CN102707617A - Method for realizing fuzzy PID (Proportion Integration Differentiation) ActiveX control - Google Patents

Abstract

The invention discloses a method for realizing a fuzzy PID (Proportion Integration Differentiation) ActiveX control. The method comprises the following steps of: 1, setting a variable, a membership function of the variable, a fuzzy reasoning method and a de-fuzziness method; 2, setting a reasoning rule, determining the fuzzy values of e and ec to obtain accurate values kp, ki and kd after de-fuzzing a PID output variable, and establishing a fuzzy control decision polling list; and 3, compiling a fuzzy PID algorithm through VB (Visual Basic), and creating an ActiveX control. Due to the adoption of the control realizing method disclosed by the invention, the logical operation program of a fuzzy reasoning process is not required to be compiled any more during development, the output values deltaKP, deltaKI and deltakd of the fuzzy adjustment amount of a PID parameter which corresponds to a basic discourse domain are not required to be calculated according to a weighted average method, a program structure is simplified, the program executing period is shortened greatly, and the response speed of a system is increased.

Description

A kind of implementation method of fuzzy algorithm ActiveX control

Technical field

The present invention relates to a kind of implementation method of fuzzy algorithm Active X control, belong to the realization and the applied technical field of advanced control method.

Background technology

Active X control triggers the control operation through incident (clicking control, keyboard input etc.); It is the employed IO interface of control container that method and attribute then provide control; And utilize the agreement support of configuration software to Windows standard A ctiveX control; The implementation algorithm module is connected with the data at configuration interface, thereby calls control algolithm easily.

The fuzzy algorithm is as a kind of advanced person's intelligent control algorithm; Compare with conventional PID control and to have the mathematical model that need not to set up controlled device; The time lag of controlled device, non-linear and time variation had certain advantages such as adaptive faculty; Simultaneously noise is also had stronger inhibition ability, promptly robustness is better, begins to take shape in the application of various fields such as industrial, civilian.At present, it is general with reference to program flow diagram shown in Figure 1 to write the fuzzy algorithm routine, comprises that mainly input variable quantizes, the fuzzy control table computing is synthetic and controlled quentity controlled variable output is calculated.Among Fig. 1, e is the deviation of actual value and setting value, and ec is the variable quantity of deviation, and Escope is that the basic domain of e is a deviation range, and Ecscope is that the basic domain of ec is the deviation variation range, and E is the universe of fuzzy sets of e, and EC is the universe of fuzzy sets of ec, △ K _P, △ K _I, △ K _DBe respectively the adjustment amount of P, I, D parameter.

At first, according to controlled device, preset sequence sampling period, E, EC, Escope, Ecscope, pid parameter initial value, goal-setting value, and △ K _P, △ K _I, △ K _DBasic domain K _PScope, K _IScope, K _DScope is the actual setting range of controlled quentity controlled variable, calculates e and ec and calculated value is limited in the basic domain scope separately;

Then, with the calculated value obfuscation of e and ec and write fuzzy control rule, obtain the fuzzy value of pid parameter adjustment amount according to E and EC through fuzzy reasoning;

At last, adopt method of weighted mean with pid parameter adjustment amount ambiguity solution to △ K _P, △ K _I, △ K _DBasic domain in go, form new pid parameter on the pid parameter value that the last program scanning computation of Period that is added to goes out, and use the increment type PID algorithm formula and calculate the final controlled quentity controlled variable that outputs to controlled device.

The calculating that the computing of frame of broken lines part fuzzy control table is synthesized and exported controlled quentity controlled variable is the emphasis and the difficult point of algorithm routine exploitation.Each sampling and the controlled quentity controlled variable that provides through FUZZY ALGORITHMS FOR CONTROL, directly controlling object must be transformed into it in receptible basic domain of controlling object and go, and computing formula is following:

ΔK _P＝C _P·kp，ΔK _I＝C _I·ki，ΔK _D＝C _D·kd

Wherein, $C_P=\frac{\left|K_P\mathrm{Scope}\right|}{\mathrm{Kp}},$ $C_I=\frac{\left|K_I\mathrm{Scope}\right|}{\mathrm{Kp}},$ $C_D=\frac{\left|K_D\mathrm{Scope}\right|}{\mathrm{Kd}},$ Δ K _P, Δ K _I, Δ K _DAdjustment amount for pid parameter; C _P, C _I, C _DScale-up factor for each adjustment amount of pid parameter; Kp, ki, kd are Δ K _P, Δ K _I, Δ K _DThe fuzzy value that in universe of fuzzy sets, obtains by the fuzzy control rule judgement.At last, according to the pid parameter adjustment amount that obtains, obtain final controlled quentity controlled variable according to increment type PID algorithm.

Therefore, when the control invokes control algolithm, must cause the COMPUTER CALCULATION amount big, it is heavier to load, and causes the hyperharmonic vibration of system easily, thus influence the fuzzy algorithm to controlled device online, control effect and control accuracy in real time.

Summary of the invention

The objective of the invention is to quantize in order to solve input variable, the fuzzy control table computing is synthetic and controlled quentity controlled variable to export the Development Engineering of inclusive fuzzy algorithm routine complicated; The problem that calculated amount is big; Utilize the gui tool calculating computing of fuzzy control table is synthetic and the output controlled quentity controlled variable in Matlab fuzzy control tool box to simplify, proposed a kind of simplification implementation method of fuzzy algorithm Active X control.

A kind of implementation method of fuzzy algorithm Active X control comprises following step:

Step 1): membership function, fuzzy reasoning method and reverse gelatinizing method that variable, variable are set;

Step 2): inference rule is set, confirms the fuzzy value of e and ec, obtain exact value kp, ki, kd after the gelatinization of PID output variable reverse, set up fuzzy control decision-making question blank;

Step 3): through VB, write the fuzzy algorithm, and create Active X control;

The invention has the advantages that: the logical operation program that the control exploitation time no longer need be write fuzzy reasoning process, need be according to the fuzzy adjustment amount of calculated with weighted average method pid parameter output valve Δ K corresponding to basic domain yet _P, Δ K _I, Δ K _D, not only simplified program structure, the program implementation cycle also shortens greatly, helps accelerating the response speed of system.

Description of drawings

Fig. 1 is the program flow diagram of the fuzzy algorithm in the background technology of the present invention;

Fig. 2 is a method flow diagram of the present invention;

Fig. 3 is the fuzzy inference system interface of fuzzy of the present invention;

Fig. 4 is the inference rule interface of fuzzy of the present invention;

Fig. 5 is the fuzzy inference rule viewer interface of fuzzy of the present invention;

Fig. 6 is the fuzzy controller Active X control development interface that the present invention is based on the VB environment.

Embodiment

To combine accompanying drawing and embodiment that the present invention is done further detailed description below.

The present invention adopts the control of Matlab gui tool to simplify implementation method, utilizes the gui tool in Matlab fuzzy control tool box, and the calculation procedure off-line computing of fuzzy control table is synthetic and the output controlled quentity controlled variable is accomplished, and only needs in program, to set up Δ K _P, Δ K _I, Δ K _DThe output question blank according to the situation of difference moment fuzzy input variable E and EC, is found corresponding Δ K _P, Δ K _I, Δ K _DValue on the pid parameter that the last sampling period that is added to calculates, obtained new pid parameter, and then calculated to constantly different according to the increment type PID algorithm formula, and the algorithm control need output to the controlled quentity controlled variable of controlled device.

The present invention is a kind of implementation method of fuzzy algorithm Active X control, and flow process is as shown in Figure 2, comprises following step:

Step 1): membership function, fuzzy reasoning method and reverse gelatinizing method that variable, variable are set;

In Matlab command window input " Fuzzy ", eject the fuzzy inference system editor window.In its Edit menu, add input variable e, ec and output variable kp, ki, kd, and be that input variable, output variable are selected membership function.E is the deviation of actual value and setting value, and ec is the variable quantity of deviation, and kp, ki, kd are Δ K _P, Δ K _I, Δ K _DThe fuzzy value that in universe of fuzzy sets, obtains by the fuzzy control rule judgement, △ K _P, △ K _I, △ K _DBe respectively the adjustment amount of P, I, D parameter, " And " of fuzzy reasoning elects minimum method as, and " Or " elects maximum method as, and the reverse gelatinizing method is elected method of weighted mean as, and is as shown in Figure 3.

Step 2): inference rule is set, confirms the fuzzy value of e and ec, obtain exact value kp, ki, kd after the gelatinization of PID output variable reverse, set up fuzzy control decision-making question blank;

In the Edit menu, select Rules, inference rule is set, as shown in Figure 4.Select the Rules order in the View menu; In " Rule Viewer " window that ejects; As shown in Figure 5, according to characteristic and control requirement, the fuzzy value of the variable quantity ec of input deviation e and deviation of controlled device; Like [O O], then can obtain exact value kp, ki, kd after the gelatinization of PID output variable reverse.

Through step 1) and step 2); Utilize the gui tool calculating of off-line completion controlled quentity controlled variable output automatically of Matlab; Behind input E, the EC, obtain the exact value after its corresponding PID output variable reverse gelatinization, according to different input E, EC; Obtain its corresponding kp, ki, kd, generate fuzzy control decision-making question blank.

Step 3): through VB, write the fuzzy algorithm, and create Active X control;

Get into the VB programmed environment, newly-built " Active X control " engineering 1, Visual Basic will provide a window for new control; Create the background of Active X control; On background, create the control interface of fuzzy algorithmic controller, as shown in Figure 6, on the control interface, text box is set; Wherein, " ratio initial value ", " initial value for integral ", " differential initial value " are the PID initial parameter value of fuzzy algorithm; " setting value " is the control target of controlled device; " measured value " is the actual value of controlled physical quantity; " the basic domain of error ", " error changes basic domain " are respectively the actual change scope of fuzzy algorithm input e and ec; " the basic domain of ratio increment ", " the basic domain of integration increment ", " the basic domain of DG differential gain " are respectively Active X control output Δ K _P, Δ K _I, Δ K _DThe actual change scope; " ratio final value ", " integration final value ", " differential final value " are respectively the pid parameter value of the final output of fuzzy algorithm; Clock is for calling the timer of fuzzy algorithm.Employing cycle mode is regularly called the fuzzy algorithm.The timer time spacing parameter that calls in the control and the sampling time of control should be consistent.The time interval is too short, and the COMPUTER CALCULATION amount is big, and it is heavier to load, and causes the hyperharmonic vibration of system easily; If the time interval is oversize, controls the adjusting time of slow system and can extend.

For each text box of creating adds attribute, incident and method, select " engineering " in the VB menu bar, " interpolation user control " then, " VB Active X control interface guide " then is for control adds attribute, incident and method.Because the controlling object that algorithm is used in the engineering reality is different; " the ratio initial value " of institute's placement algorithm, " initial value for integral ", " differential initial value ", " setting value ", " the basic domain of error ", " error changes basic domain ", " the basic domain of ratio increment ", " the basic domain of integration increment ", " the basic domain of DG differential gain " are all different, and these parameters need the user to be provided with according to the actual conditions of oneself; And " measured value ", " ratio final value ", " integration final value ", " differential final value " are the real output value that control obtains according to the fuzzy algorithm; Therefore; These parameters all are made as the attribute of control, and add reading/assignment code of property value, give attribute assignment and show through text box and detect the output valve of putting with controlled quentity controlled variable; And the fuzzy algorithm is made as the method for control, so that call.Obtain " measured value ", " ratio final value ", " integration final value ", " differential final value " according to the fuzzy algorithm, last controlled amount is exported to controlled device with controlled quentity controlled variable, and controlled device is controlled.

Described fuzzy algorithm is specially:

The user is according to actual conditions setting " ratio initial value ", " initial value for integral ", " differential initial value ", " setting value ", " the basic domain of error ", " error changes basic domain ", " the basic domain of ratio increment ", " the basic domain of integration increment ", " the basic domain of DG differential gain "; Pass through above-mentioned parameter; Obtain the value of e and ec; Fuzzy control decision-making question blank is inquired about, obtained its corresponding kp, ki, kd value.

Step 4): registration Active X control.

Carry out Windows system " beginning operation " order, this Active X control is registered in input in the dialog box that occurs " regsvr32 < path, control place>fuzzy algorithm control .ocx ".

Characteristic and control requirement according to controlled device; In gui tool, input variable e (error of actual value and setting value), ec (error rate of actual value and setting value), output variable P (ratio), I (integration), D (differential) are carried out the fuzzy quantization grade classification; Select the membership function of each variable, and to the corresponding fuzzy control rule of each fuzzy quantization grade.Like this, gui tool can be accomplished the evaluation work of controlled quentity controlled variable output automatically, and the corresponding output of P, I, D parameter was accurately measured when the user can inquire about different e and ec input in the fuzzy inference rule viewer.Therefore; Only need in the VB development sequence of Active X control, to set up corresponding fuzzy control decision-making question blank; When certain e after program satisfies obfuscation simultaneously and ec numerical value; Look into the fuzzy control decision table, can obtain P, the I of the corresponding moment, corresponding state, the output of three parameters of D is accurately measured, and calculates the control output quantity to controlled device according to the increment type PID formula then.

So far, the exploitation of the Active X control of fuzzy algorithm finishes.Utilize the corresponding moment of gui tool calculated off-line in Matlab fuzzy control tool box, the PID output of corresponding state accurately to measure; Simplified the execution amount of computing machine in procedure quantity and the operational process in the control performance history; Not only simplified program structure; The program implementation cycle also shortens greatly, helps accelerating the response speed of system, improves the control accuracy and control effect of controlled device.

At present, configuration software basically all supports the embedding of common controls to use, and therefore control algolithm is designed to the form of control, in configuration software, embeds and uses, and its data transmission is the inner data communication of program, and real-time is good.Control only provides the interface of algorithm control, and in the picture of carrying out control strategy, the user need only call these algorithm controls, and the input necessary parameter is connected with data can accomplish corresponding control strategies quickly and easily.

Claims (2)

1. the implementation method of a fuzzy algorithm Active X control is characterized in that, comprises following step:

Step 1): membership function, fuzzy reasoning method and reverse gelatinizing method that variable, variable are set;

In Matlab command window input " Fuzzy ", eject the fuzzy inference system editor window; In its Edit menu, add input variable e, ec and output variable kp, ki, kd, and be that input variable, output variable are selected membership function; E is the deviation of actual value and setting value, and ec is the variable quantity of deviation, and kp, ki, kd are Δ K _P, Δ K _I, Δ K _DThe fuzzy value that in universe of fuzzy sets, obtains by the fuzzy control rule judgement, △ K _P, △ K _I, △ K _DBe respectively the adjustment amount of P, I, D parameter, " And " of fuzzy reasoning elects minimum method as, and " Or " elects maximum method as, and the reverse gelatinizing method is elected method of weighted mean as;

Step 2): inference rule is set, confirms the fuzzy value of e and ec, obtain exact value kp, ki, kd after the gelatinization of PID output variable reverse, set up fuzzy control decision-making question blank;

In the Edit menu, select Rules, inference rule is set; Select the Rules order in the View menu; In " Rule Viewer " window that ejects; Based on the characteristic of controlled device and control requirement, the fuzzy value of the variable quantity ec of input deviation e and deviation obtains exact value kp, ki, kd after the gelatinization of PID output variable reverse;

Through step 1) and step 2); Utilize the gui tool calculating of off-line completion controlled quentity controlled variable output automatically of Matlab; Behind input E, the EC, obtain the exact value after its corresponding PID output variable reverse gelatinization, based on different input E, EC; Obtain its corresponding kp, ki, kd, generate fuzzy control decision-making question blank;

Step 3): through VB, write the fuzzy algorithm, and create Active X control;

Get into the VB programmed environment; Newly-built " Active X control " engineering, Visual Basic will provide a window for new control, create the background of Active X control; On background, create the control interface of fuzzy algorithmic controller; On the control interface, text box is set, wherein, " ratio initial value ", " initial value for integral ", " differential initial value " are the PID initial parameter value of fuzzy algorithm; " setting value " is the control target of controlled device; " measured value " is the actual value of controlled physical quantity; " the basic domain of error ", " error changes basic domain " are respectively the actual change scope of fuzzy algorithm input e and ec; " the basic domain of ratio increment ", " the basic domain of integration increment ", " the basic domain of DG differential gain " are respectively Active X control output Δ K _P, Δ K _I, Δ K _DThe actual change scope; " ratio final value ", " integration final value ", " differential final value " are respectively the pid parameter value of the final output of fuzzy algorithm; Clock is for calling the timer of fuzzy algorithm; Employing cycle mode is regularly called the fuzzy algorithm; The timer time spacing parameter that calls in the control and the sampling time of control should be consistent;

For each text box of creating adds attribute, incident and method, select " engineering " in the VB menu bar, " interpolation user control " then, " VB Active X control interface guide " then is for control adds attribute, incident and method; Because the controlling object that algorithm is used in the engineering reality is different, the user is provided with " ratio initial value ", " initial value for integral ", " differential initial value ", " setting value ", " the basic domain of error ", " error changes basic domain ", " the basic domain of ratio increment ", " the basic domain of integration increment ", " the basic domain of DG differential gain " of algorithm according to oneself actual conditions; And " measured value ", " ratio final value ", " integration final value ", " differential final value " are the real output value that control obtains according to the fuzzy algorithm; Therefore; These parameters all are made as the attribute of control, and add reading/assignment code of property value, give attribute assignment and show through text box and detect the output valve of putting with controlled quentity controlled variable; And the fuzzy algorithm is made as the method for control, so that call; Obtain " measured value ", " ratio final value ", " integration final value ", " differential final value " according to the fuzzy algorithm, last controlled amount is exported to controlled device with controlled quentity controlled variable, and controlled device is controlled;

Described fuzzy algorithm is specially:

" ratio initial value ", " initial value for integral ", " differential initial value ", " setting value ", " the basic domain of error ", " error changes basic domain ", " the basic domain of ratio increment ", " the basic domain of integration increment ", " the basic domain of DG differential gain " according to user's setting; Obtain the value of e and ec; Fuzzy control decision-making question blank is inquired about, obtained its corresponding kp, ki, kd value.

2. the implementation method of a kind of fuzzy algorithm Active X control according to claim 1 is characterized in that, also comprises step 4): registration Active X control:

Carry out Windows system " beginning operation " order, this Active X control is registered in input in the dialog box that occurs " regsvr32 < path, control place>fuzzy algorithm control .ocx ".

Images