CN118226746A - PID parameter directed self-adaptive intelligent setting method - Google Patents

PID parameter directed self-adaptive intelligent setting method Download PDF

Info

Publication number
CN118226746A
CN118226746A CN202311849283.2A CN202311849283A CN118226746A CN 118226746 A CN118226746 A CN 118226746A CN 202311849283 A CN202311849283 A CN 202311849283A CN 118226746 A CN118226746 A CN 118226746A
Authority
CN
China
Prior art keywords
pid
parameters
model
current
function
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202311849283.2A
Other languages
Chinese (zh)
Inventor
刘华伟
王学锋
楚维
黄业茂
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Today Information Technology Co ltd
Original Assignee
Shanghai Today Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Today Information Technology Co ltd filed Critical Shanghai Today Information Technology Co ltd
Priority to CN202311849283.2A priority Critical patent/CN118226746A/en
Publication of CN118226746A publication Critical patent/CN118226746A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a PID parameter directed self-adaptive intelligent setting method, which comprises the steps of constructing an improved PID model 1, constructing a PID parameter evaluation model 2 and constructing a directed self-adaptive intelligent setting model 3; compared with the prior art, the invention has the advantages that: the invention adopts an incremental improved PID model, not only can form effective control quantity, but also can effectively reduce the error accumulated in the whole process; and constructing a rapid, stable and accurate evaluation function of the response curve, and continuously optimizing the parameters of the tuning PID control through a directed self-adaptive intelligent tuning model to obtain optimal tuning parameters. The method not only can enable the parameters to be directionally adjusted towards the optimal parameter direction in the setting process, reduce the setting time, but also can find the optimal parameter according to the step length, and improve the accuracy of PID control. Actual tests show that the method has the setting efficiency improved by about 57 percent compared with the conventional method, has the control precision improved by about 63 percent, and can be applied to setting parameters of a simulation model and parameters of a control system.

Description

PID parameter directed self-adaptive intelligent setting method
Technical Field
The invention relates to the technical field of automatic control, in particular to a PID parameter directed self-adaptive intelligent setting method.
Background
In an industrial loop control system, control based on proportion (Proportion), integration (Integration) and Differentiation (Differentiation) is abbreviated as PID control, and the principle is to dynamically adjust the control amount according to the control deviation e (t) between the actual output value and the calculated reference value so that e (t) tends to 0 as quickly and accurately as possible. The mathematical expression of its control is generally:
Wherein: u (t) is the control output, and kp, ki and kd are the weight coefficients of proportion, integral and derivative respectively.
For PID control, the speed and precision of the control often depend on whether the three weight coefficients kp, ki and kd are reasonable or not, and because the three weight coefficients are generally fixed constants due to the requirement of robustness of the control system, parameter setting needs to be performed in advance, so that the optimal PID parameters are obtained, and the aim of optimizing the control system is fulfilled.
The traditional PID parameter setting method depends on artificial experience, on one hand, higher precision cannot be achieved, on the other hand, the performance of the control system depends on the degree of experience enrichment to a great extent, and therefore the method is particularly important for researching the PID parameter setting method.
Disclosure of Invention
Aiming at the defects of the existing PID parameter tuning technique method, the invention obtains the optimal tuning parameters of the PID controller by utilizing the directional self-adaptive intelligent tuning, thereby providing the directional self-adaptive intelligent tuning method for the PID parameters, which mainly comprises the following steps: and constructing an improved PID model 1, a PID parameter evaluation model 2 and a directional self-adaptive intelligent setting model 3. Building an improved PID model 1: three state quantities, namely a current error e (t), a previous error e (t-1) and a previous error e (t-2), are introduced to obtain an improved PID model:
u(t)＝Δu(t)+u(t-1)＝kp.[e(t)-e(t-1)]+ki.e(t)+kd.[e(t)-2e(t
-1)+e(k-2)]+u(t-1)
and (3) constructing a PID parameter evaluation model 2: the PID parameter evaluation function f is constructed as the sum of the rapidity function f1, the stability function f2 and the accuracy function f 3:
f= f1+f2+f3=ω1.t2+ω2.kv (t) one Viealdt +ω3.1v (t) one Vieal dt
Constructing a directional self-adaptive intelligent setting model 3: random perturbations of a specific step size (sp, si, sd) are added on the basis of the current PID parameters (kp, ki, kd), when new PID parameters (k=kp+sp, ki=ki+si, k=kd+sd)
The obtained evaluation function is superior to the previous PID parameters (kp, ki, kd), the current PID parameters are used as new PID parameters (kp=k, ki=ki, kd=k), and the PID parameters are adjusted to continuously adjust a step length to the current optimization direction; otherwise, the PID parameters are not updated, and random disturbance added with specific step length (sp, si, sd) is continuously optimized
Setting PID parameters until the PID evaluation function value f is not possible to be optimized, taking current PID parameters kp, ki and kd as optimal PID control parameters, namely: k=kp, k=ki, k=kd, fopt=f.
Compared with the prior art, the invention improves the whole process integral control of the traditional PID control, and adopts an incremental improved PID model, thereby not only forming effective control quantity, but also effectively reducing the accumulated error in the whole process; on the basis of the requirements of the traditional PID parameter setting on the rapidity, stability and accuracy of the response curve, a rapid, stability and accuracy evaluation function of the response curve is constructed, and parameters of PID control are continuously optimized through a directed self-adaptive intelligent setting model, so that optimal setting parameters are obtained. The method not only can enable the parameters to be directionally adjusted towards the optimal parameter direction in the setting process, reduce the setting time, but also can find the optimal parameter according to the step length, and improve the accuracy of PID control. Actual tests show that the method has the setting efficiency improved by about 57 percent compared with the conventional method, has the control precision improved by about 63 percent, and can be applied to setting parameters of a simulation model and parameters of a control system.
In order to solve the problems, the technical scheme of the invention is as follows: comprises the following parts: constructing an improved PID model, a PID parameter evaluation model and a directed self-adaptive intelligent setting model;
The improved PID model is constructed: three state quantities, namely a current error e (t), a previous error e (t-1) and a previous error e (t-2), are introduced to obtain an improved PID model:
u (t) =Δu (t) +u (t 1) =k p [ e (t) to e (t 1) ]+k i.e(t)+kd ] [ e (t) to 2e (t 1) +e (k 2) ]+u (t 1);
The PID parameter evaluation model is constructed: the PID parameter evaluation function f is constructed as the sum of the rapidity function f1, the stability function f2 and the accuracy function f 3:
The directional self-adaptive intelligent setting model is constructed: adding random disturbance of a specific step length (sp, si, sd) on the basis of the current PID parameters (kp, ki, kd), when the new PID parameters (k=kp+sp, ki=ki+si, k=kd+sd) are superior to the previous PID parameters (kp, ki, kd), taking the current PID parameters as the new PID parameters (kp=k, ki=ki, kd=k), and adjusting the PID parameters to continuously adjust one step length to the current optimization direction; otherwise, the PID parameters are not updated, and the random disturbance of a specific step length (sp, si, sd) is continuously added to carry out optimization PID parameter setting until the PID evaluation function value f is not optimized, and the current PID parameters kp, ki, kd are taken as the optimal PID control parameters, namely
Further, an improved PID model is constructed, wherein three state quantities, namely a current error e (t), a previous error e (t-1) and a previous error e (t-2) are introduced, and the improved PID model is obtained:
u(t)＝Δu(t)+u(t-1)＝kp.[e(t)-e(t-1)]+ki.e(t)+kd.[e(t)-2e(t-1)+e(k-2)]+u(t-1)
Wherein: u (t) is the control output quantity at the current moment, u (t-1) is the control output quantity at the last moment, and Deltau (t) is the control output increment required to be calculated.
Further, the constructed PID parameter evaluation model specifically includes:
Step one, rapidly evaluating PID control, and constructing a PID control rapid function f1 as the product of time t2 and weight coefficient omega 1 required by the second crossing of the ideal value Vieal by the response curve V (t), namely: f1 =ω1.t2;
step two, PID control stability evaluation, namely constructing a PID control stability function f2 as the product of the integral absolute value between the ideal value Vieal of the 1 st pass and the ideal value of the kth pass of the response curve V (t) and the weight coefficient omega 2, namely
Thirdly, evaluating the accuracy of PID control, namely constructing a PID control accuracy function f3 as the product of an integral absolute value and a weight coefficient omega 3 between the k+1th pass ideal value Vieal and the last (Nth) pass ideal value Vieal of the response curve V (t) after the influence curve V (t) is stable, namely:
Step four: the PID parameter evaluation function f is constructed as the sum of the rapidity function f1, the stability function f2 and the accuracy function f 3:
wherein: ω1+ω2+ω3=1.
Further, a PID directed adaptive intelligent tuning model can be constructed based on a parameter evaluation model, the principle is that random disturbance of a specific step length (sp, si, sd) is added based on current PID parameters (kp, ki, kd), when an evaluation function obtained by new PID parameters (k=kp+sp, ki=ki+si, k=kd+sd) is better than the previous PID parameters (kp, ki, kd), the current PID parameters are used as new PID parameters (kp=k, ki=ki, kd=k), and the PID parameters are adjusted to continuously adjust one step length to the current optimization direction; otherwise, not updating the PID parameters, continuing to add random disturbance of a specific step length (sp, si, sd) to carry out optimization PID parameter setting, and specifically comprising the following steps:
Initializing various parameters, namely endowing initial values to PID parameters kp, ki and kd, endowing step sizes sp, si and sd initial values according to the requirements of a PID control system on precision, and respectively endowing different weight coefficients omega 1, omega 2 and omega 3 according to the emphasis degree of a comparison example item kp, a differential item ki and an integral item kd;
Step two, according to the obtained evaluation function under the current initial PID parameters kp, ki and kd as the current evaluation function value f, changing the PID parameters kp, ki and kd to adjust the PID parameters in random steps sp, si and sd respectively to obtain new PID parameters k=kp+sp ki=ki+si and k=kd+sd, on the basis, calculating a new evaluation function value f, judging the size between f and f, and if f is less than f, executing the step three; if f is f, executing the fourth step;
Step three, taking new PID parameters k, ki as current optimal parameters, and updating the evaluation function values of the current optimal parameters, namely kp=k, ki=ki, kd=k, f=f, and recording the step-size adjustment direction sign function sgn ():
The value of sgn (-) is only 0, -1 and 1, after which Step 5 is performed
Step four, randomly selecting new PID parameters k, ki in a range of plus or minus 10 times of step length, namely:
Wherein: rand (), which is a random valued function, is able to take a random value from between two set minimum maximum values.
On the basis of new PID parameters k, ki, calculating a new evaluation function value f, comparing f with f, if f is less than f, updating the PID parameters to be new k, ki is: kp=k, ki=ki, kd=k, f=f;
step five, repeatedly executing the step two to the step four until the PID evaluation function value f is not possible to be optimized, and taking the current PID parameters kp, ki and kd as optimal PID control parameters, namely:
Compared with the prior art, the invention has the advantages that:
1. The invention improves the integral control of the whole process of the traditional PID control, adopts an incremental improved PID model, not only can form effective control quantity, but also can effectively reduce the accumulated error in the whole process; on the basis of the requirements of the traditional PID parameter setting on the rapidity, stability and accuracy of the response curve, a rapid, stability and accuracy evaluation function of the response curve is constructed, and parameters of PID control are continuously optimized through a directed self-adaptive intelligent setting model, so that optimal setting parameters are obtained. The method not only can enable the parameters to be directionally adjusted towards the optimal parameter direction in the setting process, reduce the setting time, but also can find the optimal parameter according to the step length, and improve the accuracy of PID control. Actual tests show that the method has the setting efficiency improved by about 57 percent compared with the conventional method, has the control precision improved by about 63 percent, and can be applied to setting parameters of a simulation model and parameters of a control system.
Drawings
Fig. 1 shows an overall architecture diagram of a PID parameter directed adaptive intelligent tuning method of the present invention.
FIG. 2 shows a schematic diagram representing an improved PID model in an embodiment of the invention.
FIG. 3 shows a computational schematic diagram representing a PID parameter assessment model in an embodiment of the invention.
FIG. 4 shows a flow chart of a directed adaptive smart tuning model in one embodiment of the invention.
FIG. 5 is a diagram showing simulation results of a directed adaptive intelligent tuning method using PID parameters in an embodiment of the invention.
FIG. 6 shows a graph of comparison simulation results for an applied and non-applied PID directed adaptive intelligent tuning method.
Detailed Description
Specific embodiments of the present invention will be further described below with reference to the accompanying drawings. Wherein like parts are designated by like reference numerals.
It should be noted that the words "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to directions in the drawings, and the words "inner" and "outer" refer to directions toward or away from, respectively, the geometric center of a particular component.
In order to make the contents of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
As shown in fig. 1-5, a directional adaptive intelligent tuning method for PID parameters mainly includes: building improved PID model 1, building PID parameter evaluation model 2 and building directed adaptation
The parts of the model 3 should be intelligently set.
Step 1: an improved PID model 1 was constructed. Because the conventional PID algorithm employs full process integral control,
With the increase of time, the error accumulation gradually increases, and the proportional term and the integral term are only related to the current error, so that the control parameter is difficult to set, and when the current output value deviates greatly from the reference value, an effective control quantity cannot be formed. Thus, an improved PID model is constructed in which three state quantities are introduced, namely the current error e (t),
The error e (t-1) at the previous moment and the error e (t-2) at the previous moment are used for obtaining an improved PID model:
u(t)＝Δu(t)+u(t-1)＝kp.e(t)-e(t-1)+ki.e(t)+kd.e(t)-2e(t-1)+e(k-2)+u(t-1)
Wherein: u (t) is the control output quantity at the current moment, u (t-1) is the control output quantity at the last moment, and Deltau (t) is the control output increment required to be calculated.
Step two: and constructing a PID parameter evaluation model 2. For the quality of PID control, a reasonable PID parameter evaluation model needs to be constructed to optimally set parameters of a control system. In the conventional PID parameter setting, it is generally determined whether the PID parameter of a system is optimal based on the rapidity, stability and accuracy of the response curve of the PID control system, so the PID parameter evaluation model constructed by the present patent specifically includes:
Step1: PID control rapidity evaluation. Construction of PID control rapidity function f1 as response curve V (t) th
The product of time t2 and weight coefficient ω1 required for the secondary crossing of ideal value Vieal is: f1 =ω1.t2.
Step 2: and (5) evaluating the stability of PID control. Construction of PID control stability function f2 as response curve V (t) self-correlation
Multiplication of absolute value of integral between ideal value Vieal of 1 st pass and ideal value of kth pass and weight coefficient omega 2
The product is: f2 Let ω2.kv (t) one Vieal dt.
Step 3: and (5) evaluating the accuracy of PID control. After the PID control accuracy function f3 is constructed to be stable in the influence curve V (t), the product of the integral absolute value and the weight coefficient omega 3 between the k+1th pass ideal value Vieal and the last (Nth) pass ideal value Vieal of the response curve V (t), namely: f3 =ω3.1v (t) one Vieal dt.
Step 4: the PID parameter evaluation function f is constructed as the sum of the rapidity function f1, the stability function f2 and the accuracy function f 3:
f= f1+f2+f3=ω1.t2+ω2.kv (t) one Viealdt +ω3.1v (t) one Vieal dt
Wherein: ω1+ω2+ω3=1.
Step three: and constructing a directional self-adaptive intelligent setting model 3. Based on the parameter evaluation model, a PID directed self-adaptive intelligent tuning model can be constructed, the principle is that random disturbance with specific step length (sp, si, sd) is added on the basis of the current PID parameters (kp, ki, kd), and when new PID parameters (k=kp+sp, ki=ki+si, k=kd+sd)
The obtained evaluation function is superior to the previous PID parameters (kp, ki, kd), the current PID parameters are used as new PID parameters (kp=k, ki=ki, kd=k), and the PID parameters are adjusted to continuously adjust a step length to the current optimization direction; otherwise, not updating the PID parameters, continuing to add random disturbance of a specific step length (sp, si, sd) to carry out optimization PID parameter setting, and specifically comprising the following steps:
Step1: initializing various parameters. Initial values are given to PID parameters kp, ki, kd, and initial values of step sizes sp, si, sd are given according to the requirements of the PID control system for precision, and different weight coefficients ω1, ω2 and ω3 are given according to the degree of emphasis of the comparison example term kp, the differential term ki and the integral term kd, respectively.
Step2: according to the evaluation function obtained in the second step under the current initial PID parameters kp, ki, kd as the current evaluation function value f, changing the PID parameters kp, ki, kd to make them randomly step size sp, si, sd respectively to adjust, obtaining new PID parameters k=kp+sp, ki=ki+si, k=kd+sd, on which the new evaluation function value f can be calculated. Judging the size between f and f, if f is less than f, executing Step 3; if f > f, step4 is performed.
Step 3: taking new PID parameters k, ki as current optimal parameters, and updating the evaluation function values of the current optimal parameters, namely: kp=k, ki=ki, kd=k, and f=f. And recording a step-size adjustment direction sign function sgn ():
sgn(kp)＝kkp,sgn(ki)＝ki*ki,sgn(kd)＝k kd
It can be seen that the values of sgn (-) are only 0, -1 and 1. Step 5 is then performed.
Step 4: new PID parameters k, ki x, k are randomly selected within a range of plus or minus 10 times of step length, namely: k=rand (kp-10 sp, kp+10sp), ki=rand (ki-10 si, ki+10si), k=rand (kd-10 sd, kd+10sd)
Wherein: rand (), which is a random valued function, is able to take a random value from between two set minimum maximum values. And calculating a new evaluation function value f on the basis of the new PID parameters k, ki. And comparing the magnitudes of f and f, if f is less than f, updating the PID parameters to new k, ki, k, namely: kp=k, ki=ki, kd=k, and f=f.
The method is used for enabling the PID to find the random disturbance of the range added by the optimal PID parameters more quickly in the directed self-adaptive intelligent setting process.
Step 5: step 2 to Step 4 are repeatedly executed until the PID evaluation function value f is no longer possible to be optimized, and at this time, the current PID parameter kp, ki, kd is taken as the optimal PID control parameter, namely: k=kp, k=ki, k=kd, f opt=f.
An embodiment of the present invention shows an improved PID model 1 as shown in FIG. 2. The proportion adjusting module 2-1 is affected by kp, the current error and the error at the last moment, and the output of the proportion adjusting module is kp.e (t) to e (t-1); the integral regulating module 2-2 is affected by the common influence of ki and the current error, and the output of the integral regulating module is ki.e (t); the differential mediation module 2-3 is affected by kd, the current error and the first two time errors together, and its output is kd.e (t) -2e (t-1) +e (k-2). The three regulation modules of proportion, integral and differential act together to obtain the increment delta u (t) of the control output.
An embodiment of the present invention is shown in FIG. 3, which shows a PID parameter assessment model 2. It can be seen that the functions f1, f2 and f3 representing rapidity, stability and accuracy represent the front, middle and rear segments, respectively, of the corresponding curve V (t) on the time axis. And the response curve is gradually close to the ideal value video
A directed adaptive intelligent tuning model 3 in one embodiment of the invention is shown in fig. 4. Adding random disturbance of specific step length (sp, si, sd) based on current PID parameters (kp, ki, kd), when new PID parameters
(K=kp+sp, ki=ki+si, k=kd+sd) and the resulting evaluation function is better than the previous PID parameters (kp, ki, kd), the current PID parameters are taken as new PID parameters (kp=k, ki=ki, kd=k), and the PID parameters are adjusted to the current PID parameters
Continuously adjusting a step length in the front optimization direction; otherwise, the PID parameters are not updated, and the random disturbance of a specific step length (sp, si, sd) is added to carry out optimization PID parameter setting until the PID evaluation function value f does not have any optimization any more
Possibly, the current PID parameters kp, ki, kd are taken as the optimal PID control parameters, namely:
k＝kp,k＝ki,k＝kd,f opt＝f。
The simulation result of the directional adaptive intelligent setting method using PID parameters is shown in FIG. 5. It can be seen that, along with the continuous superposition of simulation cycle times, the corresponding curve of the PID parameter directional self-adaptive intelligent setting method can be greatly improved from the aspects of rapidness, stability and accuracy.
The comparison simulation results of the PID directed adaptive intelligent tuning method of the present invention with and without PID are shown in FIG. 6. It can be seen that the corresponding curve using the PID directed adaptive intelligent tuning method is able to reach the desired value more quickly (the time for the second time of the PID control response curve using the method of the present invention to reach the desired value is 1572 seconds, whereas the time for the non-application of the method of the present invention is 684 seconds), and to stabilize around the desired value, and the final accuracy is also better than the final accuracy for the non-application of the method of the present invention (the variance between the final and desired values of the PID control response curve using the method of the present invention is 39, whereas the variance for the non-application of the method of the present invention is 107).
The foregoing description of the invention has been presented for purposes of illustration and description, and is not intended to be limiting. Several simple deductions, modifications or substitutions may also be made by a person skilled in the art to which the invention pertains, based on the idea of the invention. Any person skilled in the art, within the scope of the present invention, shall make any changes or modifications to the present invention, such as changing the name of the improved PID model, changing the name of the PID parameter evaluation model, changing the name of the directed adaptive smart tuning model, and changing the model architecture and algorithm flow representing a method for directed adaptive smart tuning of the PID parameters, and applying the same structure and algorithm to other PID control, parameter tuning scenarios, etc., shall fall within the scope of the present invention as defined by the appended claims.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.

Claims (4)

1. A PID parameter directed self-adaptive intelligent setting method is characterized in that: comprises the following parts: constructing an improved PID model, a PID parameter evaluation model and a directed self-adaptive intelligent setting model;
The improved PID model is constructed: three state quantities, namely a current error e (t), a previous error e (t-1) and a previous error e (t-2), are introduced to obtain an improved PID model:
u(t)＝Δu(t)+u(t-1)＝kp.[e(t)-e(t-1)]+ki.e(t)+kd.[e(t)-2e(t-1)+e(k-2)]+u(t-1)；
The PID parameter evaluation model is constructed: the PID parameter evaluation function f is constructed as the sum of the rapidity function f1, the stability function f2 and the accuracy function f 3:
The directional self-adaptive intelligent setting model is constructed: adding random disturbance of a specific step length (sp, si, sd) on the basis of the current PID parameters (kp, ki, kd), when the new PID parameters (k=kp+sp, ki=ki+si, k=kd+sd) are superior to the previous PID parameters (kp, ki, kd), taking the current PID parameters as the new PID parameters (kp=k, ki=ki, kd=k), and adjusting the PID parameters to continuously adjust one step length to the current optimization direction; otherwise, the PID parameters are not updated, and the random disturbance of a specific step length (sp, si, sd) is continuously added to carry out optimization PID parameter setting until the PID evaluation function value f is not optimized, and the current PID parameters kp, ki, kd are taken as the optimal PID control parameters, namely fopt＝f。
2. The directional adaptive intelligent tuning method for PID parameters according to claim 1, characterized in that: the method comprises the steps of constructing an improved PID model, wherein three state quantities, namely a current error e (t), a previous error e (t-1) and a previous error e (t-2), are introduced to obtain the improved PID model:
u(t)＝Δu(t)+u(t-1)＝kp.[e(t)-e(t-1)]+ki.e(t)+kd.[e(t)-2e(t-1)+e(k-2)]+u(t-1)
Wherein: u (t) is the control output quantity at the current moment, u (t-1) is the control output quantity at the last moment, and Deltau (t) is the control output increment required to be calculated.
3. The directional adaptive intelligent tuning method for PID parameters according to claim 1, characterized in that: the constructed PID parameter evaluation model
The body comprises:
Step one, rapidly evaluating PID control, and constructing a PID control rapid function f1 as the product of time t2 and weight coefficient omega 1 required by the second crossing of the ideal value Vieal by the response curve V (t), namely: f1 =ω1.t2;
step two, PID control stability evaluation, namely constructing a PID control stability function f2 as the product of the integral absolute value between the ideal value Vieal of the 1 st pass and the ideal value of the kth pass of the response curve V (t) and the weight coefficient omega 2, namely
Thirdly, evaluating the accuracy of PID control, namely constructing a PID control accuracy function f3 as the product of an integral absolute value and a weight coefficient omega 3 between the k+1th pass ideal value Vieal and the last (Nth) pass ideal value Vieal of the response curve V (t) after the influence curve V (t) is stable, namely:
Step four: the PID parameter evaluation function f is constructed as the sum of the rapidity function f1, the stability function f2 and the accuracy function f 3:
wherein: ω1+ω2+ω3=1.
4. The directional adaptive intelligent tuning method for PID parameters according to claim 1, characterized in that: the PID directed self-adaptive intelligent setting model can be constructed on the basis of a parameter evaluation model, and the principle is that random disturbance of a specific step length (sp, si, sd) is added on the basis of current PID parameters (kp, ki, kd), when an evaluation function obtained by new PID parameters (k=kp+sp, ki=ki+si, k=kd+sd) is superior to the previous PID parameters (kp, ki, kd), the current PID parameters are used as new PID parameters (kp=k, ki=ki, kd=k), and the PID parameters are adjusted to continuously adjust one step length to the current optimization direction; otherwise, not updating the PID parameters, continuing to add random disturbance of a specific step length (sp, si, sd) to carry out optimization PID parameter setting, and specifically comprising the following steps:
Initializing various parameters, namely endowing initial values to PID parameters kp, ki and kd, endowing step sizes sp, si and sd initial values according to the requirements of a PID control system on precision, and respectively endowing different weight coefficients omega 1, omega 2 and omega 3 according to the emphasis degree of a comparison example item kp, a differential item ki and an integral item kd;
Step two, according to the obtained evaluation function under the current initial PID parameters kp, ki and kd as the current evaluation function value f, changing the PID parameters kp, ki and kd to adjust the PID parameters in random steps sp, si and sd respectively to obtain new PID parameters k=kp+sp ki=ki+si and k=kd+sd, on the basis, calculating a new evaluation function value f, judging the size between f and f, and if f is less than f, executing the step three; if f is f, executing the fourth step;
Step three, taking new PID parameters k, ki as current optimal parameters, and updating the evaluation function values of the current optimal parameters, namely kp=k, ki=ki, kd=k, f=f, and recording the step-size adjustment direction sign function sgn ():
The value of sgn (-) is only 0, -1 and 1, after which Step 5 is performed
Step four, randomly selecting new PID parameters k, ki in a range of plus or minus 10 times of step length, namely:
Wherein: rand (), which is a random valued function, is able to take a random value from between two set minimum maximum values.
On the basis of new PID parameters k, ki, calculating a new evaluation function value f, comparing f with f, if f is less than f, updating the PID parameters to be new k, ki is: kp=k, ki=ki, kd=k, f=f;
step five, repeatedly executing the step two to the step four until the PID evaluation function value f is not possible to be optimized, and taking the current PID parameters kp, ki and kd as optimal PID control parameters, namely:
fopt＝f。
CN202311849283.2A 2023-12-29 2023-12-29 PID parameter directed self-adaptive intelligent setting method Pending CN118226746A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311849283.2A CN118226746A (en) 2023-12-29 2023-12-29 PID parameter directed self-adaptive intelligent setting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311849283.2A CN118226746A (en) 2023-12-29 2023-12-29 PID parameter directed self-adaptive intelligent setting method

Publications (1)

Publication Number Publication Date
CN118226746A true CN118226746A (en) 2024-06-21

Family

ID=91498591

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311849283.2A Pending CN118226746A (en) 2023-12-29 2023-12-29 PID parameter directed self-adaptive intelligent setting method

Country Status (1)

Country Link
CN (1) CN118226746A (en)

Similar Documents

Publication Publication Date Title
CN104391444B (en) A kind of based on the discrete system mononeuric PID setting method of improvement
CN108489015B (en) Air conditioning system temperature control method based on pole allocation and Pade approximation
CN109283844B (en) Smith estimation compensation control method based on pole allocation
CN105807632B (en) A kind of furnace temp controller design method based on improvement dahlin algorithm
CN110955143B (en) Compound control method for first-order inertia pure lag process
KR101849464B1 (en) Automatic Tuning Method for PID controllers
CN105388764A (en) Electro-hydraulic servo PID control method and system based on dynamic matrix feed-forward prediction
CN111752143A (en) Adjustable inertia integral control method
CN118226746A (en) PID parameter directed self-adaptive intelligent setting method
CN114371618A (en) Extended state observer compensation method based on neural network and active disturbance rejection controller
US20160282829A1 (en) Controller and control method
JP4648448B2 (en) Closed loop process control device including PID regulator
KR102415186B1 (en) Control device with adjustable control behavior
US10844799B2 (en) Plant control system
CN109695893B (en) Method, device, equipment and system for controlling oxygen concentration in boiler system
CN108757925B (en) Stepless transmission slip rate control method and system based on active disturbance rejection control
CN108919638B (en) High-order system model reduction and PID control method based on frequency domain
CN109634220B (en) Six-degree-of-freedom robot motion control method and system
WO2022034658A1 (en) Control device
CN111368982A (en) Novel time delay PD controller design method based on BM neural network
CN116520680B (en) Anti-interference PID controller setting method
JP2020047075A (en) Plant control device
JPWO2020003403A1 (en) Control device and control method
CN114415509B (en) Servo elastic active inhibition method for remote rocket projectile
JP7553368B2 (en) Control device and control method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination