steady state of the permanent magnet synchronous motor, reduces the amplitude fluctuation of the load torque identification result, and greatly improves the robustness of the extended disturbance observer. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 This is a flowchart of the mechanical parameter identification process of the permanent magnet synchronous motor of the present invention.
[0048] Figure 2 It is a schematic diagram of the load torque parameter identification experiment results using the method of the present invention.
[0049] Figure 3 Schematic diagram of the results of the moment of inertia parameter identification experiment using the method of the present invention. DETAILED DESCRIPTION
[0050] In order to describe the present invention more specifically, the technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0051] like Figure 1 As shown, the present invention is based on the permanent magnet synchronous motor mechanical parameter identification method of the extended disturbance observer, and the specific process is as follows:
[0052] First, the initial mechanical and electrical parameters of the permanent magnet synchronous motor are obtained, including the number of motor pole pairs. p , d-axis stator inductance L d , q-axis stator inductance L q , stator resistance R s , Magnetic Link ψ f , initial value of moment of inertia J (0), viscous friction coefficient B And the Coulomb friction coefficient C ; Obtain the actual mechanical position angle of the motor through the encoder q m And calculate the mechanical angular velocity of the motor oh m .
[0053] This embodiment uses an encoder and a current sensor to obtain the actual mechanical position angle of the motor. q m And the stator current of the three axes abc in the motor abc three-phase coordinate system i a , i b and i c The mechanical angular velocity of the motor oh m By measuring the actual mechanical position angle q m Perform differential operation to obtain the specific expression as follows:
[0054]
[0055] D-axis stator current in the motor dq two-phase coordinate system i d and q-axis stator current i q Then, by the three-phase stator current i a , i b and i c After coordinate transformation, the specific expression is as follows:
[0056] ;
[0057]
[0058] Where: M abc / dq is the transformation matrix from the abc three-phase coordinate system to the dq two-phase coordinate system, oh e is the electromagnetic angular velocity of the permanent magnet synchronous motor, oh e = p m , t Indicates time.
[0059] Then, the permanent magnet synchronous motor load torque T L and the inverse moment of inertia v The extended state equation of the permanent magnet synchronous motor is constructed as an extended state variable; the mechanical angular velocity of the permanent magnet synchronous motor is oh m , Load torque T L and the inverse moment of inertia vThe speed estimation equation of the extended disturbance observer is constructed for the observed object; the speed estimation equation is used to make a difference with the extended state equation to obtain the relationship between the load torque estimation error and the inverse moment of inertia estimation error, and finally the extended disturbance observation model is obtained according to the relationship, thereby realizing the load torque T L and moment of inertia J Observation.
[0060] Permanent magnet synchronous motor load torque T L and the inverse moment of inertia v As an extended state variable, the extended state equation of the permanent magnet synchronous motor is constructed as follows:
[0061] ;
[0062]
[0063] in: T f and T e are the friction torque and electromagnetic torque of the permanent magnet synchronous motor respectively, T e =1.5 pψ f i q , v =1 / J , sgn( ) represents the sign function, , , They are oh m , T L , v The first derivative of p , ψ f , B , C are the number of motor pole pairs, flux linkage, viscous friction coefficient, and Coulomb friction coefficient of the permanent magnet synchronous motor respectively; the switching frequency of the controller is relatively high, and the load torque and the inverse of the moment of inertia are considered to be constant within a control cycle, that is, , .
[0064] The mechanical angular velocity of the permanent magnet synchronous motor oh m , Load torque T L and the inverse moment of inertia v As the observation object, the speed estimation equation of the constructed extended disturbance observer is:
[0065] ;
[0066]
[0067] in: , , They are oh m , v , T L The observed value of for The first derivative of k is the speed estimation model gain ( k <0), e 1 indicates mechanical angular velocity estimation error.
[0068] According to the speed estimation equation of the extended disturbance observer of the permanent magnet synchronous motor, the estimated error relationship between the observed value and the actual value of the first-order derivative of the angular velocity is obtained by subtracting the extended state equation:
[0069]
[0070] Where: for e The first derivative of 1, e 2 is the observed value of load torque The actual value of the load torque T L The estimated error between , e 3 is the observed value of the inverse of the load torque The actual value of the inverse of the load torque v The estimated error between .
[0071] According to the estimated error relationship between the observed value and the actual value of the first-order derivative of angular velocity, the relationship between the load torque estimation error and the inverse moment of inertia estimation error can be obtained:
[0072]
[0073] According to the relationship between the load torque estimation error and the inverse moment of inertia estimation error, the extended disturbance observation model can be obtained:
[0074]
[0075] in: k t and k v are load torque gain and moment of inertia gain (k t <0, k v <0), and respectively and The first derivative of .
[0076] In order to suppress the influence of the coupling error between load torque and moment of inertia on its estimation accuracy and improve the robustness of the extended disturbance observer, the moment of inertia trigger is designed using the dynamic change information of the motor to ensure that the moment of inertia identification works when the permanent magnet synchronous motor is in dynamic change and remains stationary when the permanent magnet synchronous motor is in steady state. q 1:
[0077] ;
[0078]
[0079] in: q 1 is the moment of inertia trigger, r is the trigger proportional gain, d is the trigger threshold, l m The trigger factor contains the speed change information. is a given angular velocity.
[0080] The relationship between the moment of inertia trigger and the moment of inertia gain is:
[0081]
[0082] Considering that the observer system is stable e 2 and e 3 approaches zero, and the extended perturbation observation model is simplified to:
[0083]
[0084] The expressions for converting the observed values of load torque and moment of inertia are:
[0085]
[0086] in: is the moment of inertia J The observed value of T L (0) is the initial value of the load torque and T L (0)=0, v (0)=1 / J (0), J(0) is the initial value of the moment of inertia.
[0087] Finally, the load torque observation value and rotational inertia observation value of the permanent magnet synchronous motor were obtained.
[0088] Verification example:
[0089] In order to verify the effectiveness and superiority of the permanent magnet synchronous motor mechanical parameter identification method of the present invention, we conducted an experimental verification. In the experiment, a surface-mounted permanent magnet synchronous motor was used as an example. The parameters are shown in Table 1:
[0090] Table 1
[0091]
[0092] The given speed of the permanent magnet synchronous motor is set to a trapezoidal speed, with the upper and lower limits of 130rad / s and 70rad / s respectively, the rise and fall time of the trapezoidal speed are both 125ms, and the waveform frequency of the trapezoidal speed is 2Hz. At the initial moment, the motor is started without load, and the load torque and moment of inertia are 0Nm and 0.01388kgm respectively. 2 ; When the motor is running for 0~4s, the load torque and moment of inertia are 5Nm and 0.01388kgm respectively 2 , the load torque and moment of inertia are 10Nm and 0.01388kgm respectively at 4~8s 2 , the load torque and moment of inertia are 0Nm and 0.00504kgm respectively at 8~11s 2 , the load torque and moment of inertia are 7Nm and 0.01388kgm respectively at 11~15s 2 .
[0093] The experimental results of simultaneous identification of load torque and moment of inertia are as follows: Figure 2 and Figure 3 As shown in the experimental results, it can be seen that the load torque and moment of inertia can converge to the actual load torque value and moment of inertia value within 0.35s, and the identification error is less than 1.5% of the actual value. When the method of the present invention is used at 4-8s, the fluctuation amplitudes of the load torque and moment of inertia are 0.7Nm and 0.00096kgm respectively. 2 Therefore, the motor mechanical parameter identification method of the present invention has a faster convergence speed and higher identification accuracy.
[0094] On the whole, the permanent magnet synchronous motor mechanical parameter identification based on the extended disturbance observer in the present invention can simultaneously identify the load torque and the moment of inertia online, and has strong robustness. It can also be seen from the experimental results in the verification example that the method of the present invention can simultaneously identify the two parameters, and both parameter identifications have a faster convergence speed and higher identification accuracy.
[0095] The above description of the embodiments is to facilitate the understanding and application of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments. Improvements and modifications made by those skilled in the art to the present invention based on the disclosure of the present invention should be within the protection scope of the present invention.
Claims
1. A method for identifying mechanical parameters of a permanent magnet synchronous motor based on an extended disturbance observer comprises the following steps: (1) Preliminary acquisition of the mechanical and electrical parameters of the permanent magnet synchronous motor, and the actual mechanical position angle θ of the permanent magnet synchronous motor is collected through the encoder and current sensor respectively. m and the three-phase stator current i a 、i b and i c , and then calculate the mechanical angular velocity ω of the permanent magnet synchronous motor m and the d-axis stator current i d and the q-axis stator current i q ; (2) Taking the load torque T of the permanent magnet synchronous motor as an example L And the inverse of the moment of inertia v is used as the extended state variable to construct the extended state equation of the permanent magnet synchronous motor, which is expressed as follows: T f =Bω m +Csgn(ω m ) in: T f and T e are the friction torque and electromagnetic torque of the permanent magnet synchronous motor, T e =1.5pψ f i q , v = 1 / J, sgn() represents the sign function, They are ω m , T L , the first-order derivative of v, p, ψ f , B, and C are the motor pole pair number, flux linkage, viscous friction coefficient, and Coulomb friction coefficient of the permanent magnet synchronous motor, respectively; (3) With mechanical angular velocity ω m , load torque T L And the inverse of the moment of inertia v is used as the observation object to construct the speed estimation equation of the extended disturbance observer, which is expressed as follows: in: They are ω m , v, T L The observed value of for The first-order derivative of , k is the speed estimation model gain, and e1 represents the mechanical angular velocity estimation error; (4) The speed estimation equation is used to make a difference with the extended state equation to obtain the relationship between the load torque estimation error and the inverse moment of inertia estimation error. The specific implementation method is as follows: First, the speed estimation equation is subtracted from the extended state equation to obtain the estimated error relationship of the first-order derivative of the mechanical angular velocity as follows: Then, according to the above estimation error relationship, the relationship between the load torque estimation error and the inverse moment of inertia estimation error is converted as follows: in: is the first-order derivative of e1, and e2 is the load torque estimation error, that is, e3 The estimation error of the inverse moment of inertia is (5) According to the relationship in step (4), the extended perturbation observation model is obtained by transformation, and its expression is as follows: Where: k t and k v are the given load torque gain and moment of inertia gain respectively, and They are and The first derivative of ; (6) Using the extended disturbance observation model to realize the load torque T L and moment of inertia J are observed.
2. The method for mechanical parameter identification of a permanent magnet synchronous motor based on an extended disturbance observer according to claim 1 is characterized in that: The mechanical parameters include the number of motor pole pairs p, the initial value of the moment of inertia J(0), the viscous friction coefficient B and the Coulomb friction coefficient C. The electrical parameters include the d-axis stator inductance L d , q-axis stator inductance L q , stator resistance R s and magnetic flux ψ f .
3. The method for mechanical parameter identification of a permanent magnet synchronous motor based on an extended disturbance observer according to claim 1 is characterized in that: In step (1), the mechanical angular velocity ω is calculated by the following formula: m : Wherein: t represents time.
4. The method for mechanical parameter identification of a permanent magnet synchronous motor based on an extended disturbance observer according to claim 1 is characterized in that: In step (1), the d-axis stator current i is calculated by the following formula: d and the q-axis stator current i q : Where: M abc / dq is the transformation matrix from the abc three-phase coordinate system to the dq two-phase coordinate system, ω e is the electromagnetic angular velocity of the permanent magnet synchronous motor and ω e =pω m , p is the number of motor pole pairs, and t represents time.
5. The method for mechanical parameter identification of a permanent magnet synchronous motor based on an extended disturbance observer according to claim 1, characterized in that: The optimization expression of the extended perturbation observation model is as follows: Where: k t is the given load torque gain, and They are and The first-order derivative of , q1 is the moment of inertia trigger, ρ is the given trigger proportional gain, δ is the given trigger threshold, l m As the trigger factor, is the given mechanical angular velocity reference value.
6. The method for mechanical parameter identification of a permanent magnet synchronous motor based on an extended disturbance observer according to claim 5 is characterized in that: In step (6), the load torque T L The specific expression for observing the moment of inertia J is as follows: in: is the observed value of the moment of inertia J, T L (0) is the initial value of the load torque and T L (0)=0, v(0)=1 / J(0), J(0) is the initial value of the moment of inertia.
Citation Information
Patent Citations
Method and apparatus for identifying inertial parameters of object to be identified
CN105259786A
Method and device for identifying dynamic rotational inertia of motor servo system, and motor servo system
CN110266232A
Method for identifying parameters of permanent magnet synchronous motor based on second-order sliding mode observer
CN110557070A
Permanent magnet synchronous motor mechanical parameter identification method based on extended sliding mode observer
CN114006557A
Parameter identification method for permanent magnet synchronous motor
CN114204864A
