CN102298325B - Variable parameter control method of sine instruction - Google Patents

Abstract

The invention discloses a variable parameter control method of a sine instruction. According to the invention, real-time determination is carried out on frequencies and amplitudes of an instruction signal of a sine instruction control system; after a certain operation period, when processing on the frequencies and the amplitudes of the sine instruction is completed, adjustment is carried out on coefficients of a proportional P and a differential term D of an integration separation PID algorithm according to different frequencies and amplitudes, so that a control algorithm of a variable parameter is realized and thus a high control precision of the system is reached. According to the invention, a problem on an amplitude frequency response characteristic of a servo system in a wide range is resolved; moreover, a principle employed in the invention is simple and practical as well as reliability is high.

Description

A kind of variable parameter control method of sinusoidal instruction

Technical field

The present invention relates to the variable parameter control method of a kind of parameter control method, particularly a kind of sinusoidal instruction.

Background technology

Sinusoidal command signal is a kind of common instruction type in the servo-control system, especially has comparatively widely to use in the servocontrol of armament systems.The PID control method is simple, robustness good, reliability is high, it is the control method that is most widely used at present, but control object generally has non-linear, time uncertainty in actual application, when amplitude-frequency wider range of sinusoidal command signal, conventional non-variable parameter PID control method is difficult to the control effect that reaches desirable in the full frequency band scope because the control parameter is definite value.And for intelligence control methods such as fuzzies, its algorithm is realized relative all more complicated with parameter adjustment, and operand is large, and the arithmetic capability of system is had relatively high expectations.

Summary of the invention

The object of the present invention is to provide a kind of variable parameter control method of sinusoidal instruction, the solution servo-drive system satisfies the amplitude-frequency response characteristic problem in the full frequency band scope under relatively simple algorithm.

A kind of concrete steps of variable parameter control method of sinusoidal instruction are:

The first step is built the sinusoidal command control system of variable element

The sinusoidal command control system of variable element comprises: controller, feedback potentiometer, motor, speed reduction unit and signal processing module, frequency judge module, coefficient adjustment module, PID computing module, controlled quentity controlled variable output module; The feedback signal terminal of controller signals receiving end difference and instruction input end and feedback potentiometer links to each other, controller output end links to each other with the drive wire of motor, motor rotary shaft gear and the engagement of speed reduction unit input end, the output terminal of speed reduction unit and the engagement of rudderpost gear, feedback potentiometer is coaxial to be installed on the rudderpost; Each functional module realizes by the arithmetic processor of controller inside; The function of signal processing module is: realize the reading of signal, filtering; The function of frequency judge module is: frequency and the amplitude of processing and decision instruction signal; The function of coefficient adjustment module is: scale-up factor and the differential coefficient of adjusting PID; The function of PID computing module is: carry out the PID computing; The function of controlled quentity controlled variable output module is: finish the output controlled quentity controlled variable.

Second step signal processing module acquisition instructions and feedback signal

Signal processing module reads in respectively instruction and feedback signal, then instruction and feedback signal is sent to the step low-pass Butterworth filter, carries out the signal filtering computing.

The 3rd synchronizing frequency judge module processing instruction signal amplitude and frequency

The frequency judge module is processed the command signal that collects and is judged; At first the decision instruction signal is in uphill process or decline process, when command signal value increases continuously, be judged as uphill process, otherwise be judged as the decline process, the continuous several times variation range is (ε is the fluctuating error of AD conversion) during less than definite value ε, then is judged as nonsinusoidal signal; When being judged as uphill process, after maximal value occurring, signal value begins to reduce gradually, then thinks the top flex point that sinusoidal signal occurs, puts the maximal value zone bit, this moment the value of recording the time and maximal value; When being judged as the decline process, after minimum value occurring, signal value begins to increase gradually, then thinks the bottom flex point that sinusoidal signal occurs, puts minimum value zone bit and the value of recording the time and minimum value; After maximal value and minimum value all occurred, then the mistiming by its time of occurrence calculated the frequency of signal, and the difference by maximal value and minimum value calculates amplitude.

The 4th step coefficient adjustment module is calculated the PID coefficient

, be updated in the adjustment amount function of proportional P and differential term D as the input of coefficient adjustment module according to the frequency of the command signal that obtains and amplitude, calculate it and adjust coefficient μ (A, f); Its functional relation can be expressed as:

(1)

In the formula (1), k _ABe the coefficient of relationship of proportional P or differential term D and amplitude A, k _fCoefficient of relationship for proportional P or differential term D and frequency f.

The 5th step PID computing module carries out the PID computing

The coefficient adjustment module calculates to adjust and sends to the PID computing module behind the coefficient μ (A, f) and carry out the PID computing;

The PID control that each cycle of sinusoidal instruction adopts integration to separate; In beginning during trace command, system deviation is cancelled first integral action during greater than setting value, adopts PD control; When deviation adds integral action during less than setting value again, adopt PID control;

According to actual conditions, artificial setting threshold e〉0; As | e (k) |〉during e, adopt PD control; When | e (k) |≤e, adopt PID control; Its control algolithm can be expressed as:

(2)

In the formula (2), T is control cycle; Kp is scale-up factor; Ki is integral coefficient; Kd is differential coefficient; U (k) is the controller output in k cycle; E (k) is the error signal in k cycle; E (k-1) is the error signal in k-1 cycle; β is the integration separation coefficient, as | e (k) | get 0 during e, when | e (k) | get 1 during≤e;

When each sinusoidal instruction cycle does not finish, the parameter constant that keeps last time, every through half instruction cycle, finish the judgement of offset of sinusoidal instruction cycle and amplitude, calculate the adjustment coefficient μ (A of corresponding P or D coefficient according to the coefficient adjustment module, f), thereby the value of kp and ki is regulated, realized VARIABLE PARAMETER PID CONTROL; Its control algolithm is expressed as:

(3)

The 6th step controlled quentity controlled variable output module output controlled quentity controlled variable

The controlled quentity controlled variable u (k) of the motor that formula (3) is calculated exports to motor, thus the control of completion system.

The present invention distinguishes by amplitude and the frequency of offset of sinusoidal instruction, accordingly the control parameter of Kind of Integration Separation PID Controller's is adjusted, and has realized a kind of control algolithm of variable element.The principle of the invention is simple, practical, and reliability is higher, is applied in the Electrodynamic Rudder System of certain model at present.

Embodiment

A kind of concrete steps of variable parameter control method of sinusoidal instruction are:

The first step is built the sinusoidal command control system of variable element

The sinusoidal command control system of variable element comprises: controller, feedback potentiometer, motor, speed reduction unit and signal processing module, frequency judge module, coefficient adjustment module, PID computing module, controlled quentity controlled variable output module; The feedback signal terminal of controller signals receiving end difference and instruction input end and feedback potentiometer links to each other, controller output end links to each other with the drive wire of motor, motor rotary shaft gear and the engagement of speed reduction unit input end, the output terminal of speed reduction unit and the engagement of rudderpost gear, feedback potentiometer is coaxial to be installed on the rudderpost; Each functional module realizes by the arithmetic processor of controller inside; The function of signal processing module is: realize the reading of signal, filtering; The function of frequency judge module is: frequency and the amplitude of processing and decision instruction signal; The function of coefficient adjustment module is: scale-up factor and the differential coefficient of adjusting PID; The function of PID computing module is: carry out the PID computing; The function of controlled quentity controlled variable output module is: finish the output controlled quentity controlled variable.

Second step: signal processing module acquisition instructions and feedback signal

Signal processing module reads in respectively instruction and feedback signal, then instruction and feedback signal is sent to the step low-pass Butterworth filter, carries out the signal filtering computing.

The 3rd step: frequency judge module processing instruction signal amplitude and frequency

The frequency judge module is processed the command signal that collects and is judged.Command signal value increases continuously, is uphill process, and the continuous several times variation range is (ε is the fluctuating error of AD conversion) during less than definite value ε, then is nonsinusoidal signal; In the uphill process, maximal value appears and after, signal value begins to reduce gradually, the top flex point of sinusoidal signal then occurs, puts the maximal value zone bit, this moment the value of recording the time and maximal value; In the decline process, minimum value appears and after, signal value begins to increase gradually, the bottom flex point of sinusoidal signal then occurs, puts minimum value zone bit and the value of recording the time and minimum value.After maximal value and minimum value all occurred, the mistiming by its time of occurrence calculated the frequency of signal, and the difference by maximal value and minimum value calculates amplitude.

The 4th step: the coefficient adjustment module is calculated the PID coefficient

, be updated in the adjustment amount function of proportional P and differential term D as the input of coefficient adjustment module according to the frequency of the command signal that obtains and amplitude, calculate it and adjust coefficient μ (A, f).Its functional relation can be expressed as:

(1)

In the formula (1), k _ABe the coefficient of relationship of proportional P or differential term D and amplitude A, k _fCoefficient of relationship for proportional P or differential term D and frequency f.

The 5th step: the PID computing module carries out the PID computing

The coefficient adjustment module calculates to adjust and sends to the PID computing module behind the coefficient μ (A, f) and carry out the PID computing.

The PID control that each cycle of sinusoidal instruction adopts integration to separate.In beginning during trace command, system deviation is cancelled first integral action during greater than setting value, adopts PD control; When deviation adds integral action during less than setting value again, adopt PID control.

According to actual conditions, artificial setting threshold e〉0; As | e (k) |〉during e, adopt PD control; When | e (k) |≤e, adopt PID control.Its control algolithm can be expressed as:

(2)

In the formula (2), T is control cycle; Kp is scale-up factor; Ki is integral coefficient; Kd is differential coefficient; U (k) is the controller output in k cycle; E (k) is the error signal in k cycle; E (k-1) is the error signal in k-1 cycle; β is the integration separation coefficient, as | e (k) | get 0 during e, when | e (k) | get 1 during≤e.

When each sinusoidal instruction cycle does not finish, the parameter constant that keeps last time, every through half instruction cycle, finish the judgement of offset of sinusoidal instruction cycle and amplitude, calculate the adjustment coefficient μ (A of corresponding P or D coefficient according to the coefficient adjustment module, f), thereby the value of kp and ki is regulated, realized VARIABLE PARAMETER PID CONTROL.Its control algolithm can be expressed as:

(3)

The 6th step: controlled quentity controlled variable output module output controlled quentity controlled variable

The controlled quentity controlled variable u (k) of the motor that formula (3) is calculated exports to motor, thus the control of completion system.

Claims (1)

1. a variable parameter control method that is used for the sinusoidal instruction of servo-drive system of armament systems is characterized in that the concrete steps of this method are;

The first step is built the sinusoidal command control system of variable element

The sinusoidal command control system of variable element comprises: controller, feedback potentiometer, motor, speed reduction unit, signal processing module, frequency judge module, coefficient adjustment module, PID computing module and controlled quentity controlled variable output module; The feedback signal terminal of controller signals receiving end difference and instruction input end and feedback potentiometer links to each other, controller output end links to each other with the drive wire of motor, motor rotary shaft gear and the engagement of speed reduction unit input end, the output terminal of speed reduction unit and the engagement of rudderpost gear, feedback potentiometer is coaxial to be installed on the rudderpost; Each functional module realizes by the arithmetic processor of controller inside; The function of signal processing module is: realize the reading of signal, filtering; The function of frequency judge module is: frequency and the amplitude of processing and decision instruction signal; The function of coefficient adjustment module is: scale-up factor and the differential coefficient of adjusting PID; The function of PID computing module is: carry out the PID computing; The function of controlled quentity controlled variable output module is: finish the output controlled quentity controlled variable;

Second step signal processing module acquisition instructions and feedback signal

Signal processing module reads in respectively instruction and feedback signal, then instruction and feedback signal is sent to the step low-pass Butterworth filter, carries out the signal filtering computing;

The 3rd synchronizing frequency judge module processing instruction signal amplitude and frequency

The frequency judge module is processed the command signal that collects and is judged; At first the decision instruction signal is in uphill process or decline process, when command signal value increases continuously, is judged as uphill process, otherwise be judged as the decline process, the continuous several times variation range is during less than definite value ε, and ε is the fluctuating error of AD conversion, then is judged as nonsinusoidal signal; When being judged as uphill process, after maximal value occurring, signal value begins to reduce gradually, then thinks the top flex point that sinusoidal signal occurs, puts the maximal value zone bit, this moment the value of recording the time and maximal value; When being judged as the decline process, after minimum value occurring, signal value begins to increase gradually, then thinks the bottom flex point that sinusoidal signal occurs, puts minimum value zone bit and the value of recording the time and minimum value; After maximal value and minimum value all occurred, then the mistiming by its time of occurrence calculated the frequency of signal, and the difference by maximal value and minimum value calculates amplitude;

The 4th step coefficient adjustment module is calculated the PID coefficient

, be updated in the adjustment amount function of proportional P and differential term D as the input of coefficient adjustment module according to the frequency of the command signal that obtains and amplitude, calculate it and adjust coefficient μ (A, f); Its functional relation is expressed as:

μ(A,f)＝1+k _A+k _f (1)

In the formula (1), k _ABe the coefficient of relationship of proportional P or differential term D and amplitude A, k _fCoefficient of relationship for proportional P or differential term D and frequency f;

The 5th step PID computing module carries out the PID computing

The coefficient adjustment module calculates to adjust and sends to the PID computing module behind the coefficient μ (A, f) and carry out the PID computing;

The PID control that each cycle of sinusoidal instruction adopts integration to separate; In beginning during trace command, system deviation is cancelled first integral action during greater than setting value, adopts PD control; When deviation adds integral action during less than setting value again, adopt PID control;

According to actual conditions, artificial setting threshold e〉0; As | e (k) |〉during e, adopt PD control; When | e (k) |≤e, adopt PID control; Its control algolithm is expressed as:

$u\left(k\right)=k_{p}e\left(k\right)+\mathrm{\β}\×k_i\underset{j=0}{\overset{k}{\mathrm{\Σ}}}e\left(j\right)T+k_d(e\left(k\right)-e(k-1))/T---\left(2\right)$

In the formula (2), T is control cycle; Kp is scale-up factor; Ki is integral coefficient; Kd is differential coefficient; U (k) is the controller output in k cycle; E (k) is the error signal in k cycle; E (k-1) is the error signal in k-1 cycle; β is the integration separation coefficient, as | e (k) | during e, get 0; When | e (k) | during≤e, get 1;

When each sinusoidal instruction cycle does not finish, the parameter constant that keeps last time, every through half instruction cycle, finish the judgement of offset of sinusoidal instruction cycle and amplitude, calculate the adjustment coefficient μ (A of corresponding P or D coefficient according to the coefficient adjustment module, f), thereby the value of kp and ki is regulated, realized VARIABLE PARAMETER PID CONTROL; Its control algolithm is expressed as:

$u\left(k\right)=k_p\×\mathrm{\μ}(A,f)e\left(k\right)+\mathrm{\β}\×k_i\underset{j=0}{\overset{k}{\mathrm{\Σ}}}e\left(j\right)T+k_d\×\mathrm{\μ}(A,f)(e\left(k\right)-e(k-1))/T---\left(3\right)$

The 6th step controlled quentity controlled variable output module output controlled quentity controlled variable

The controlled quentity controlled variable u (k) of the motor that formula (3) is calculated exports to motor, thus the control of completion system.