CN113872476B - PI parameter optimization method and device for doubly-fed induction motor - Google Patents

Abstract

The application relates to the technical field of power system control, and discloses a PI parameter optimization method and device of a doubly-fed induction motor. In the method, operation parameters of a doubly-fed induction motor are firstly obtained, and a doubly-fed induction motor system model is constructed, wherein in the doubly-fed induction motor system model, a double-loop PI control frame is adopted for the angular speed of a generator rotor and the reactive power of the generator. And then updating the PI parameters by using an IASO algorithm according to the doubly-fed induction motor system model. And finally, feeding back the generator rotor voltage output by the PI controller to a doubly-fed induction motor system model for the next iteration until the algorithm converges to obtain the optimal PI parameter. The method and the device can accurately optimize all PI parameters through the steps, have short optimizing time and high optimizing precision, and effectively solve the technical problems in the prior art.

Description

PI parameter optimization method and device for doubly-fed induction motor

Technical Field

The application relates to the technical field of power system control, in particular to a PI parameter optimization method and device of a doubly-fed induction motor.

Background

In recent years, fossil energy crisis and environmental change are severe, and energy structure transformation is promoted in countries around the world. The wind power generation has the remarkable advantages of cleanness, no pollution, abundant resources and the like, and is rapidly developed.

With the rapid development of wind power generation technology, doubly-fed induction motors (DFIGs) are widely used in the modern wind power generation industry with their excellent performance. The doubly-fed induction motor can not only realize the adjustment of the active power of the power grid, but also adjust the reactive power of the power grid.

The current conventional maximum power tracking method of the doubly-fed induction motor is realized based on a double-loop PI controller. However, parameters of the PI controller in the doubly-fed induction motor are mutually affected, and in actual operation, the parameters of the PI controller are mostly optimized by experience of staff, so that the optimization effect is poor.

Disclosure of Invention

The application discloses a PI parameter optimization method and device of a doubly-fed induction motor, which are used for solving the technical problems that in the prior art, a conventional doubly-fed induction motor maximum power tracking method is realized based on a double-loop PI controller, however, parameters of the PI controller in the doubly-fed induction motor are mutually influenced, and in actual operation, the parameters of the PI controller are optimized mostly by virtue of experience of staff, so that the optimization effect is poor.

The first aspect of the application discloses a PI parameter optimization method of a doubly-fed induction motor, which comprises the following steps:

Obtaining doubly-fed induction machine operating parameters, the doubly-fed induction machine operating parameters comprising: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage and generator rotor q-axis voltage;

determining an objective function of the PI parameter according to the operation parameter of the doubly-fed induction motor and a preset limiting condition;

generating a doubly-fed induction machine system model according to the objective function, a pre-built wind turbine model, a pre-built generator model and a pre-built rotating shaft system model;

updating the PI parameters by using an IASO algorithm according to the doubly-fed induction motor system model;

updating the generator reactive power error and the generator rotor angular speed error;

updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular speed error of the generator rotor and the updated PI parameter;

updating the doubly-fed induction motor system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor;

and carrying out iterative updating on the doubly-fed induction motor system model to determine the optimal PI parameter.

Optionally, the updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular speed error of the generator rotor and the updated PI parameter includes:

determining a generator rotor d-axis current reference value and a generator rotor q-axis current reference value according to the updated generator reactive power error and the updated generator rotor angular speed error;

and updating the generator rotor d-axis voltage and the generator rotor q-axis voltage according to the generator rotor d-axis current reference value, the generator rotor q-axis current reference value, a preset generator rotor d-axis voltage compensation term and a preset generator rotor q-axis voltage compensation term.

Optionally, the updating the doubly-fed induction machine system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor includes:

updating the objective function according to the updated generator rotor d-axis voltage and the updated generator rotor q-axis voltage;

and updating the doubly-fed induction motor system model according to the updated objective function.

Optionally, the updating the doubly-fed induction machine system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor includes:

and feeding back the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor to the doubly-fed induction motor system model for updating after pulse width modulation.

Optionally, the limiting condition includes: an outer loop proportional gain limit, an outer loop integral gain limit, an inner loop proportional gain limit, an inner loop integral gain limit, a wind speed limit, a grid voltage limit, a reactive power limit, a generator rotor d-axis voltage limit, and a generator rotor q-axis voltage limit.

Optionally, the step of constructing the wind turbine model comprises:

acquiring the mechanical rotation speed of a wind turbine, the radius of the wind turbine and the wind speed, and determining the tip speed ratio according to the mechanical rotation speed of the wind turbine, the radius of the wind turbine and the wind speed;

obtaining a pitch angle, and determining a wind energy utilization coefficient according to the tip speed ratio and the pitch angle;

generating the wind turbine model based on the air density, the wind turbine radius, the wind energy utilization coefficient, and the wind speed.

Optionally, the step of constructing the generator model includes:

acquiring a mechanical rotation speed of a generator, a synchronous angular speed of the generator, an angular speed of a rotor of the generator, an internal voltage of a d-axis of the generator, an internal voltage of a q-axis of the generator, a stator current of a d-axis of the generator, a stator terminal voltage of a q-axis of the generator, a rotor terminal voltage of a d-axis of the generator, a rotor terminal voltage of a q-axis of the generator, mutual inductance between stator and rotor of the generator, stator inductance of the generator, rotor inductance of the generator, stator resistance of the generator and rotor resistance of the generator, and determining a dynamic equation of the generator;

determining electromagnetic torque generated by a generator according to the generator d-axis internal voltage, the generator q-axis internal voltage, the generator synchronous angular speed, the generator d-axis stator current and the generator q-axis stator current;

determining reactive power of the generator according to the generator d-axis stator terminal voltage, the generator q-axis stator terminal voltage, the generator d-axis stator current and the generator q-axis stator current;

and generating the generator model according to the generator dynamic equation, the electromagnetic torque generated by the generator and the reactive power of the generator.

Optionally, the step of constructing the rotating shaft system model includes:

determining an equivalent inertia constant of the rotating shaft system according to a preset wind turbine inertia constant and a preset generator inertia constant;

determining a mechanical torque from the wind turbine model and the generator rotor angular speed;

and generating the rotating shaft system model according to the angular speed of the generator rotor, the preset integrated damping of the concentrated inertia system, the mechanical torque and the electromagnetic torque generated by the generator.

Optionally, the updating the PI parameter according to the doubly-fed induction machine system model by using an IAS0 algorithm includes:

for any PI parameter, acquiring a plurality of initial candidate parameters of the any PI parameter according to a preset first random number;

determining candidate parameter objective function values according to the doubly-fed induction motor system model and a plurality of initial candidate parameters;

determining a current optimal candidate parameter and a current worst candidate parameter according to the candidate parameter objective function value;

for any initial candidate parameter of any PI parameter, acquiring an interaction candidate parameter from a plurality of initial candidate parameters of the any PI parameter, wherein the interaction candidate parameter refers to any initial candidate parameter interacted with the any initial candidate parameter;

Determining a Euclidean distance of a candidate parameter according to the any initial candidate parameter and the interaction candidate parameter;

determining a minimum distance ratio of candidate parameters according to the candidate parameter objective function value and the current optimal candidate parameter;

determining a candidate parameter distance ratio according to the candidate parameter Euclidean distance, the candidate parameter objective function value, the plurality of initial candidate parameters, the current optimal candidate parameter, the candidate parameter minimum distance ratio and a preset candidate parameter maximum distance ratio;

determining interaction force among candidate parameters according to the candidate parameter distance ratio, wherein the interaction force among candidate parameters is the interaction force between any initial candidate parameter and the interaction candidate parameter;

determining the total interaction force of any initial candidate parameter according to the interaction force among the candidate parameters and the plurality of initial candidate parameters;

obtaining a fixed key length from any initial candidate parameter to the current optimal candidate parameter, and determining geometric constraint of any initial candidate parameter according to the any initial candidate parameter, the current optimal candidate parameter and the fixed key length from any initial candidate parameter to the current optimal candidate parameter;

Determining constraint force of any initial candidate parameter according to geometric constraint of the any initial candidate parameter;

determining the quality of any initial candidate parameter according to the candidate parameter objective function value, the current optimal candidate parameter, the current worst candidate parameter, any PI parameter and the interaction candidate parameter;

updating the acceleration of any initial candidate parameter according to the total interaction force of any initial candidate parameter, the constraint force of any initial candidate parameter and the mass of any initial candidate parameter;

acquiring the initial speed of any initial candidate parameter, and updating the initial speed of any initial candidate parameter according to the initial speed of any initial candidate parameter, the acceleration of any initial candidate parameter and a preset random array;

updating any initial candidate parameter according to the updated speed of the any initial candidate parameter;

and updating the PI parameter according to any updated initial candidate parameter.

The second aspect of the application discloses a PI parameter optimization apparatus for a doubly-fed induction motor, where the PI parameter optimization apparatus for a doubly-fed induction motor is applied to the PI parameter optimization method for a doubly-fed induction motor disclosed in the first aspect of the application, and the PI parameter optimization apparatus for a doubly-fed induction motor includes:

The operation parameter acquisition module is used for acquiring operation parameters of the doubly-fed induction motor, and the operation parameters of the doubly-fed induction motor comprise: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage and generator rotor q-axis voltage;

the objective function determining module is used for determining an objective function of the PI parameter according to the operation parameter of the doubly-fed induction motor and a preset limiting condition;

the doubly-fed induction motor system model generation module is used for generating a doubly-fed induction motor system model according to the objective function, a pre-built wind turbine model, a pre-built generator model and a pre-built rotating shaft system model;

the PI parameter updating module is used for updating the PI parameter by using an IAS0 algorithm according to the doubly-fed induction motor system model;

the error updating module is used for updating the reactive power error of the generator and the angular speed error of the rotor of the generator;

the generator rotor voltage updating module is used for updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular speed error of the generator rotor and the updated PI parameter;

The system model updating module is used for updating the doubly-fed induction motor system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor;

and the PI parameter optimization module is used for carrying out iterative updating on the doubly-fed induction motor system model to determine the optimal PI parameter.

Optionally, the generator rotor voltage update module includes:

a rotor current reference value determining unit, configured to determine a generator rotor d-axis current reference value and a generator rotor q-axis current reference value according to the updated generator reactive power error and the updated generator rotor angular speed error;

and the rotor voltage updating unit is used for updating the generator rotor d-axis voltage and the generator rotor q-axis voltage according to the generator rotor d-axis current reference value, the generator rotor q-axis current reference value, a preset generator rotor d-axis voltage compensation term and a preset generator rotor q-axis voltage compensation term.

Optionally, the system model updating module includes:

an objective function updating unit, configured to update the objective function according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor;

And the system model updating unit is used for updating the doubly-fed induction motor system model according to the updated objective function.

Optionally, the system model updating module includes:

and the pulse width modulation unit is used for feeding the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor back to the doubly-fed induction motor system model for updating after pulse width modulation.

Optionally, the doubly-fed induction machine system model generating module includes:

the blade tip speed ratio acquisition unit is used for acquiring the mechanical rotation speed of the wind turbine, the radius of the wind turbine and the wind speed, and determining the blade tip speed ratio according to the mechanical rotation speed of the wind turbine, the radius of the wind turbine and the wind speed;

the wind energy utilization coefficient acquisition unit is used for acquiring a pitch angle and determining a wind energy utilization coefficient according to the tip speed ratio and the pitch angle;

a wind turbine model generation unit for generating the wind turbine model based on the air density, the wind turbine radius, the wind energy utilization coefficient and the wind speed.

Optionally, the doubly-fed induction machine system model generating module includes:

the generator dynamic equation determining unit is used for obtaining the mechanical rotation speed of the generator, the synchronous angular speed of the generator, the angular speed of the generator rotor, the internal voltage of the generator d axis, the internal voltage of the generator q axis, the stator current of the generator d axis, the stator end voltage of the generator q axis, the rotor end voltage of the generator d axis, the rotor end voltage of the generator q axis, the mutual inductance between the stator and the rotor of the generator, the stator inductance of the generator, the rotor inductance of the generator, the stator resistance of the generator and the rotor resistance of the generator, and determining the generator dynamic equation;

An electromagnetic torque determining unit configured to determine an electromagnetic torque generated by a generator based on the generator d-axis internal voltage, the generator q-axis internal voltage, the generator synchronous angular velocity, a generator d-axis stator current, and the generator q-axis stator current;

the generator reactive power determining unit is used for determining the generator reactive power according to the generator d-axis stator terminal voltage, the generator q-axis stator terminal voltage, the generator d-axis stator current and the generator q-axis stator current;

and the generator model generating unit is used for generating the generator model according to the generator dynamic equation, the electromagnetic torque generated by the generator and the generator reactive power.

Optionally, the doubly-fed induction machine system model generating module includes:

the equivalent inertia constant acquisition unit is used for determining an equivalent inertia constant of the rotating shaft system according to a preset wind turbine inertia constant and a preset generator inertia constant;

a mechanical torque acquisition unit for determining a mechanical torque from the wind turbine model and the generator rotor angular speed;

and the rotating shaft system model generating unit is used for generating the rotating shaft system model according to the angular speed of the generator rotor, the preset integrated damping of the concentrated inertia system, the mechanical torque and the electromagnetic torque generated by the generator.

Optionally, the PI parameter updating module includes:

the initial candidate parameter acquisition unit is used for acquiring a plurality of initial candidate parameters of any PI parameter according to a preset first random number aiming at the PI parameter;

the objective function value calculation unit is used for determining candidate parameter objective function values according to the doubly-fed induction motor system model and a plurality of initial candidate parameters;

a candidate parameter screening unit, configured to determine a current optimal candidate parameter and a current worst candidate parameter according to the candidate parameter objective function value;

an interaction candidate parameter obtaining unit, configured to obtain, for any initial candidate parameter of the any PI parameter, an interaction candidate parameter from a plurality of initial candidate parameters of the any PI parameter, where the interaction candidate parameter is any initial candidate parameter having an interaction with the any initial candidate parameter;

the Euclidean distance determining unit is used for determining the Euclidean distance of the candidate parameter according to any initial candidate parameter and the interaction candidate parameter;

the minimum distance ratio obtaining unit is used for determining a candidate parameter minimum distance ratio according to the candidate parameter objective function value and the current optimal candidate parameter;

A candidate parameter distance ratio determining unit, configured to determine a candidate parameter distance ratio according to the candidate parameter euclidean distance, the candidate parameter objective function value, the plurality of initial candidate parameters, the current optimal candidate parameter, the candidate parameter minimum distance ratio, and a preset candidate parameter maximum distance ratio;

an interaction force obtaining unit, configured to determine an interaction force between candidate parameters according to the candidate parameter distance ratio, where the interaction force between candidate parameters is an interaction force between the any initial candidate parameter and the interaction candidate parameter;

a total interaction force obtaining unit, configured to determine a total interaction force of any one of the initial candidate parameters according to the interaction force between the candidate parameters and the plurality of initial candidate parameters;

the geometric constraint acquisition unit is used for acquiring the fixed key length from any initial candidate parameter to the current optimal candidate parameter, and determining the geometric constraint of any initial candidate parameter according to the any initial candidate parameter, the current optimal candidate parameter and the fixed key length from any initial candidate parameter to the current optimal candidate parameter;

A constraint force determining unit, configured to determine a constraint force of the any initial candidate parameter according to a geometric constraint of the any initial candidate parameter;

a quality obtaining unit, configured to determine a quality of the any initial candidate parameter according to the candidate parameter objective function value, the current optimal candidate parameter, the current worst candidate parameter, the any PI parameter, and the interaction candidate parameter;

the acceleration updating unit is used for updating the acceleration of any initial candidate parameter according to the total interaction force of any initial candidate parameter, the constraint force of any initial candidate parameter and the mass of any initial candidate parameter;

the initial speed updating unit is used for acquiring the initial speed of any initial candidate parameter and updating the initial speed of any initial candidate parameter according to the initial speed of any initial candidate parameter, the acceleration of any initial candidate parameter and a preset random array;

an initial candidate parameter updating unit, configured to update any initial candidate parameter according to the updated speed of the any initial candidate parameter;

And the PI parameter updating unit is used for updating the PI parameter according to any one of the updated initial candidate parameters.

The application relates to the technical field of power system control, and discloses a PI parameter optimization method and device of a doubly-fed induction motor. In the method, operation parameters of a doubly-fed induction motor are firstly obtained, and a doubly-fed induction motor system model is constructed, wherein in the doubly-fed induction motor system model, a double-loop PI control frame is adopted for the angular speed of a generator rotor and the reactive power of the generator. And then updating the PI parameters by using an IASO algorithm according to the doubly-fed induction motor system model. And finally, feeding back the generator rotor voltage output by the PI controller to a doubly-fed induction motor system model for the next iteration until the algorithm converges to obtain the optimal PI parameter. The method and the device can accurately optimize all PI parameters through the steps, have short optimizing time and high optimizing precision, and effectively solve the technical problems in the prior art.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.

Fig. 1 is a schematic workflow diagram of a PI parameter optimization method for a doubly-fed induction machine according to an embodiment of the present disclosure;

fig. 2 is a schematic structural diagram of a PI parameter optimizing apparatus for a doubly-fed induction machine according to an embodiment of the present disclosure.

Detailed Description

In order to solve the technical problems that in the prior art, the conventional maximum power tracking method of the doubly-fed induction motor is realized based on a double-loop PI controller, however, parameters of the PI controller in the doubly-fed induction motor are mutually influenced, in actual operation, the parameters of the PI controller are mostly optimized by experience of staff, and the optimization effect is poor, the application discloses a PI parameter optimization method and device of the doubly-fed induction motor through the following two embodiments.

The first embodiment of the application discloses a PI parameter optimization method of a doubly-fed induction motor, referring to a workflow diagram shown in fig. 1, the PI parameter optimization method of the doubly-fed induction motor includes:

step S101, obtaining operation parameters of a doubly-fed induction motor, wherein the operation parameters of the doubly-fed induction motor comprise: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage, and generator rotor q-axis voltage.

In some embodiments of the present application, the doubly-fed induction machine operating parameters include: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage, generator rotor q-axis voltage, wind speed, grid voltage, and reactive power.

Step S102, determining an objective function of the PI parameter according to the operation parameter of the doubly-fed induction motor and a preset limiting condition.

In some embodiments of the present application, the limiting conditions include: an outer loop proportional gain limit, an outer loop integral gain limit, an inner loop proportional gain limit, an inner loop integral gain limit, a wind speed limit, a grid voltage limit, a reactive power limit, a generator rotor d-axis voltage limit, and a generator rotor q-axis voltage limit.

The doubly-fed induction motor can realize variable-speed power generation, wind power resources with different wind speeds can be better utilized, and under the optimal control condition, the variable-speed wind power system can obtain the maximum wind power in a large wind speed range. The period runs at the maximum power point, that is, MPPT. To achieve the MPPT effect, the objective function value is required to be minimum within the constraints. Consider three examples: step wind speed, random wind speed and grid voltage drop, and constructing an objective function of PI parameters, as follows:

Wherein,representing generator reactive power error,/->Indicating the angular velocity error of the generator rotor, u _qr Represents the generator rotor q-axis voltage, u _dr Representing the generator rotor d-axis voltage, in a relatively slow outer control loop, the outer loop proportional gain (K _P1 ，K _P2 ) And integral gain (K) _I1 ，K _I2 ) Respectively limited to [0,0.5 ]]And [0,2 ]]And in a relatively fast inner control loop, the proportional gain (K _P3 ，K _P4 ) And integral gain (K) _I3 ，K _I4 ) The range of the values of (2) is limited to 0, 15 respectively]And [0, 50 ]]Between them. T is the total computation time for each case. Wind speed v _wind Varying between 8m/s and 12m/s, the grid voltage v _s Limited between 0.2p.u. and 1.0p.u., reactive power Q _s Limited to between-1.0 p.u. and 1.0p.u. Furthermore, the weight coefficient ω ₁ ＝ω ₂ ＝1/16。

Step S103, generating a doubly-fed induction motor system model according to the objective function, a pre-constructed wind turbine model, a pre-constructed generator model and a pre-constructed rotating shaft system model;

in some embodiments of the present application, the step of constructing the wind turbine model comprises:

obtaining a wind turbine mechanical speed, a wind turbine radius and a wind speed, and determining a tip speed ratio based on the wind turbine mechanical speed, the wind turbine radius and the wind speed.

And obtaining a pitch angle, and determining a wind energy utilization coefficient according to the tip speed ratio and the pitch angle.

Generating the wind turbine model based on the air density, the wind turbine radius, the wind energy utilization coefficient, and the wind speed.

In particular, in a wind turbine model, the mechanical power P captured by the wind turbine _m Is determined by the following formula:

wherein ρ represents air density, R represents wind turbine radius, v _wind Represents wind speed, C _p (lambda, beta) represents the function of the tip speed ratio lambda and the pitch angle beta, also known as the wind energy utilization coefficient, omega _m Representing the mechanical rotational speed of the wind turbine c ₁ ＝0.5176，c ₂ ＝116，c ₃ ＝0.4，c ₄ ＝5，c ₅ ＝21，c ₆ ＝0.0068。

In some embodiments of the present application, the step of constructing the generator model includes:

acquiring a mechanical rotation speed of a generator, a synchronous angular speed of the generator, an angular speed of a rotor of the generator, an internal voltage of a d-axis of the generator, an internal voltage of a q-axis of the generator, a stator current of a d-axis of the generator, a stator terminal voltage of a q-axis of the generator, a rotor terminal voltage of a d-axis of the generator, a rotor terminal voltage of a q-axis of the generator, mutual inductance between stator and rotor of the generator, stator inductance of the generator, rotor inductance of the generator, stator resistance of the generator and rotor resistance of the generator, and determining a dynamic equation of the generator;

Determining electromagnetic torque generated by a generator according to the generator d-axis internal voltage, the generator q-axis internal voltage, the generator synchronous angular speed, the generator d-axis stator current and the generator q-axis stator current;

determining reactive power of the generator according to the generator d-axis stator terminal voltage, the generator q-axis stator terminal voltage, the generator d-axis stator current and the generator q-axis stator current;

and generating the generator model according to the generator dynamic equation, the electromagnetic torque generated by the generator and the reactive power of the generator.

Specifically, in the generator model, the dynamic equation of the generator is as follows:

K _mrr ＝L _m /L _rr ；

R ₂ ＝R ₁ K _mrr ² ；

R ₁ ＝R ₂ +R _s ；

wherein omega _b Represents the mechanical rotation speed omega of the generator _s Represents synchronous angular velocity omega of generator _r Representing the angular speed of the generator rotor; e' _ds And e' _qs Respectively representing the internal voltage of the d axis of the generator and the internal voltage of the q axis of the generator; i.e _ds And i _qs Respectively representing the d-axis stator current of the generator and the q-axis stator current of the generator; u (u) _ds And u _qs Respectively representing the d-axis stator terminal voltage of the generator and the q-axis stator terminal voltage of the generator; u (u) _dr And u _qr Respectively representing the end voltage of the d-axis rotor of the generator and the end voltage of the q-axis rotor of the generator; l (L) _m Representing mutual inductance between stator and rotor of generator, L _rr Representing the inductance of the generator rotor, L _ss Represents the inductance of the stator of the generator, R _r Representing the resistance of the generator rotor, R _s Representing the generator stator resistance.

The electromagnetic torque produced by the generator is determined by the following formula:

T _e ＝(e′ _qs /ω _s )i _qs +(e′ _ds /ω _s )i _ds ；

wherein T is _e Representing the electromagnetic torque produced by the generator.

The generator reactive power is determined by the following formula:

Q _s ＝u _qs i _ds -u _ds i _qs ＝u _qs i _ds ；

in some embodiments of the present application, the step of constructing the spindle system model includes:

and determining an equivalent inertia constant of the rotating shaft system according to the preset wind turbine inertia constant and the preset generator inertia constant.

Determining a mechanical torque from the wind turbine model and the generator rotor angular speed.

And generating the rotating shaft system model according to the angular speed of the generator rotor, the preset integrated damping of the concentrated inertia system, the mechanical torque and the electromagnetic torque generated by the generator.

In particular, spindle systems typically employ equivalent inertial constantsThe number is H _m Modeling a centralized inertia system of (1), as follows

H _m ＝H _t +H _g ；

Wherein H is _t Indicating the inertia constant of the wind turbine, H _g Representing the generator inertia constant.

The electromechanical dynamic equation can be expressed as follows:

wherein omega _a Indicating the rotational angular velocity of the polymerization system, setting ω _a ＝ω _r The method comprises the steps of carrying out a first treatment on the surface of the D represents the comprehensive damping of the centralized inertia system, T _m For mechanical torque, satisfy T _m ＝P _m /ω _a 。

And generating a doubly-fed induction motor system model according to the wind turbine model, the generator model, the rotating shaft system model, the objective function and the limiting conditions.

And step S104, updating the PI parameters by using an IASO algorithm according to the doubly-fed induction motor system model.

Further, the updating the PI parameter by using an IASO algorithm according to the doubly-fed induction machine system model includes:

and for any PI parameter, acquiring a plurality of initial candidate parameters of the any PI parameter according to a preset first random number.

In some embodiments of the present application, the obtaining, according to a preset first random number, a plurality of initial candidate parameters of the arbitrary PI parameter includes:

and acquiring a plurality of initial candidate parameters of any PI parameter according to the preset PI parameter upper limit, the preset PI parameter lower limit and the first random number, wherein the first random number is a random number in a [0,1] interval.

In particular, the four coupled PI parameters, denoted K, respectively, need to be optimally adjusted _Pi And K _Ii . The parameters that need to be set here are: k (K) _P1 、K _P2 、K _I1 、K _I2 、K _P3 、K _P4 、K _I3 、K _I4 。

In some embodiments of the present application, all PI parameters are optimized using a modified atomic search optimization (improved atom search optimizer, IASO) algorithm.

Firstly, an initial population of an IASO algorithm is constructed based on an optimization space of atoms, the initial population is uniformly and randomly initialized in a solution space, and a plurality of initial candidate parameters of any PI parameter are determined through the following formula:

wherein,the ith initial candidate parameter of the PI parameter to be optimized is represented by the d th initial candidate parameter, the number n of the initial candidate parameters is determined according to the actual application scene, UB represents the upper limit of the PI parameter, LB represents the lower limit of the PI parameter, and r _id Is [0,1 ]]The random number of the interval, d, represents the number of PI parameters.

And determining candidate parameter objective function values according to the doubly-fed induction motor system model and a plurality of initial candidate parameters.

Specifically, the objective function value of the candidate parameter, i.e., the objective function value of each initial candidate parameter, is determined using the objective function.

And determining the current optimal candidate parameter and the current worst candidate parameter according to the candidate parameter objective function value.

Specifically, the current optimal candidate parameter X of any initial candidate parameter is determined according to the candidate parameter objective function value _best And the current worst candidate parameter X _worst The iteration process, x, is then performed _best，d (k) Current optimal candidate parameter, x, representing the PI parameter of the d-th desired optimization at the kth iteration _worst，d (k) The current worst candidate parameter for the PI parameter representing the d-th desired optimization at the kth iteration.

The current optimal candidate parameters are determined, and the next step is to update the initial candidate parameters of any initial candidate parameters for the next iteration.

For any initial candidate parameter of any PI parameter, an interaction candidate parameter is obtained from a plurality of initial candidate parameters of any PI parameter, where the interaction candidate parameter is any initial candidate parameter having interaction with the any initial candidate parameter.

And determining the Euclidean distance of the candidate parameter according to the any initial candidate parameter and the interaction candidate parameter.

In particular, the Lennard-Jones (L-J) potential energy is typically used to characterize the force between two interacting atoms. Determining a candidate parameter Euclidean distance by the following formula:

r _ij，d ＝||x _jd -x _id ||；

wherein x is _jd The jth initial candidate parameter, x, representing the d PI parameter _id The ith initial candidate parameter, r, representing the d-th PI parameter _ij，d The j-th initial candidate parameter and the euclidean distance of the i-th initial candidate parameter, which represent the d-th PI parameter.

And determining a minimum distance ratio of the candidate parameters according to the candidate parameter objective function value and the current optimal candidate parameter.

Specifically, the candidate parameter minimum distance ratio h _min (k) Is determined by the following formula:

wherein h is _min The initial value is 1.2, which is used to determine the candidate parameter minimum distance ratio at the first iteration.

The candidate parameter maximum distance ratio is set to 1.24. When the h value is less than 1.12, atoms tend to repel close atoms; conversely, an atom attracts a leaving atom. The improved IASO algorithm proposed in this embodiment must satisfy: when a better solution cannot be found at present, performing extensive global searching based on the current solution; if a better solution is found, a deep local exploration is made and thus an improvement in this section is made.

And determining a candidate parameter distance ratio according to the Euclidean distance of the candidate parameter, the candidate parameter objective function value, the plurality of initial candidate parameters, the current optimal candidate parameter, the candidate parameter minimum distance ratio and the preset candidate parameter maximum distance ratio.

Specifically, the candidate parameter distance ratio is determined by the following formula:

wherein h is _ij，d (k) Representing a candidate parameter distance ratio, in particular the distance ratio between the jth initial candidate parameter and the ith initial candidate parameter of the (d) th PI parameter, k _max Representing the maximum number of iterations.

And determining interaction force among candidate parameters according to the candidate parameter distance ratio, wherein the interaction force among candidate parameters is the interaction force between any initial candidate parameter and the interaction candidate parameter.

Specifically, the interaction force between candidate parameters is determined by the following formula:

F _ij，d (k)＝-η(k)[2(h _ij，d (k)) ¹³ -(h _ij，d (k)) ⁷ ]；

wherein F is _ij，d (k) The interaction force between candidate parameters, in particular the interaction force of the j-th initial candidate parameter of the d-th PI parameter to the i-th initial candidate parameter, alpha represents depth weight and is determined in advance according to an actual application scene.

The input variables η and h directly affect the value of F. In practice, the repulsive and attractive forces can be controlled by adjusting the upper and lower limits of h.

And determining the total interaction force of any initial candidate parameter according to the interaction force among the candidate parameters and the plurality of initial candidate parameters.

Specifically, the total interaction force for any of the initial candidate parameters is determined by the following formula:

wherein F is _id (k) The total interaction force of the ith initial candidate parameter representing the d PI parameter.

And obtaining the fixed key length from any initial candidate parameter to the current optimal candidate parameter, and determining the geometric constraint of any initial candidate parameter according to the any initial candidate parameter, the current optimal candidate parameter and the fixed key length from any initial candidate parameter to the current optimal candidate parameter.

In particular, geometric constraints are introduced to preserve the structure of polyatomic molecules. To simulate this constraint simply, it is assumed that each atom is covalently bound to the best atom. Thus, the geometric constraint θ of the ith initial candidate parameter for the ith PI parameter _id (k) Is determined by the following formula:

wherein b _{i，best_d} The fixed key length of the i-th initial candidate parameter to the current optimal candidate parameter, representing the d-th PI parameter.

And determining the constraint force of any initial candidate parameter according to the geometric constraint of any initial candidate parameter.

In some embodiments of the present application, the determining the constraint force of the any initial candidate parameter according to the geometric constraint of the any initial candidate parameter includes:

and determining the Lagrangian multiplier according to the preset multiplier weight.

And determining the constraint force of any initial candidate parameter according to the geometric constraint of any initial candidate parameter and the Lagrange multiplier.

Specifically, the constraint force of any of the initial candidate parameters is determined by the following formula:

where β 'represents the multiplier weight, λ' (k) represents the Lagrangian multiplier, G _id (k) Representing the constraint of the ith initial candidate parameter for the ith PI parameter at the kth iteration, The representation gradient operator (full differentiation in all directions in space) is a differential operator in the calculus, called Hamilton operator, used to represent gradients and divergences.

And determining the quality of any initial candidate parameter according to the candidate parameter objective function value, the current optimal candidate parameter, the current worst candidate parameter, any PI parameter and the interaction candidate parameter.

Specifically, the quality of any of the initial candidate parameters is determined by the following formula:

wherein m is _id (k) Representing the quality of the ith initial candidate parameter for the ith PI parameter at the kth iteration.

And updating the acceleration of any initial candidate parameter according to the total interaction force of any initial candidate parameter, the constraint force of any initial candidate parameter and the mass of any initial candidate parameter.

Specifically, in the IAS0 algorithm, each atom moves to a new position under the combined action of the interaction force and the geometric constraint. According to Newton's second law, the acceleration of any one of the initial candidate parameters is updated by the following formula:

wherein a is _id (k) Representing the acceleration of the ith initial candidate parameter for the ith PI parameter at the kth iteration.

The initial speed of any initial candidate parameter is obtained, and the initial speed of any initial candidate parameter is updated according to the initial speed of any initial candidate parameter, the acceleration of any initial candidate parameter and a preset random array.

In some embodiments of the present application, the random array includes a plurality of second random numbers, the number of the second random numbers being the same as the number of the initial candidate parameters of the arbitrary PI parameter, wherein the second random numbers are uniformly distributed in the range of [0,1 ].

Specifically, the initial speed of any of the initial candidate parameters is updated by the following formula:

v _id (k+1)＝c·v _id (k)+a _id (k)；

wherein v is _id (k) Representing the speed of the ith initial candidate parameter for the ith PI parameter at the kth iteration, c representing the second random number.

And updating any initial candidate parameter according to the updated speed of the any initial candidate parameter.

Specifically, any of the initial candidate parameters is updated by the following formula:

x _id (k+1)＝x _id (k)+v _id (k+1)；

and updating the PI parameter according to any updated initial candidate parameter.

In each iteration process, the current optimal candidate parameter is the PI parameter in the current iteration.

Step S105, updating the generator reactive power error and the generator rotor angular speed error.

And step S106, updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular speed error of the generator rotor and the updated PI parameter.

In some embodiments of the present application, the updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular velocity error of the generator rotor, and the updated PI parameter includes:

and determining a generator rotor d-axis current reference value and a generator rotor q-axis current reference value according to the updated generator reactive power error and the updated generator rotor angular speed error.

And updating the generator rotor d-axis voltage and the generator rotor q-axis voltage according to the generator rotor d-axis current reference value, the generator rotor q-axis current reference value, a preset generator rotor d-axis voltage compensation term and a preset generator rotor q-axis voltage compensation term.

Specifically, a rotor-side converter (RSC) implements MPPT using conventional PI-based VC, and an external control loop obtains dq-axis rotor current reference values by independently adjusting generator rotor angular speed and generator reactive power, respectively And->While the inner control loop adjusts the two currents and adds the compensation term u _qr2 And u _dr2 Finally obtaining the generator rotor voltage u output by the PI controller _dr And u _dr . The above-described framework constitutes four coupled PI control loops.

And respectively inputting the angular speed error of the generator rotor and the reactive power error of the generator of the doubly-fed induction motor into a double-loop PI controller of the optimal parameters obtained in the iteration, thereby obtaining the rotor voltage output by the PI controller.

And step S107, updating the doubly-fed induction motor system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor.

Further, the updating the doubly-fed induction machine system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor includes:

and updating the objective function according to the updated generator rotor d-axis voltage and the updated generator rotor q-axis voltage.

And updating the doubly-fed induction motor system model according to the updated objective function.

In some embodiments of the present application, the updating the doubly-fed induction machine system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor includes:

And feeding back the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor to the doubly-fed induction motor system model for updating after pulse width modulation.

Specifically, the rotor voltage output by the PI controller is subjected to pulse width modulation (pulse width modulation, PWM) and then fed back to the doubly-fed induction motor for the next iteration.

And S108, carrying out iterative updating on the doubly-fed induction motor system model, and determining the optimal PI parameters.

And carrying out the next iteration on each PI parameter by adopting an improved IASO algorithm based on the objective function obtained after the current iteration.

And continuously cycling the steps until convergence to obtain an optimal PI controller parameter set.

The embodiment can realize maximum power tracking of wind energy by only measuring the angular speed of the generator rotor and the reactive power of the generator, and has simple control structure. The improved IASO algorithm provided by the application is quick and stable in optimizing, can extract wind energy to the greatest extent, and greatly improves fault ride-through (FRT) capability.

According to the PI parameter optimization method for the doubly-fed induction motor, operation parameters of the doubly-fed induction motor are obtained first, a doubly-fed induction motor system model is built, and in the doubly-fed induction motor system model, a double-loop PI control frame is adopted for the angular speed of a generator rotor and the reactive power of the generator. And then updating the PI parameters by using an IASO algorithm according to the doubly-fed induction motor system model. And finally, feeding back the generator rotor voltage output by the PI controller to a doubly-fed induction motor system model for the next iteration until the algorithm converges to obtain the optimal PI parameter. The method and the device can accurately optimize all PI parameters through the steps, have short optimizing time and high optimizing precision, and effectively solve the technical problems in the prior art.

The following are device embodiments of the present application, which may be used to perform method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

The second embodiment of the application discloses a PI parameter optimization apparatus of a doubly-fed induction motor, where the PI parameter optimization apparatus of a doubly-fed induction motor is applied to the PI parameter optimization method of a doubly-fed induction motor disclosed in the first embodiment of the application, see a schematic structural diagram shown in fig. 2, and the PI parameter optimization apparatus of a doubly-fed induction motor includes:

an operation parameter obtaining module 10, configured to obtain operation parameters of a doubly-fed induction machine, where the operation parameters of the doubly-fed induction machine include: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage, and generator rotor q-axis voltage.

And the objective function determining module 20 is configured to determine an objective function of the PI parameter according to the operation parameter of the doubly-fed induction machine and a preset limiting condition.

A doubly-fed induction machine system model generation module 30 for generating a doubly-fed induction machine system model based on the objective function, a pre-built wind turbine model, a pre-built generator model, and a pre-built rotor shaft system model.

Further, the doubly-fed induction machine system model generation module 30 includes:

the blade tip speed ratio acquisition unit is used for acquiring the mechanical rotation speed of the wind turbine, the radius of the wind turbine and the wind speed, and determining the blade tip speed ratio according to the mechanical rotation speed of the wind turbine, the radius of the wind turbine and the wind speed.

The wind energy utilization coefficient acquisition unit is used for acquiring the pitch angle and determining the wind energy utilization coefficient according to the tip speed ratio and the pitch angle.

A wind turbine model generation unit for generating the wind turbine model based on the air density, the wind turbine radius, the wind energy utilization coefficient and the wind speed.

Further, the doubly-fed induction machine system model generation module 30 includes:

the generator dynamic equation determining unit is used for obtaining the mechanical rotation speed of the generator, the synchronous angular speed of the generator, the angular speed of the generator rotor, the internal voltage of the generator d axis, the internal voltage of the generator q axis, the current of the generator d axis stator, the current of the generator q axis stator, the voltage of the generator d axis stator end, the voltage of the generator q axis stator end, the voltage of the generator d axis rotor end, the voltage of the generator q axis rotor end, mutual inductance between the generator stator and the generator rotor, the inductance of the generator stator, the inductance of the generator rotor, the resistance of the generator stator and the resistance of the generator rotor, and determining the generator dynamic equation.

And the electromagnetic torque determining unit is used for determining electromagnetic torque generated by the generator according to the generator d-axis internal voltage, the generator q-axis internal voltage, the generator synchronous angular speed, the generator d-axis stator current and the generator q-axis stator current.

And the generator reactive power determining unit is used for determining the generator reactive power according to the generator d-axis stator terminal voltage, the generator q-axis stator terminal voltage, the generator d-axis stator current and the generator q-axis stator current.

And the generator model generating unit is used for generating the generator model according to the generator dynamic equation, the electromagnetic torque generated by the generator and the generator reactive power.

Further, the doubly-fed induction machine system model generation module 30 includes:

and the equivalent inertia constant acquisition unit is used for determining the equivalent inertia constant of the rotating shaft system according to the preset wind turbine inertia constant and the preset generator inertia constant.

A mechanical torque acquisition unit for determining a mechanical torque from the wind turbine model and the generator rotor angular speed.

And the rotating shaft system model generating unit is used for generating the rotating shaft system model according to the angular speed of the generator rotor, the preset integrated damping of the concentrated inertia system, the mechanical torque and the electromagnetic torque generated by the generator.

And the PI parameter updating module 40 is configured to update the PI parameter according to the doubly-fed induction machine system model by using an IAS0 algorithm.

Further, the PI parameter updating module 40 includes:

the initial candidate parameter acquisition unit is used for acquiring a plurality of initial candidate parameters of any PI parameter according to a preset first random number aiming at the any PI parameter.

And the objective function value calculation unit is used for determining candidate parameter objective function values according to the doubly-fed induction motor system model and a plurality of initial candidate parameters.

And the candidate parameter screening unit is used for determining the current optimal candidate parameter and the current worst candidate parameter according to the candidate parameter objective function value.

An interaction candidate parameter obtaining unit, configured to obtain, for any initial candidate parameter of the any PI parameter, an interaction candidate parameter from a plurality of initial candidate parameters of the any PI parameter, where the interaction candidate parameter is any initial candidate parameter that has an interaction with the any initial candidate parameter.

And the Euclidean distance determining unit is used for determining the Euclidean distance of the candidate parameter according to any initial candidate parameter and the interaction candidate parameter.

And the minimum distance ratio acquisition unit is used for determining a candidate parameter minimum distance ratio according to the candidate parameter objective function value and the current optimal candidate parameter.

And the candidate parameter distance ratio determining unit is used for determining a candidate parameter distance ratio according to the Euclidean distance of the candidate parameter, the objective function value of the candidate parameter, the plurality of initial candidate parameters, the current optimal candidate parameter, the minimum distance ratio of the candidate parameter and the preset maximum distance ratio of the candidate parameter.

And the interaction force acquisition unit is used for determining interaction force among candidate parameters according to the candidate parameter distance ratio, wherein the interaction force among candidate parameters is the interaction force between any initial candidate parameter and the interaction candidate parameter.

And the total interaction force acquisition unit is used for determining the total interaction force of any initial candidate parameter according to the interaction force among the candidate parameters and the plurality of initial candidate parameters.

The geometric constraint obtaining unit is used for obtaining the fixed key length from any initial candidate parameter to the current optimal candidate parameter, and determining the geometric constraint of any initial candidate parameter according to the any initial candidate parameter, the current optimal candidate parameter and the fixed key length from any initial candidate parameter to the current optimal candidate parameter.

And the constraint force determining unit is used for determining the constraint force of any initial candidate parameter according to the geometric constraint of any initial candidate parameter.

And the quality acquisition unit is used for determining the quality of any initial candidate parameter according to the candidate parameter objective function value, the current optimal candidate parameter, the current worst candidate parameter, any PI parameter and the interaction candidate parameter.

And the acceleration updating unit is used for updating the acceleration of any initial candidate parameter according to the total interaction force of any initial candidate parameter, the constraint force of any initial candidate parameter and the mass of any initial candidate parameter.

The initial speed updating unit is used for acquiring the initial speed of any initial candidate parameter and updating the initial speed of any initial candidate parameter according to the initial speed of any initial candidate parameter, the acceleration of any initial candidate parameter and a preset random array.

And the initial candidate parameter updating unit is used for updating any initial candidate parameter according to the updated speed of any initial candidate parameter.

And the PI parameter updating unit is used for updating the PI parameter according to any one of the updated initial candidate parameters. An error update module 50 for updating the generator reactive power error and the generator rotor angular speed error.

The generator rotor voltage update module 60 is configured to update the generator rotor d-axis voltage and the generator rotor q-axis voltage according to the updated generator reactive power error, the updated generator rotor angular speed error, and the updated PI parameter.

Further, the generator rotor voltage update module 60 includes:

and the rotor current reference value determining unit is used for determining a generator rotor d-axis current reference value and a generator rotor q-axis current reference value according to the updated generator reactive power error and the updated generator rotor angular speed error.

And the rotor voltage updating unit is used for updating the generator rotor d-axis voltage and the generator rotor q-axis voltage according to the generator rotor d-axis current reference value, the generator rotor q-axis current reference value, a preset generator rotor d-axis voltage compensation term and a preset generator rotor q-axis voltage compensation term.

The system model updating module 70 is configured to update the doubly-fed induction machine system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor.

Further, the system model updating module 70 includes:

and the objective function updating unit is used for updating the objective function according to the updated generator rotor d-axis voltage and the updated generator rotor q-axis voltage.

And the system model updating unit is used for updating the doubly-fed induction motor system model according to the updated objective function.

Further, the system model updating module 70 includes:

and the pulse width modulation unit is used for feeding the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor back to the doubly-fed induction motor system model for updating after pulse width modulation.

And the PI parameter optimization module 80 is used for iteratively updating the doubly-fed induction motor system model to determine an optimal PI parameter.

The foregoing detailed description has been provided for the purposes of illustration in connection with specific embodiments and exemplary examples, but such description is not to be construed as limiting the application. Those skilled in the art will appreciate that various equivalent substitutions, modifications and improvements may be made to the technical solution of the present application and its embodiments without departing from the spirit and scope of the present application, and these all fall within the scope of the present application. The scope of the application is defined by the appended claims.

Claims (9)

1. The PI parameter optimization method of the doubly-fed induction motor is characterized by comprising the following steps of:

obtaining doubly-fed induction machine operating parameters, the doubly-fed induction machine operating parameters comprising: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage and generator rotor q-axis voltage;

determining an objective function of the PI parameter according to the operation parameter of the doubly-fed induction motor and a preset limiting condition;

generating a doubly-fed induction machine system model according to the objective function, a pre-built wind turbine model, a pre-built generator model and a pre-built rotating shaft system model;

updating the PI parameters by using an IASO algorithm according to the doubly-fed induction motor system model;

updating the generator reactive power error and the generator rotor angular speed error;

updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular speed error of the generator rotor and the updated PI parameter;

updating the doubly-fed induction motor system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor;

Performing iterative updating on the doubly-fed induction motor system model to determine an optimal PI parameter;

the updating of the PI parameters using an IASO algorithm according to the doubly-fed induction machine system model includes:

for any PI parameter, acquiring a plurality of initial candidate parameters of the any PI parameter according to a preset first random number;

determining candidate parameter objective function values according to the doubly-fed induction motor system model and a plurality of initial candidate parameters;

determining a current optimal candidate parameter and a current worst candidate parameter according to the candidate parameter objective function value;

for any initial candidate parameter of any PI parameter, acquiring an interaction candidate parameter from a plurality of initial candidate parameters of the any PI parameter, wherein the interaction candidate parameter refers to any initial candidate parameter interacted with the any initial candidate parameter;

determining a Euclidean distance of a candidate parameter according to the any initial candidate parameter and the interaction candidate parameter;

determining a minimum distance ratio of candidate parameters according to the candidate parameter objective function value and the current optimal candidate parameter;

determining a candidate parameter distance ratio according to the candidate parameter Euclidean distance, the candidate parameter objective function value, the plurality of initial candidate parameters, the current optimal candidate parameter, the candidate parameter minimum distance ratio and a preset candidate parameter maximum distance ratio;

Determining interaction force among candidate parameters according to the candidate parameter distance ratio, wherein the interaction force among candidate parameters is the interaction force between any initial candidate parameter and the interaction candidate parameter;

determining the total interaction force of any initial candidate parameter according to the interaction force among the candidate parameters and the plurality of initial candidate parameters;

obtaining a fixed key length from any initial candidate parameter to the current optimal candidate parameter, and determining geometric constraint of any initial candidate parameter according to the any initial candidate parameter, the current optimal candidate parameter and the fixed key length from any initial candidate parameter to the current optimal candidate parameter;

determining constraint force of any initial candidate parameter according to geometric constraint of the any initial candidate parameter;

determining the quality of any initial candidate parameter according to the candidate parameter objective function value, the current optimal candidate parameter, the current worst candidate parameter, any PI parameter and the interaction candidate parameter;

updating the acceleration of any initial candidate parameter according to the total interaction force of any initial candidate parameter, the constraint force of any initial candidate parameter and the mass of any initial candidate parameter;

Acquiring the initial speed of any initial candidate parameter, and updating the initial speed of any initial candidate parameter according to the initial speed of any initial candidate parameter, the acceleration of any initial candidate parameter and a preset random array;

updating any initial candidate parameter according to the updated speed of the any initial candidate parameter;

and updating the PI parameter according to any updated initial candidate parameter.

2. The PI parameter optimization method for a doubly-fed induction machine according to claim 1, wherein said updating said generator rotor d-axis voltage and said generator rotor q-axis voltage based on said updated generator reactive power error, said updated generator rotor angular speed error, and said updated PI parameters comprises:

determining a generator rotor d-axis current reference value and a generator rotor q-axis current reference value according to the updated generator reactive power error and the updated generator rotor angular speed error;

and updating the generator rotor d-axis voltage and the generator rotor q-axis voltage according to the generator rotor d-axis current reference value, the generator rotor q-axis current reference value, a preset generator rotor d-axis voltage compensation term and a preset generator rotor q-axis voltage compensation term.

3. The PI parameter optimization method for a doubly-fed induction machine according to claim 1, wherein said updating said doubly-fed induction machine system model based on said updated generator rotor d-axis voltage and said updated generator rotor q-axis voltage comprises:

updating the objective function according to the updated generator rotor d-axis voltage and the updated generator rotor q-axis voltage;

and updating the doubly-fed induction motor system model according to the updated objective function.

4. The PI parameter optimization method for a doubly-fed induction machine according to claim 1, wherein said updating said doubly-fed induction machine system model based on said updated generator rotor d-axis voltage and said updated generator rotor q-axis voltage comprises:

and feeding back the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor to the doubly-fed induction motor system model for updating after pulse width modulation.

5. The PI parameter optimization method for a doubly-fed induction machine according to claim 1, wherein said constraints include: an outer loop proportional gain limit, an outer loop integral gain limit, an inner loop proportional gain limit, an inner loop integral gain limit, a wind speed limit, a grid voltage limit, a reactive power limit, a generator rotor d-axis voltage limit, and a generator rotor q-axis voltage limit.

6. The PI parameter optimization method for a doubly-fed induction machine according to claim 1, wherein said step of constructing said wind turbine model comprises:

acquiring the mechanical rotation speed of a wind turbine, the radius of the wind turbine and the wind speed, and determining the tip speed ratio according to the mechanical rotation speed of the wind turbine, the radius of the wind turbine and the wind speed;

obtaining a pitch angle, and determining a wind energy utilization coefficient according to the tip speed ratio and the pitch angle;

generating the wind turbine model based on the air density, the wind turbine radius, the wind energy utilization coefficient, and the wind speed.

7. The PI parameter optimization method for a doubly-fed induction machine according to claim 1, wherein the step of constructing said generator model comprises:

acquiring a mechanical rotation speed of a generator, a synchronous angular speed of the generator, an angular speed of a rotor of the generator, an internal voltage of a d-axis of the generator, an internal voltage of a q-axis of the generator, a stator current of a d-axis of the generator, a stator terminal voltage of a q-axis of the generator, a rotor terminal voltage of a d-axis of the generator, a rotor terminal voltage of a q-axis of the generator, mutual inductance between stator and rotor of the generator, stator inductance of the generator, rotor inductance of the generator, stator resistance of the generator and rotor resistance of the generator, and determining a dynamic equation of the generator;

Determining electromagnetic torque generated by a generator according to the generator d-axis internal voltage, the generator q-axis internal voltage, the generator synchronous angular speed, the generator d-axis stator current and the generator q-axis stator current;

determining reactive power of the generator according to the generator d-axis stator terminal voltage, the generator q-axis stator terminal voltage, the generator d-axis stator current and the generator q-axis stator current;

and generating the generator model according to the generator dynamic equation, the electromagnetic torque generated by the generator and the reactive power of the generator.

8. The PI parameter optimization method for a doubly-fed induction machine according to claim 7, wherein said step of constructing said rotating shaft system model comprises:

determining an equivalent inertia constant of the rotating shaft system according to a preset wind turbine inertia constant and a preset generator inertia constant;

determining a mechanical torque from the wind turbine model and the generator rotor angular speed;

and generating the rotating shaft system model according to the angular speed of the generator rotor, the preset integrated damping of the concentrated inertia system, the mechanical torque and the electromagnetic torque generated by the generator.

9. PI parameter optimizing apparatus for a doubly-fed induction machine, wherein the PI parameter optimizing apparatus for a doubly-fed induction machine is applied to the PI parameter optimizing method for a doubly-fed induction machine according to any one of claims 1 to 8, the PI parameter optimizing apparatus for a doubly-fed induction machine comprising:

the operation parameter acquisition module is used for acquiring operation parameters of the doubly-fed induction motor, and the operation parameters of the doubly-fed induction motor comprise: generator reactive power error, generator rotor angular speed error, generator rotor d-axis voltage and generator rotor q-axis voltage;

the objective function determining module is used for determining an objective function of the PI parameter according to the operation parameter of the doubly-fed induction motor and a preset limiting condition;

the doubly-fed induction motor system model generation module is used for generating a doubly-fed induction motor system model according to the objective function, a pre-built wind turbine model, a pre-built generator model and a pre-built rotating shaft system model;

the PI parameter updating module is used for updating the PI parameter by using an IASO algorithm according to the doubly-fed induction motor system model;

the error updating module is used for updating the reactive power error of the generator and the angular speed error of the rotor of the generator;

The generator rotor voltage updating module is used for updating the d-axis voltage of the generator rotor and the q-axis voltage of the generator rotor according to the updated reactive power error of the generator, the updated angular speed error of the generator rotor and the updated PI parameter;

the system model updating module is used for updating the doubly-fed induction motor system model according to the updated d-axis voltage of the generator rotor and the updated q-axis voltage of the generator rotor;

and the PI parameter optimization module is used for carrying out iterative updating on the doubly-fed induction motor system model to determine the optimal PI parameter.