Adaptive Offline Parameter Identification Method of Permanent Magnet Synchronous Motor Based on Hill Climbing Algorithm

Abstract

The present invention discloses an adaptive permanent magnet synchronous motor offline parameter identification method based on a ramp algorithm. The method utilizes the current loop of the motor to dynamically change the control parameters and the control loop to realize offline parameter identification. The parameters include stator resistance, direct-axis inductance, quadrature-axis inductance and rotor flux. Since vector control is widely used in permanent magnet synchronous motors, the present invention utilizes its own vector control loop for identification. Different from the traditional identification method, the identification order of the present invention is quadrature-axis inductance, rotor flux, direct-axis inductance and stator resistance. By changing the current loop PI link and the control loop, the above parameters can be accurately identified according to the mathematical model of the permanent magnet synchronous motor. Compared with the traditional identification method, the technical solution has excellent robustness, can avoid the coupling influence between parameters during the identification process, and improve the accuracy of identification.

Description

Technical Field

[0001] The present invention relates to the field of motor offline parameter identification, and in particular to an adaptive permanent magnet synchronous motor offline parameter identification method based on a ramp algorithm. Background Art

[0002] Permanent magnet synchronous motors are gaining more and more popularity in both small and medium power applications and high voltage and high power applications due to their high efficiency and easy control. However, no matter how different their application scenarios are, vector control is basically used in industry. Vector control often requires the conversion of three-phase electrical quantities to a synchronous rotating coordinate system for calculation, which results in the coupling of AC and DC axes in the motor model, making control difficult. In order to solve this problem in engineering, decoupling is often performed in the control loop to minimize the coupling effect of AC and DC axes. However, the decoupling of the control system requires the parameters of the motor, but in many cases, the parameters provided by the motor manufacturer are not accurate or are not provided. It can be seen that the precise control of parameter values ​​has a great impact on the performance of the control system.

[0003] At the same time, apart from the decoupling requirement, the adjustment of vector control itself also depends on the parameters of the motor. For example, in the adjustment of the control parameters of the motor inner loop, the motor is often equivalent to a first-order delay system, whose time constant is determined by the stator inductance and resistance. It can be seen that the adjustment of the PI controller is very dependent on the accurate understanding of the motor parameters. In addition, in the advanced control of the motor such as the sensorless vector control, weak magnetic control, maximum torque current ratio, etc., the parameters of the motor have a great influence on the control performance. Therefore, the offline identification of the permanent magnet synchronous motor parameters is particularly important.

[0004] Permanent magnet synchronous motors are often driven by frequency converters, and identification is performed before the motor is started and put into operation. In many cases, the motor needs to be identified when it is stationary. In general, the idea of ​​offline identification is to give the motor an excitation signal by controlling the frequency converter, and then measure the response and perform calculations to finally obtain the identification value. Depending on the excitation signal, measurement signal, calculation algorithm, nonlinear compensation, etc., the identification algorithm is also different, and each has its own advantages and disadvantages. However, most methods are one-time calculations, and some are repeated multiple times for accuracy and then averaged. In traditional offline identification methods, resistance often has a great influence on the results, and resistance is affected by conditions such as temperature. These factors combined often affect the accuracy of the identification results.

[0005] Therefore, it is particularly important to develop an efficient and accurate adaptive permanent magnet synchronous motor offline parameter identification method based on ramp algorithm. Summary of the invention

[0006] The purpose of the present invention is to provide an adaptive permanent magnet synchronous motor offline parameter identification method based on a ramp algorithm.

[0007] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0008] The present invention realizes offline identification of motor parameters by dynamically changing the control parameters and control loop of the permanent magnet synchronous motor current loop, wherein the motor parameters include stator resistance, quadrature-axis inductance, direct-axis inductance and rotor flux, and specifically includes the following steps:

[0009] The permanent magnet synchronous motor adopts vector control, and the quadrature and direct axes are completely decoupled in the control loop;

[0010] By changing the control circuit and the direct-axis PI controller, the quadrature-axis steady-state error is made zero, and the quadrature-axis inductance is calculated after the direct-axis has a steady-state error;

[0011] Identification of rotor flux is to modify the current loop control so that the direct axis uses a PI controller and the quadrature axis uses a low-gain P controller, and the rotor flux is identified after decoupling the direct axis inductance and flux;

[0012] The direct-axis inductance is identified by setting constraints based on the identified quadrature-axis inductance and rotor flux;

[0013] Calculate the stator resistance R, the formula is as follows:

[0014]

[0015] Where K dP is the proportional controller gain, is the given value of the direct axis current, i d is the actual direct axis current.

[0016] Furthermore, in the quadrature-axis inductance identification, when the rated value When the system reaches a steady state:

[0017]

[0018] Where ω is the electrical speed in rad / s, K dP is the proportional controller gain; i d is the actual direct-axis current, is the given value of the quadrature axis current, L q is the actual value of the quadrature-axis inductance, is the estimated value of quadrature-axis inductance;

[0019] The steady-state error of the quadrature axis depends on the difference between the actual and predicted quadrature axis inductance. During the identification process, the quadrature-axis inductance estimate is dynamically adjusted when i d = 0, that is,

[0020] Furthermore, before the rotor flux identification, the coupling term and the interference of the back electromotive force need to be eliminated before the flux identification, which specifically includes:

[0021] Give current instruction In steady state, there is d = 0, the coupling term is ignored

[0022] ωL d and In steady state, only The term affects the steady-state error of the current;

[0023] From this we get:

[0024]

[0025] Change the input proportional gain to K qP , and at the same time Change to Then we have:

[0026]

[0027] That is, when the stator resistance is unknown, the value of the first term in equation (4) is determined, and after the value of equation (5) is determined, it is deduced that:

[0028]

[0029] The steady-state error of the quadrature-axis current depends on the difference between the actual and predicted flux linkages.

[0030] During the identification process, the flux estimation value is dynamically adjusted when

[0031] i′ q = 0, that is,

[0032] Among them, i′ q is i minus (4) q , is the estimated value of magnetic flux, λ f is the actual value, K qp is the quadrature-axis proportional controller gain, R is the inductance, and ω is the electrical speed.

[0033] Furthermore, the constraint condition of the direct axis inductance is that the reference signal becomes and The direct-axis inductance is

[0034]

[0035] i n exist When the minimum value is reached, slowly adjust the estimated L d ;

[0036] Among them, i n is the modulus of the current, K dP is the direct-axis proportional controller gain, K qP is the quadrature-axis proportional controller gain; R is the inductance, ω c is the electrical speed, i d is the actual direct axis current, i q is the actual quadrature axis current, is the given value of direct axis current.

[0037] In a second aspect, an embodiment of the present application further provides an electronic device, including:

[0038] a processor; and a memory arranged to store computer-executable instructions.

[0039] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs.

[0040] The beneficial effects of the present invention are:

[0041] The present invention is an adaptive permanent magnet synchronous motor offline parameter identification method based on a ramp algorithm. Compared with the prior art, the present invention has the following technical effects:

[0042] 1. The present invention proposes an adaptive offline parameter identification method. The algorithm can consciously find the optimal solution of the motor parameters through a feedback loop. The result is accurate and has a wide range of applications.

[0043] 2. The present invention can solve the parameter identification problem of permanent magnet synchronous motors, thereby improving the dynamic performance of the control system. Compared with traditional identification methods, this technical solution has excellent robustness and can avoid the coupling effect between parameters during the identification process. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a permanent magnet synchronous motor model and a schematic diagram of control decoupling thereof of the present invention;

[0045] Figure 2 It is a schematic diagram of a circuit for identifying a quadrature-axis inductance control of the present invention;

[0046] Figure 3 is a schematic diagram of a series PI controller of the present invention;

[0047] Figure 4 is a schematic diagram of a steady-state direct-axis control loop when identifying a quadrature-axis inductance according to the present invention;

[0048] Figure 5 It is a transfer block diagram of a quadrature-axis steady-state system during flux linkage identification of the present invention;

[0049] Figure 6 It is a block diagram of the identification direct-axis inductance control of the present invention; DETAILED DESCRIPTION

[0050] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0051] The present invention provides an adaptive permanent magnet synchronous motor offline parameter identification method based on a ramp algorithm, comprising the following steps:

[0052] like Figure 1 As shown, in this embodiment, the present invention realizes offline identification of the current loop of the motor by dynamically changing the control parameters and the control loop, stator resistance, quadrature-axis inductance, direct-axis inductance and rotor flux, specifically including the following steps:

[0053] Use vector control to transform the coordinate system into the synchronous coordinate system;

[0054] Complete decoupling of the quadrature and direct axes in the control loop;

[0055] By changing the control circuit, the quadrature axis steady-state error is made zero, and the quadrature axis inductance is calculated after the direct axis has a steady-state error;

[0056] Identifying the rotor flux includes modifying the current loop control so that the direct axis adopts a PI controller and the quadrature axis adopts a low-gain P controller, and then identifying the rotor flux;

[0057] The direct-axis inductance is identified by setting constraints based on the identified quadrature-axis inductance and rotor flux;

[0058] The stator resistance is calculated based on the direct-axis inductance, quadrature-axis inductance and rotor flux.

[0059] The present invention intends to use the current loop of the motor itself to identify the parameters of the permanent magnet synchronous motor offline by dynamically changing the control parameters and the control loop. The parameters of the permanent magnet synchronous motor are mainly stator resistance, quadrature-axis inductance, direct-axis inductance and rotor flux. The identification methods of the present invention are described below:

[0060] Example 1 Regarding the quadrature-axis inductance L q Calculation

[0061] The quadrature-axis inductance identification method of the present invention mainly utilizes the structure of the vector control current loop itself and dynamically adjusts the control parameters through a processor to achieve the purpose of parameter identification.

[0062] like Figure 1As shown in the figure, the right half is the permanent magnet motor model. q The axes are coupled to each other, where ωL q and ωL d is a coupling term. In order to achieve independent control of the independent quadrature and direct axes, decoupling is desired. The usual practice is to perform decoupling in the control loop. Taking the direct axis (d axis) as an example, adding and subtracting a in Represents an estimated value.

[0063] Since vector control is performed in a synchronous coordinate system, the current reference signal is theoretically a DC quantity, and zero steady-state error can be achieved through a traditional PI controller.

[0064] By changing Figure 1 for Figure 2 , it can be seen that the direct axis (d axis) has removed the integral link, and only retains the proportional link with lower gain, while the quadrature axis (q axis) is still retained as the PI link. The purpose of this is that when the system reaches a steady state, the quadrature axis can achieve zero error, while the direct axis has a steady-state error, so that the quadrature axis inductance can be calculated. The above steps can be achieved by simply changing the processor program.

[0065] For the tuning of the quadrature axis PI controller, the following Figure 3 The parameter adjustment of the series PI controller follows the principle of zero-pole cancellation, that is, to ensure that the current loop is an integral link, which is a setting principle generally followed in vector control. Based on this, the parameters of the quadrature-axis PI controller are:

[0066]

[0067] K p =ω c L q (2)

[0068] where ω c is the bandwidth of the current loop, in rad / s.

[0069] By adjusting the current loop in this way, the steady-state error of the quadrature axis (q-axis) can be made zero, so that the coupling effect is reflected on the direct axis (d-axis). When the system is in steady state, the quadrature axis is completely decoupled, and the direct axis control loop is equivalent to Figure 3 , since the output current can be considered to be a stable value in steady state, the poles of the motor model can be ignored.

[0070] The input to the system is: is the rated value, When the system reaches a steady state, it can have:

[0071]

[0072] Where ω is the electrical speed in rad / s, K dP is the proportional controller gain.

[0073] It can be seen that the steady-state error of the quadrature axis depends on the difference between the actual value and the expected value of the quadrature axis inductance. That is to say, during the identification process, the estimated value is dynamically adjusted. when i d = 0, that is,

[0074] It can be seen that compared with the traditional identification method, this method has a specific advantage that it does not depend on other parameters K dP , ω or R.

[0075] Example 2 Regarding rotor flux λ f Calculation

[0076] Further modify the current loop through the program, that is, replace Figure 2 In the controller shown in the figure, the quadrature axis uses a low-gain P controller, while the direct axis uses a PI controller. At this time, the steady-state error of the direct axis is zero, while for the quadrature axis, there is always a steady-state error due to the coupling and back electromotive force. Because at this time there are both and therefore Figure 3 It needs to be identified in two steps:

[0077] First, give the current instruction In steady state, there is d = 0, the coupling term ωL can be ignored d and Then in steady state, the entire cross axis is equivalent to Figure 5 , at this time only The term can affect the steady-state error of the current.

[0078] From this we can get:

[0079]

[0080] By observing the equation, we can see that the first term is an interference term and its influence needs to be eliminated. q , that is, it can completely eliminate i d The effect on the q-axis can be used to clarify the effect of the first term in the equation. Figure 4 It can be seen that if the proportional gain is changed to K qP , and at the same time Change to Then we have:

[0081]

[0082] That is, the value of the first term in equation (4) can be determined without knowing the stator resistance.

[0083] After the value of equation (5) is determined, it can be subtracted from equation (4) to obtain:

[0084]

[0085] Obviously, we can now identify L q The same way to identify λ f .

[0086] Direct axis inductance L d

[0087] After completing the first two steps, you can proceed to the direct-axis inductance L d 's identification, at this time still based on Figure 5 Set up the control system, and the reference signal given to the system becomes

[0088] and The control system now becomes Figure 6 .

[0089] The direct-axis inductance is

[0090]

[0091]

[0092] You can see that I n exist When it reaches the minimum value, then we can estimate L by slowly adjusting d This identification method can also avoid the interference of R and has excellent robustness.

[0093] Example 4 Calculation of stator resistance R

[0094] According to equation (8), we can get

[0095]

[0096] Obviously, R can be calculated directly.

[0097] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0098] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0099] 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, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive permanent magnet synchronous motor offline parameter identification method based on a ramp algorithm, characterized in that: Offline identification of motor parameters is achieved by dynamically changing the control parameters of the permanent magnet synchronous motor current loop and the control loop. The motor parameters include stator resistance, quadrature-axis inductance, direct-axis inductance and rotor flux. Specifically, the following steps are included: The permanent magnet synchronous motor adopts vector control, and the quadrature and direct axes are completely decoupled in the control loop; By changing the control circuit and the direct-axis PI controller, the quadrature-axis steady-state error is made zero, and the quadrature-axis inductance is calculated after the direct-axis has a steady-state error; Identification of rotor flux is to modify the current loop control so that the direct axis uses a PI controller and the quadrature axis uses a low-gain P controller, and the rotor flux is identified after decoupling the direct axis inductance and flux; The direct-axis inductance is identified by setting constraints based on the identified quadrature-axis inductance and rotor flux; Calculate the stator resistance R, the formula is as follows: Where K dP is the proportional controller gain, is the given value of the direct axis current, i d is the actual direct axis current.

2. The method for offline parameter identification of an adaptive permanent magnet synchronous motor based on a ramp algorithm according to claim 1 is characterized in that: In the quadrature-axis inductance identification, when the rated value When the system reaches a steady state: Where ω is the electrical speed in rad / s, K dP is the proportional controller gain; i d is the actual direct-axis current, is the given value of the quadrature axis current, L q is the actual value of the quadrature-axis inductance, is the estimated value of quadrature-axis inductance; The steady-state error of the quadrature axis depends on the difference between the actual and predicted quadrature axis inductance. During the identification process, the quadrature-axis inductance estimate is dynamically adjusted when i d = 0, that is, 3. The method for offline parameter identification of an adaptive permanent magnet synchronous motor based on a ramp algorithm according to claim 1 is characterized in that: Before identifying the rotor flux, it is necessary to eliminate the interference of coupling terms and back electromotive force and then identify the flux, including: Give current instruction In steady state, there is d = 0, the coupling term ωL is ignored d and In steady state, only The term affects the steady-state error of the current; From this we get: Change the input proportional gain to K qP , and at the same time Change to Then we have: That is, when the stator resistance is unknown, the value of the first term in equation (4) is determined, and after the value of equation (5) is determined, it is deduced that: The steady-state error of the quadrature-axis current depends on the difference between the actual and predicted flux linkages. During the identification process, the flux estimation value is dynamically adjusted when i′ q = 0, that is, Among them, i′ q is i minus (4) q , is the estimated value of magnetic flux, λ f is the actual value, K qP is the quadrature-axis proportional controller gain, R is the inductance, and ω is the electrical speed.

4. The method for offline parameter identification of an adaptive permanent magnet synchronous motor based on a ramp algorithm according to claim 1 is characterized in that: The constraint condition of the direct axis inductance is that the reference signal becomes and The direct-axis inductance is i n exist When the minimum value is reached, slowly adjust the estimated L d ; Among them, i n is the modulus of the current, K dP is the direct-axis proportional controller gain, K qP is the quadrature-axis proportional controller gain; R is the inductance, ω c is the electrical speed, i d is the actual direct axis current, i q is the actual quadrature axis current, is the given value of direct axis current.

5. An electronic device comprising: processor; and a memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1 to 4.

6. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute any one of the methods of claims 1 to 4.

Images