CN109639199A - Suppression method of PMSM speed and torque ripple under three-phase asymmetric fault - Google Patents

Abstract

The present invention discloses permanent magnet synchronous motor revolving speed, method for suppressing torque ripple under a kind of asymmetrical three-phase failure, according under malfunction, the characteristics of revolving speed, torque cycle pulsative oscillation, the method of the invention designs a kind of iterative learning controller based on rotor-position, a kind of iterative law of learning with forgetting factor is constructed, and convergence is carried out to iterative learning controller.By to quadrature axis reference currenti _qrefPeriodical compensation is carried out, realizes that revolving speed, torque tightly follow given value.The method of the invention is applicable not only under malfunction motor constant speed and operates, revolving speed when for variable-speed operation, torque periodic swinging also have good inhibitory effect.

Description

Suppression method of PMSM speed and torque ripple under three-phase asymmetric fault

technical field

The invention relates to the field of new energy vehicle motor drive, in particular to a method for suppressing rotational speed and torque ripple of a permanent magnet synchronous motor under a three-phase asymmetric fault.

Background technique

In recent years, with the deepening of people's understanding of the concept of environmental protection, governments around the world have vigorously developed and supported new energy electric vehicles. Permanent Magnet Synchronous Motor (PMSM) is widely used in new energy vehicles due to its advantages of high torque density, high power density and high performance. When a three-phase asymmetric fault occurs in a permanent magnet synchronous motor, the rotational speed and torque will pulsate and oscillate, which will affect the safety of the entire vehicle, and even endanger personal safety, resulting in irreversible serious consequences.

So far, the methods of speed and torque ripple suppression are mainly concentrated in the mathematical model of the motor in the symmetrical state, and there has not been much research and analysis on the motor with asymmetrical three-phase stator windings. Moreover, the pulsation suppression of permanent magnet synchronous motors in the traditional sense is mainly based on the Iterative Learning Control (ILC) strategy in the time domain. The iteration period is generally proportional to the rotor rotation period, so it is limited to constant speed motion.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to provide a method for suppressing the rotational speed and torque pulsation under the asymmetrical fault state of the three-phase stator winding of a permanent magnet synchronous motor, which can achieve pulsation suppression no matter how the rotational speed changes.

Description of drawings

Fig. 1 is the overall block diagram of the vector control of ILC speed and torque ripple suppression based on rotor position.

Figure 2 is a block diagram of an iterative learning controller based on rotor position.

Detailed ways

The present invention will be described in detail below with reference to the accompanying drawings. The method for suppressing the rotational speed and torque ripple of a permanent magnet synchronous motor under a three-phase asymmetric fault of the present invention includes the following steps.

S01) Select the iteration period

Through the torque oscillation analysis of the speed and torque of the permanent magnet synchronous motor, it can be found that no matter how the speed changes, there is a periodic relationship between the speed and torque and the rotor position θ _m in the three-phase asymmetric state. Considering that the rotational speed and torque pulsation are periodic functions of θ _m , the method of the present invention designs an iterative learning controller based on rotor position. The overall block diagram of ILC speed and torque ripple suppression vector control based on rotor position is shown in Figure 1. Different from normal vector control, an iterative learning controller related to the rotor position θ _m is connected in parallel with the speed loop. The current i _qref is periodically compensated.

S02) Constructing learning law

The ILC of the present invention constructs a PI-type closed-loop iterative learning law with forgetting factor, which is expressed as:

(1)

In formula (1), α is the forgetting factor, α ∈ [0,1]; Δ i _q,k+1 ( θ _m ) and Δ i _q,k ( θ _m ) represent the k +1th and kth times, respectively q-axis reference compensation current during iteration, e _k+1 ( θ _m ) is the motor speed deviation at the k +1th iteration, θ _m is the mechanical angle the motor has rotated, θ _m ∈[0,2π], subscript k is the number of iterations, k = 1, 2, 3...; K ₀ and K ₁ are the learning gains of the proportional term and integral term of e _k+1 ( θ _m ), respectively.

The block diagram of the iterative learning controller based on rotor position is shown in Figure 2, where n _r * and n _{r , k+1} are the expected speed in rpm and the feedback value of the motor speed at the k +1th iteration, respectively; e _{n , k+1} ( θ _m ) and e _k+1 ( θ _m ) are the motor speed errors in rpm and rad/s units at the k +1th iteration, respectively; K ₃ =π/30; Memory is the memory used for Stores the compensation current for k iterations.

S03) Determine the convergence condition

When the friction coefficient of the motor is 0, the following function can be constructed according to the equation of motion:

(2)

Among them, ω _{r ,k} ( θ _m ) is the motor speed; J is the motor rotational inertia; T _L is the load torque; y _k ( θ _m ) system output; b=3p _n Ψ _f / 2 ; p _n is the motor pole logarithm; Ψ _f is the permanent magnet flux linkage.

From formula (2), we get:

(3)

The tracking error is:

(4)

In formula (4), ω _r *( θ _m ) and ω _{r ,k} ( θ _m ) are the expected speed in rad/s and the feedback value of the motor speed at the k +1th iteration, respectively.

From formula (2), formula (3) and formula (4), we get:

(5)

From formula (5), we can get:

(6)

According to the definition of infinite norm and the absolute value inequality theorem, we can get:

(7)

After finishing formula (7), we can get:

(8)

For ILC convergence, equation (8) must satisfy the following two constraints:

a) and convergence;

b) order , then |λ|<1;

Obviously, and Convergence to the constants C ₁ and C ₂ , where C ₁ ∈[0,2π] and C ₂ ∈[0,2π], so the constraint condition a is satisfied; in conclusion, the iterative learning controller based on the rotor position only needs to converge Satisfying condition b, the constraint can be simplified to:

(9)

The value range of the forgetting factor α is generally between 0.05 and 0.1, which is used to ensure the robustness of the system controller; reasonable K ₀ and K ₁ should not only ensure the learning convergence, but also ensure the convergence speed.

The above description is only the basic principle of the present invention, and improvements and substitutions made by those skilled in the art according to the present invention belong to the protection scope of the present invention.

Claims (1)

1. A PMSM rotation speed torque ripple suppression method under three-phase asymmetric fault is characterized in that: the method comprises the following steps:

s01) considering that the rotation speed and the torque ripple areθ _mThe method designs an iterative learning controller based on the rotor position; ILC speed and torque ripple suppression vector control based on rotor position is generally shown in FIG. 1. unlike normal vector control, a speed loop is connected in parallel with a rotor positionθ _mAssociated iterative learning controller for referencing q-axis parametersExamination currenti _qrefCarrying out periodic compensation;

s02) the method constructs a PI type closed loop iterative learning law with forgetting factors, which is expressed as:

（1）

in the formula (1), the reaction mixture is,αis a factor of forgetting to forget,α∈[0,1]；Δi _q,k+1(θ _m) And Δi _q,k(θ _m) Respectively representk+1 times andkthe q-axis reference compensation current at the time of the sub-iteration,e _k+1(θ _m) Is the firstkThe motor speed deviation in +1 iteration,θ _mis the mechanical angle that the motor has rotated through,θ _m∈[0,2π]subscriptkIs the number of iterations that are to be performed,k=1,2,3…；K₀and K₁Are respectively ase _k+1(θ _m) The learning gain of the proportional term and the integral term;

s03) determining a convergence condition

When the motor friction coefficient is 0, the following function can be constructed according to the motion equation:

（2）

wherein,ω _r,k(θ _m) Is the motor speed;Jis the rotational inertia of the motor;T _L is the load torque;y _k (θ _m) Outputting the system;b=3p _n Ψ _f /2；p _n the number of pole pairs of the motor is;Ψ _f is a permanent magnet flux linkage;

obtained by the formula (2):

（3）

the tracking error is:

（4）

in the formula (4), the reaction mixture is,ω _r *(θ _m) Andω _r,k(θ _m) Respectively the desired rotation speed and the second rotation speed in rad/s unitkThe motor rotating speed feedback value during +1 iteration;

is obtained by the following formula (2), formula (3) and formula (4):

（5）

from formula (5):

（6）

the method can be obtained according to the definition of infinite norm and the theorem of absolute value inequality:

（7）

the formula (7) is arranged to obtain:

（8）

ILC converges, equation (8) must satisfy the following two constraints:

a）andconverging;

b) order toThen | λ #<1;

It is clear that,andconverge on constant C₁And C₂Wherein, C₁∈[0,2π]And C₂∈[0,2π]So that the constraint a is satisfied; in summary, the iterative learning controller based on the rotor position only needs to satisfy the condition b, and the constraint condition can be simplified as follows:

（9）

forgetting factorαThe value range of (A) is generally between 0.05 and 0.1, and the robustness of a system controller is ensured; reasonable K₀And K₁Not only the learning convergence but also the convergence speed should be ensured.