1. A robust disturbance rejection composite control method for overcoming effects of periodically moving parts, comprising:

step one, setting an angular velocity measurement period as h and recording a time sequence as t₁,t₂,…,t_k…; the current angular velocity measurement is ω (k)；

Step two, taking the satellite angular velocity measurement sequences omega (k-N +1), … omega (k-1) and omega (k) as input, and outputting the first harmonic frequency f of the periodic motion interference by adopting a fast Fourier transform method_dAccording to said f_dCalculating the motion disturbance period Deltat_d；

Thirdly, calculating the angular acceleration estimated value z at the current moment according to the angular velocity measured value omega (k-1)₂(k) (ii) a Calculating the amplitude T of the periodic disturbance torque_d(k) (ii) a And obtaining the maximum amplitude T of the periodic disturbance torque_d,max；

Step four, the amplitude T of the periodic disturbance torque_d(k) From 0 to 0.1 × T_d,maxIs denoted as t_sPeriod of same motion disturbance Δ t_dAmplitude T of internal and periodic disturbance torque_d(k) From T_d,maxDown to 0.1 × T_d,maxIs denoted as t_f(ii) a Calculating each dynamic interference period delta t_dInner feedforward compensation duration Δ t_c(ii) a Number of sampling points N for calculating interference period_dAnd the number of sampling points N in the feedforward compensation duration_c；

Step five, according to the sampling point number N of the interference period_dAnd the number of sampling points N in the feedforward compensation duration_cCalculating feedforward control quantity T_c；

Step six, obtaining the maximum amplitude omega of the angular speed error by utilizing an extreme method according to the angular speed measurement sequences omega (k-N +1), … omega (k-1) and omega (k)_max；

Step seven, according to the maximum amplitude T of the disturbance moment estimated value_d,maxMaximum amplitude ω of the sum angular velocity error_maxCalculating the feedforward compensation time value delta t of the disturbance moment_lead(ii) a And calculates a delay value deltat relative to the start time of the current calculation cycle_lag；

Step eight, calculating the delay value delta t relative to the starting moment of the current calculation period_lagFeedforward compensation duration Δ t_cFeedforward control amount T_cThree parameters constitute the feedforward control law, namely delta t after the beginning of the current calculation cycle_lag～Δt_lag+Δt_cTime-of-flight, output feed-forwardControl quantity T_cAnd the feedforward compensation mark S_c1, placing; at Δ t_lag～Δt_lag+Δt_cAt a time outside the time interval, the feedforward control quantity T is_cSet 0 and set the feedforward compensation flag S_cSetting 0;

step nine, according to the feedforward compensation mark S_cThe value of (3) is adjusted to the feedback control law;

and step ten, adding the feedback control quantity output by the feedback control law and the feedforward control quantity to obtain a final composite control quantity, and controlling the satellite by using the composite control quantity.

2. The method according to claim 1, wherein in the second step, the fast fourier transform method is a radix-2 type fast fourier transform algorithm decimated in time; according to f_dCalculating the motion disturbance period Deltat_dThe method comprises the following steps:

Δt_d＝1/f_d。

3. the method as claimed in claim 1, wherein in step three, the angular acceleration at the current time z is estimated₂(k) The calculation method comprises the following steps:

in the formula, e (×) is observer error;

z₁() is the observer first order state quantity;

h is an angular velocity measurement period;

ω (k-1) is the angular velocity measurement;

β₁a first preset gain coefficient;

u (k) is a closed-loop control quantity;

j is the uniaxial moment of inertia of the star;

β₂a second preset gain factor;

fal (x, α, δ) is a non-linear function;

the nonlinear function fal (x, α, δ) is calculated by the formula:

in the formula, both α and δ are predetermined coefficients.

4. The method of claim 3, wherein: in the third step, the amplitude T of the periodic disturbance torque_d(k) The calculation method comprises the following steps:

T_d(k)＝u(k)-Jz₂(k)

and obtaining the maximum amplitude T of the periodic disturbance torque by an extreme value judgment method_d,max。

5. The method of claim 4, wherein: in the fourth step, the feedforward compensation duration Δ t_cThe calculation method comprises the following steps:

Δt_c＝t_f-t_s。

6. the method of claim 5, wherein: in the fourth step, the number of sampling points in the interference period is N_dThe calculation method comprises the following steps:

wherein Z (x) is rounded;

number of sampling points N within feedforward compensation duration_cThe calculation method comprises the following steps:

7. the method of claim 6, wherein: in the fifth step, the process is carried out,feedforward control quantity T_cThe calculation method comprises the following steps:

8. the method of claim 7, wherein: in the seventh step, the disturbance moment feedforward compensation time value delta t_leadThe calculation method comprises the following steps: establishing a fuzzy logic system; the input variable of the fuzzy logic system is the maximum amplitude T of the periodic disturbance torque_d,maxMaximum amplitude ω of the sum angular velocity error_maxThe output variable is the feedforward compensation time value delta t of the disturbance moment_lead(ii) a The rules of fuzzy logic are shown in table 1:

TABLE 1

In the table, S represents Small; m represents in; b represents large; s1 indicates smaller; s2 indicates very small; b1 indicates larger; b2 denotes very large;

maximum amplitude T of periodic disturbance torque_d,maxHas a discourse field of [0,0.15 ]]；

Maximum amplitude ω of angular velocity error_maxHas a discourse field of [0,0.01 ]]；

Disturbance moment feedforward compensation time value delta t_leadHas a discourse field of [0,4.3 ]]。

9. The method of claim 8, wherein: in the seventh step, the delay value delta t relative to the starting time of the current calculation period_lagThe calculation method comprises the following steps:

Δt_lag＝Δt_d-Δt_lead

in the formula,. DELTA.t_dIs a motion disturbance period.

10. The method of claim 9, wherein: in the ninth step, the feedback control law adopts a PID control rate, and the specific method for adjusting the PID control law is as follows:

when S is_cWhen the value is equal to 0, namely during the non-feedforward compensation period, adopting a PID control law of normal gain; when S is_cWhen the value is 1, namely during the feedforward torque compensation, a PID control law with higher gain is adopted, namely, the proportional coefficient and the differential coefficient of the PID control law are appropriately increased on the basis of a normal value.

Background

Large long-life satellites usually carry a large number of different types of loads, the dynamic characteristics of the satellites are complex, and the control requirements are high. For imaging loads, the attitude stability of the satellite platform is an important technical index. Under the influence of various interference factors such as flexible vibration, liquid shaking, component movement and the like, the improvement of the star attitude stability is difficult. Among the numerous sources of interference on the planet, low frequency motion disturbances of periodically moving parts are common and have significant effects. Common periodic moving parts on a satellite are a scanning camera, a scanning antenna, a solar panel driving mechanism (SADA), and the like. Especially the SADA, which is an important component required for almost all three-axis stable satellites. The SADA is typically driven by a stepper motor, with a step disturbance torque on stepping being significant. In the high-stability remote sensing satellite, a subdivision driving mode is mostly adopted to reduce single-step motion interference of the SADA. This requires special design in hardware, and the system cost is relatively high. Therefore, for the periodic motion disturbance, it is necessary to improve the control method to enhance the capability of resisting the motion disturbance as much as possible.

For unknown periodic interference, the existing control methods are mainly classified into 3 types: 1) when the interference cannot be measured, but the period is known, repeated control based on an internal model principle is generally adopted, which is feedback control, but when large unmodeled dynamics exist, the stability of the system is not easy to guarantee; 2) when the disturbance can be directly measured or is known in advance (the conditions are usually difficult to satisfy in engineering), a control method mainly based on feedforward is generally adopted, and the control method comprises an inverse dynamics method, a finite impulse response method and the like; 3) for satellite attitude control with unknown (cycle unknown or amplitude unknown) deterministic interference, the method of the existing document is mainly to establish an interference model by adopting a system identification method, and the uncertainty of model parameters is ensured by improving the robustness of a feedback control law.

Disclosure of Invention

The technical problem solved by the invention is as follows: the method overcomes the defects of the prior art, provides a strong anti-interference composite control method for overcoming the influence of periodic moving parts, is used for satellite attitude control with periodic interference, and can effectively improve the attitude control precision and attitude stability of a system. The whole method has strong systematicness, clear process and easy realization.

The technical scheme of the invention is as follows:

a strong anti-interference composite control method for overcoming the influence of a periodic moving part comprises the following steps:

step one, setting an angular velocity measurement period as h and recording a time sequence as t₁,t₂,…,t_k…; the current angular velocity measurement is ω (k);

step two, taking the satellite angular velocity measurement sequences omega (k-N +1), … omega (k-1) and omega (k) as input, and outputting the first harmonic frequency f of the periodic motion interference by adopting a fast Fourier transform method_dAnd calculating the motion disturbance period Deltat_d；

Step three, calculating the angular acceleration estimated value z at the current moment₂(k) (ii) a Calculating the amplitude T of the periodic disturbance torque_d(k) (ii) a And obtaining the maximum amplitude T of the periodic disturbance torque_d,max；

Step four, the amplitude T of the periodic disturbance torque_d(k) From 0 to 0.1 × T_d,maxIs denoted as t_sPeriod of same motion disturbance Δ t_dInternal, periodic disturbance forceMagnitude T of moment_d(k) From T_d,maxDown to 0.1 × T_d,maxIs denoted as t_f(ii) a Calculating each dynamic interference period delta t_dInner feedforward compensation duration Δ t_c(ii) a Number of sampling points N for calculating interference period_dAnd the number of sampling points N in the feedforward compensation duration_c；

Step five, according to the sampling point number N of the interference period_dAnd the number of sampling points N in the feedforward compensation duration_cCalculating feedforward control quantity T_c；

Step six, obtaining the maximum amplitude omega of the angular speed error by utilizing an extreme method according to the angular speed measurement sequences omega (k-N +1), … omega (k-1) and omega (k)_max；

Step seven, according to the maximum amplitude T of the disturbance moment estimated value_d,maxMaximum amplitude ω of the sum angular velocity error_maxCalculating the feedforward compensation time value delta t of the disturbance moment_lead(ii) a And calculates a delay value deltat relative to the start time of the current calculation cycle_lag；

Step eight, calculating the delay value delta t relative to the starting moment of the current calculation period_lagFeedforward compensation duration Δ t_cFeedforward control amount T_cThree parameters constitute the feedforward control rate, i.e. at after the start of the current calculation cycle_lag～Δt_lag+Δt_cIn a time period, a feedforward control quantity T is output_cAnd the feedforward compensation mark S_c1, placing; at Δ t_lag～Δt_lag+Δt_cAt a time outside the time interval, the feedforward control quantity T is_cSet 0 and set the feedforward compensation flag S_cSetting 0;

step nine, according to the feedforward compensation mark S_cThe value of (3) is adjusted to the feedback control law;

and step ten, adding the control quantity output by the feedback control law and the control quantity output by the feedforward control rate to obtain a final composite control quantity, and controlling the satellite by using the composite control quantity.

In the above strong anti-interference composite control method for overcoming the influence of the periodically moving parts, in the second step, the fast Fourier transform method is time-wise decimatedA radix-2 type fast fourier transform algorithm; period of motion disturbance Δ t_dThe calculation method comprises the following steps:

Δt_d＝1/f_d。

in the above-mentioned strong disturbance rejection composite control method for overcoming the influence of periodically moving parts, in the third step, the angular acceleration estimated value z at the current moment₂(k) The calculation method comprises the following steps:

in the formula, e (×) is observer error;

z₁() is the observer first order state quantity;

h is an angular velocity measurement period;

ω (k-1) is the angular velocity measurement;

β₁a first preset gain coefficient;

u (k) is a closed-loop control quantity;

j is the uniaxial moment of inertia of the star;

β₂a second preset gain factor;

fal (x, α, δ) is a non-linear function;

the nonlinear function fal (x, α, δ) is calculated by the formula:

in the formula, both α and δ are predetermined coefficients.

In the above-mentioned strong disturbance rejection composite control method for overcoming the influence of the periodically moving component, in the third step, the amplitude T of the periodically disturbing moment_d(k) The calculation method comprises the following steps:

T_d(k)＝u(k)-Jz₂(k)

and obtaining the maximum amplitude T of the periodic disturbance torque by an extreme value judgment method_d,max。

In one of the above-mentioned parts against cyclic movementsAnd in the fourth step, the feedforward compensation time length delta t_cThe calculation method comprises the following steps:

Δt_c＝t_f-t_s。

in the above-mentioned strong anti-interference composite control method for overcoming the influence of the periodically moving part, in the fourth step, the number of sampling points N of the interference period_dThe calculation method comprises the following steps:

wherein Z (x) is rounded;

number of sampling points N within feedforward compensation duration_cThe calculation method comprises the following steps:

in the above strong disturbance rejection composite control method for overcoming the influence of the periodically moving part, in the fifth step, the feedforward control quantity T_cThe calculation method comprises the following steps:

in the above strong disturbance rejection composite control method for overcoming the influence of the periodically moving part, in the seventh step, the disturbance moment feedforward compensation time value delta t_leadThe calculation method comprises the following steps: establishing a fuzzy logic system; the input variable of the fuzzy logic system is the maximum amplitude T of the periodic disturbance torque_d,maxMaximum amplitude ω of the sum angular velocity error_maxThe output variable is the feedforward compensation time value delta t of the disturbance moment_lead(ii) a The rules of fuzzy logic are shown in table 1:

TABLE 1

In the table, S represents Small; m represents in; b represents large; s1 indicates smaller; s2 indicates very small; b1 indicates larger; b2 denotes very large;

maximum amplitude T of periodic disturbance torque_d,maxHas a discourse field of [0,0.15 ]]；

Maximum amplitude ω of angular velocity error_maxHas a discourse field of [0,0.01 ]]；

Disturbance moment feedforward compensation time value delta t_leadHas a discourse field of [0,4.3 ]]。

In the above-mentioned strong disturbance rejection composite control method for overcoming the influence of the periodically moving part, in the seventh step, the delay value delta t relative to the starting time of the current calculation period_lagThe calculation method comprises the following steps:

Δt_lag＝Δt_d-Δt_lead

in the formula,. DELTA.t_dIs a motion disturbance period.

In the aforementioned strong disturbance rejection composite control method for overcoming the influence of the periodic moving part, in the ninth step, the feedback control law adopts a PID control rate, and the specific method for adjusting the PID control law is as follows:

when S is_cWhen the value is equal to 0, namely during the non-feedforward compensation period, adopting a PID control law of normal gain; when S is_cWhen the value is 1, namely during the feedforward torque compensation, a PID control law with higher gain is adopted, namely, the proportional coefficient and the differential coefficient of the PID control law are appropriately increased on the basis of a normal value.

Compared with the prior art, the invention has the beneficial effects that:

(1) the method does not need system identification, and does not have excessive requirements on the prior knowledge of the model;

(2) the invention realizes the accurate compensation and inhibition of the periodic motion interference of the satellite moving parts and has better anti-interference effect.

Drawings

FIG. 1 is a flow chart of the strong anti-interference composite control of the present invention;

fig. 2 is a schematic structural diagram of a robust disturbance rejection complex control system according to an embodiment of the present invention.

Detailed Description

The invention is further illustrated by the following examples.

Aiming at the influence of on-satellite periodic moving parts, a novel feedforward and feedback composite control method with strong anti-interference capability is provided; the method comprehensively utilizes Fast Fourier Transform (FFT) and an Extended State Observer (ESO) to estimate periodic interference, on the basis, an equivalent feedforward compensation law is designed according to the angular momentum conservation principle, a fuzzy logic system is utilized to adjust the feedforward compensation time value of the interference moment, and the gain of a PID controller is dynamically adjusted according to whether the interference moment is in a feedforward compensation period or not. Compared with the existing method, the method does not need system identification and excessive model prior knowledge, and realizes better anti-interference effect by a simple control law.

As shown in fig. 1, the strong disturbance rejection composite control method for overcoming the influence of a periodic moving part specifically includes the following steps:

step one, setting an angular velocity measurement period as h and recording a time sequence as t₁,t₂…, tk, …; the current angular velocity measurement is ω (k).

Step two, taking the satellite angular velocity measurement sequences omega (k-N +1), … omega (k-1) and omega (k) as input, and outputting the first harmonic frequency f of the periodic motion interference by adopting a fast Fourier transform method_dAnd calculating the motion disturbance period Deltat_d(ii) a The fast Fourier transform method is a radix-2 type fast Fourier transform algorithm which is selected according to time; period of motion disturbance Δ t_dThe calculation method comprises the following steps:

Δt_d＝1/f_d。

step three, calculating the angular acceleration estimated value z at the current moment₂(k) (ii) a Calculating the amplitude T of the periodic disturbance torque_d(k) (ii) a And obtaining the maximum amplitude T of the periodic disturbance torque_d,max(ii) a Angular acceleration estimate z at the present time₂(k) The calculation method comprises the following steps:

in the formula, e (×) is observer error;

z₁() is the observer first order state quantity;

h is an angular velocity measurement period;

ω (k-1) is the angular velocity measurement;

β₁a first preset gain coefficient;

u (k) is a closed-loop control quantity;

j is the uniaxial moment of inertia of the star;

β₂a second preset gain factor;

fal (x, α, δ) is a non-linear function;

the nonlinear function fal (x, α, δ) is calculated by the formula:

in the formula, both α and δ are predetermined coefficients.

Amplitude T of periodic disturbance torque_d(k) The calculation method comprises the following steps:

T_d(k)＝u(k)-Jz₂(k)

and obtaining the maximum amplitude T of the periodic disturbance torque by an extreme value judgment method_d,max。

Step four, the amplitude T of the periodic disturbance torque_d(k) From 0 to 0.1 × T_d,maxIs denoted as t_sPeriod of same motion disturbance Δ t_dAmplitude T of internal and periodic disturbance torque_d(k) From T_d,maxDown to 0.1 × T_d,maxIs denoted as t_f(ii) a Calculating each dynamic interference period delta t_dInner feedforward compensation duration Δ t_c(ii) a Number of sampling points N for calculating interference period_dAnd the number of sampling points N in the feedforward compensation duration_c(ii) a Feedforward compensation duration Δ t_cThe calculation method comprises the following steps:

Δt_c＝t_f-t_s。

number of sampling points N of interference period_dThe calculation method comprises the following steps:

wherein Z (x) is rounded;

number of sampling points N within feedforward compensation duration_cThe calculation method comprises the following steps:

step five, according to the sampling point number N of the interference period_dAnd the number of sampling points N in the feedforward compensation duration_cCalculating feedforward control quantity T_c(ii) a Feedforward control quantity T_cThe calculation method comprises the following steps:

step six, obtaining the maximum amplitude omega of the angular speed error by utilizing an extreme method according to the angular speed measurement sequences omega (k-N +1), … omega (k-1) and omega (k)_max。

Step seven, according to the maximum amplitude T of the disturbance moment estimated value_d,maxMaximum amplitude ω of the sum angular velocity error_maxCalculating the feedforward compensation time value delta t of the disturbance moment_lead(ii) a And calculates a delay value deltat relative to the start time of the current calculation cycle_lag(ii) a Disturbance moment feedforward compensation time value delta t_leadThe calculation method comprises the following steps: establishing a fuzzy logic system; the input variable of the fuzzy logic system is the maximum amplitude T of the periodic disturbance torque_d,maxMaximum amplitude ω of the sum angular velocity error_maxThe output variable is the feedforward compensation time value delta t of the disturbance moment_lead(ii) a The rules of fuzzy logic are shown in table 1:

TABLE 1

In the table, S represents Small; m represents in; b represents large; s1 indicates smaller; s2 indicates very small; b1 indicates larger; b2 denotes very large;

maximum amplitude T of periodic disturbance torque_d,maxHas a discourse field of [0,0.15 ]]；

Maximum amplitude ω of angular velocity error_maxHas a discourse field of [0,0.01 ]]；

Disturbance moment feedforward compensation time value delta t_leadHas a discourse field of [0,4.3 ]]。

Delay value delta t relative to the start time of the current calculation cycle_lagThe calculation method comprises the following steps:

Δt_lag＝Δt_d-Δt_lead

in the formula,. DELTA.t_dIs a motion disturbance period.

Step eight, calculating the delay value delta t relative to the starting moment of the current calculation period_lagFeedforward compensation duration Δ t_cFeedforward control amount T_cThree parameters constitute the feedforward control rate, i.e. at after the start of the current calculation cycle_lag～Δt_lag+Δt_cIn a time period, a feedforward control quantity T is output_cAnd the feedforward compensation mark S_c1, placing; at Δ t_lag～Δt_lag+Δt_cAt a time outside the time interval, the feedforward control quantity T is_cSet 0 and set the feedforward compensation flag S_cSetting 0;

step nine, according to the feedforward compensation mark S_cThe value of (3) is adjusted to the feedback control law; the feedback control law adopts PID control rate, and the specific method for adjusting the PID control law is as follows:

when S is_cWhen the value is equal to 0, namely during the non-feedforward compensation period, adopting a PID control law of normal gain; when S is_cWhen the value is 1, namely during the feedforward torque compensation, a PID control law with higher gain is adopted, namely, the proportional coefficient and the differential coefficient of the PID control law are appropriately increased on the basis of a normal value.

And step ten, adding the control quantity output by the feedback control law and the control quantity output by the feedforward control rate to obtain a final composite control quantity, and controlling the satellite by using the composite control quantity.

Examples

The embodiments of the present invention will be described by taking a satellite with a flywheel as an execution component. The satellite pitch axis control is taken as an example for explanation. Assuming that the satellite pitch axis inertia is 5000kg.m2, the maximum resultant moment of the flywheel mass in the pitch axis is 0.2 Nm. The satellite movable part is SADA driven by a stepping motor through a reduction gear, and the motion period delta t of the satellite movable part is_dMaximum moment T acting on star body of 4.32s_d,maxAbout 0.1Nm, and a dynamic course of single step motion of about 1 s. And the sampling period h of the controller is 0.1 s. The following technical key points and calculation formulas are realized by on-board software and calculated in real time. Fig. 2 is a schematic structural diagram of a robust disturbance rejection complex control system according to an embodiment of the present invention. In the present system, it is preferred that,

(1) the periodic interference frequency is estimated from the angular velocity measurement information using a fast fourier transform.

A conventional time-decimated radix-2 FFT algorithm is employed. The data length (number of sequence points) N is taken to be 128. The input of the FFT algorithm module is satellite angular velocity measurement information sequences omega (k-N +1), … omega (k-1) and omega (k), and the output is the first harmonic frequency (fundamental frequency) f of periodic motion interference_d. F is obtained by FFT algorithm_dThen, the period value of the motion disturbance is obtained according to the following formula:

Δt_d＝1/f_d

(2) and designing a second-order extended state observer, obtaining a star acceleration estimated value from satellite angular velocity measurement information, and further calculating to obtain the amplitude of the periodic disturbance moment.

Designing a nonlinear discrete second-order extended state observer to obtain an angular acceleration estimated value z at the current moment₂(k) The formula is as follows:

in the formula, e (×) is observer error;

z₁() is the observer first order state quantity;

h is an angular velocity measurement period;

ω (k-1) is the angular velocity measurement;

β₁a first preset gain coefficient;

u (k) is a closed-loop control quantity;

j is the uniaxial moment of inertia of the star;

β₂a second preset gain factor;

fal (x, α, δ) is a non-linear function;

the relevant design factors are chosen as follows: beta is a₁＝30，β₂＝110。

According to the output result of the extended state observer, the amplitude T of the periodic disturbance torque is further calculated_d(k) The formula is as follows:

T_d(k)＝Jz₂(k)-u(k)

according to the accumulated time length (in this example, 5 Δ t_d) The maximum amplitude T of the periodic disturbance torque is calculated by a conventional extreme value determination method (in this example, the simplest sequential comparison method is used)_d,max。

(3) And calculating the equivalent disturbance moment feedforward compensation value in each compensation period.

3.1) will rise from 0 to 0.1T_d,maxIs denoted as t_sFrom T_d,maxDown to 0.1 × T_d,maxIs denoted as t_f. Will be Δ t_c＝t_f-t_sAs a period of each interference Δ t_dThe feed forward compensation duration in. Respectively calculating the number of sampling points in the interference period and the compensation duration, wherein the formula is as follows:

in the formula, Z (×) represents a rounding operation.

3.2) calculating a feedforward control quantity, wherein the formula is as follows:

(4) and designing a two-dimensional fuzzy logic system for adjusting the feedforward compensation moment value of the disturbance moment.

The fuzzy logic system is designed by the following steps:

4-1) determining the discourse domain of input and output variables and defining corresponding fuzzy sets

The discourse domain of the input variable and the output variable is respectively as follows:

T_d,max＝[0,0.15](Nm)，ω_max＝[0,0.01](°/s)，Δt_lead＝[0,4.3](s)；

the fuzzy sets are respectively:

T_d,max＝{S2,S1,M,B1,B2}

ω_max＝{S,M,B}

Δt_lead＝{S2,S1,M,B1,B2}

4-2) establishing a fuzzy rule base

The fuzzy rule base is described in the form of a fuzzy rule table, as shown in table 1.

4-3) determining membership functions of variables

In the invention, the membership function of each variable has no special requirement, so the same membership function is adopted. The membership function adopts a triangular membership function. The expression is as follows:

in the formula, x_iAs an input quantity, c_iIs the value of the ith equipartition point in the domain of discourse, b_iIs an adjustable parameter.

4-4) establishing a fuzzy system adopting a single-value fuzzy machine, a Mamdani inference machine and a central average ambiguity resolver

In the fuzzy inference engine, the inference type adopts a Mamdani fuzzy implication minimum operation method, and (and) operation adopts an intersection method (a small method), or (also/or) operation adopts a union method (a large method), and a maximum-minimum method is used for synthesis.

(5) And performing feedforward torque compensation according to the disturbance torque feedforward compensation moment value.

Feed-forward compensation time value delta t given by fuzzy logic system_leadConverting the delay value relative to the starting moment of the current calculation cycle into a delay value, wherein the conversion formula is as follows:

Δt_lag＝Δt_d-Δt_lead

(6) and dynamically adjusting the gain of the PID feedback control law according to whether the PID feedback control law is in the feedforward compensation period or not.

The feedback control law uses the most common PID control law.

During the non-feedforward compensation period, the PID control law of normal gain is adopted. For this example, the PID control law normal parameters are as follows: k_p＝30、K_i＝0.3、K_d40. During the feedforward torque compensation, a PID control law with higher gain is adopted, namely the proportional coefficient and the differential coefficient of the PID control law are properly increased on the basis of a normal value. For this example, K is taken during the feed-forward torque compensation_p＝50、K_i＝0.3、K_d＝65。

Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.