CN103941590A - Process fuzzy controller system and program designing method based on mathematic models - Google Patents
Process fuzzy controller system and program designing method based on mathematic models Download PDFInfo
- Publication number
- CN103941590A CN103941590A CN201410172061.6A CN201410172061A CN103941590A CN 103941590 A CN103941590 A CN 103941590A CN 201410172061 A CN201410172061 A CN 201410172061A CN 103941590 A CN103941590 A CN 103941590A
- Authority
- CN
- China
- Prior art keywords
- fuzzy controller
- process fuzzy
- host computer
- module
- data
- 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
Links
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention discloses a process fuzzy controller system based on mathematic models. The process fuzzy controller system comprises an upper computer, a communicating module, a process fuzzy controller, a controlling device, a programming module, a program debugging module, a control index performance analyzing module, a data outputting module and a data collecting module. The upper computer and the process fuzzy controller are in communication to carry out regression analysis or program and debug programs of a BP neural network model. The process fuzzy controller can collect in-site signals, the data are sent to the upper computer through communication, the upper computer can program according to an analyzing model to achieve fuzzy control operation, then the data are sent to the process fuzzy controller through communication to be output through an output channel of the process fuzzy controller to achieve fuzzy control on the site, the programs are corrected and developed according to analysis on debugging results, and furthermore, programs of the process fuzzy controller can be programmed and transplanted to the process fuzzy controller. Thus, the control effect of the controlling device is improved, the product yield and the product quality are improved, and the production cost is reduced.
Description
Technical field
The present invention relates to the analog quantity control technology field of production equipment control device, relate to specifically a kind of process fuzzy controller system and Programming method based on mathematical model.
Background technology
At present, the existing process controller of China mainly adopts traditional PID (ratio, integration, differential) to regulate, its product mainly contains two kinds of forms, a kind of is fixed routine regulator, another kind is programmable controller, when system model is clearer and more definite, reasonably parameter just can reach satisfied control effect; When system model is indefinite, pid parameter is set very difficult, is having a strong impact on product yield and the product quality of enterprise, causes cost of products to increase, and reduces the performance of enterprises.
Fuzzy control technology is widely used in controlling in the modern times, in process control, it is controlled successful and is better than traditional PID adjustment and controls, and existing fuzzy control technology is mainly the application of the occasions such as the instrument that builds based on computer system, high-end chip and professional fuzzy logic control chip; Because the design of process fuzzy controller and application relate to complicated fuzzy reasoning computing, cause its practical application difficulty very large; So complex calculation process is for adopting the process controller of the microprocessors such as MCS51, MCS96 to realize, for having the Programmable Logic Controller PLC of analog quantity regulatory function, it realizes difficulty and programming amount is also very large, and its scan period is long causes controlling in real time weak effect.
This patent is by utilizing the identification of fuzzy control simulation data to process fuzzy controller, process of establishing fuzzy controller mathematical model, in establishment corresponding program migration process controller, thereby on the basis that does not increase control chip cost, realize the fuzzy control of control system.
Summary of the invention
The object of this invention is to provide a kind of process fuzzy controller system and Programming method based on mathematical model, by the analysis to existing control device control method, control device is proposed based on mathematical model etc. fuzzy control process control method, improve the control effect of control device, and then raising product yield and product quality, reduce production costs, improve the performance of enterprises.
For the existing problem of background technology, the present invention by the following technical solutions:
A kind of process fuzzy controller system based on mathematical model, comprise host computer, communication module, process fuzzy controller, control device, program composition module, program debug module, control index performance evaluation module, data outputting module, data acquisition module, described host computer is connected by communication module with process fuzzy controller, the input end of data acquisition module is connected with control device, data acquisition module is input to process fuzzy controller by the data input quantity of collection, process fuzzy controller receives data input quantity and is input to host computer by communication module, host computer is input to the data input quantity of reception to control index performance evaluation module, control index performance evaluation module is analyzed and is exported analytical model to data input quantity and is input to host computer, host computer is input to program composition module by the analytical model of reception, program composition module receiving and analyzing model output program to host computer, host computer is input to program debug module by programming of reception and debugs, program debug module is input to host computer by tune-up data, host computer receives tune-up data and is input to program composition module, program composition module is input to host computer by revised program, host computer is input to process fuzzy controller by revised program, process fuzzy controller is controlled output quantity through data outputting module output control device is controlled.
Described process fuzzy controller comprises microprocessor CPU chip, memory RAM chip, storer rom chip, direct supply interface module, communication interface module, A/D ALT-CH alternate channel, D/A ALT-CH alternate channel, microprocessor CPU chip and memory RAM chip, communication interface module interconnection, the output terminal of storer rom chip connects microprocessor CPU chip, direct supply interface module is connected with microprocessor CPU chip, being connected of the input of the output terminal of A/D ALT-CH alternate channel and microprocessor CPU chip, the output terminal of microprocessor CPU chip is connected with the input end of D/A ALT-CH alternate channel, the process control amount of on-site control device is converted to normal voltage or current signal by sensor, through A/D ALT-CH alternate channel, be converted to digital signal input process fuzzy controller, and the digital quantity after microprocessor CPU chip is processed, by D/A ALT-CH alternate channel, be converted to normal voltage or current signal, on-site control device is controlled, according to the deviation of controlled system and differential variation tendency design fuzzy controller thereof, and carry out modeling and simulation, perfect, on perfect process fuzzy controller basis, process fuzzy controller is carried out to regretional analysis or the simplification of BP neural network model, adopt host computer and process fuzzy controller serial ports, bus, the mode of the communications such as Industrial Ethernet is carried out regretional analysis or BP neural network model programming and debugging, process fuzzy controller gathers on-site signal, by communication, data are sent to host computer, host computer is programmed and is realized fuzzy control operation according to analytical model, result reaches process fuzzy controller through communication again, output channel output by process fuzzy controller realizes on-the-spot fuzzy control, according to the analysis of debug results, program is modified with perfect, and then compilation process fuzzy controller program migrate to process fuzzy controller.For the Process Control System that adopts ipc monitor, directly at host computer, carry out regretional analysis or the program design of BP neural network model, and then realize Intelligent Fuzzy Control.
According to foregoing invention content description, the Programming method of this invention comprises the steps: according to the deviation of controlled system and differential variation tendency design process fuzzy controller thereof, and carry out modeling and simulation, perfect, on perfect process fuzzy controller basis, process fuzzy controller is carried out to regretional analysis or the simplification of BP neural network model, adopt host computer and process fuzzy controller serial ports, bus, the mode of the communications such as Industrial Ethernet is carried out regretional analysis or BP neural network model programming and debugging, process fuzzy controller gathers on-site signal, by communication, data are sent to host computer, host computer is programmed and is realized fuzzy control operation according to analytical model, result reaches process fuzzy controller through communication again, output channel output by process fuzzy controller realizes on-the-spot fuzzy control, according to the analysis of debug results, program is modified with perfect, and then compilation process fuzzy controller program migrate to process fuzzy controller.For the Process Control System that adopts ipc monitor, directly at host computer, to carry out regretional analysis or the program design of BP neural network model, and then realize Intelligent Fuzzy Control, described method comprises following concrete steps:
Step 1: by controlling index performance evaluation module, control device is carried out to qualitative analysis, gather and output data by process fuzzy controller, utilize host computer to differentiate the closed-loop control effect regulating without traditional PI D (ratio, integration, differential);
Step 2: according to control device is carried out to qualitative analysis, for the deviation x of setting value and measured value
1, deviation differential x
2carrying out change of scale inputs as process fuzzy controller, the controlled quentity controlled variable that control device is controlled is carried out change of scale and is exported as process fuzzy controller, choose leg-of-mutton membership function input, output are carried out to fuzzy partition, formulate fuzzy control rule, the process fuzzy controller that design is controlled control device;
Step 3: gather and output data by process fuzzy controller, utilize the control effect of host computer analytic process fuzzy controller to control device;
Step 4: revise fuzzy control rule, until process fuzzy controller reaches equipment performance index request to the control effect of control device, improve process fuzzy controller;
Step 5: perfect process fuzzy controller is carried out to control device quantitative test emulation, the input of gatherer process fuzzy controller, output data;
Step 6: get design process fuzzy controller regression model and BP network model according to input, output analysis of data collected and sieve;
Step 7: the model parameter of deterministic process fuzzy controller regression model and BP network model;
Step 8: carry out respectively process fuzzy controller regression model, BP network model and process fuzzy controller prototype emulation comparison, simulated effect is misfitted, and jumps to step 6, repeats above step, until simulated effect coincide;
Step 9: according to regression model or BP network model, gather and output data by process fuzzy controller, establishment host computer fuzzy control program also moves debugging to control device;
Step 10: the control effect to control device does not reach equipment performance index request, jumps to step 5, repeats above step, until reach equipment performance index request;
Step 11: the program of compilation process fuzzy controller based on regression model or BP network model is also downloaded to process fuzzy controller;
Step 12: finish.
By process fuzzy controller image data, host computer carries out quantities conversion to the difference of setting data and process fuzzy controller image data, be resent to process fuzzy controller, by the output of process fuzzy controller data, control device is controlled, utilized host computer to differentiate the closed-loop control effect without traditional PID adjustment.
By process fuzzy controller, gather and output data, according to control device is carried out to qualitative analysis, the process fuzzy controller that utilizes host computer without the closed-loop control effect design of traditional PID adjustment, control device to be controlled.
By process fuzzy controller, gather and output data, utilize host computer establishment fuzzy control on-line monitoring program, the real-time control effect of analytic process fuzzy controller to control device.
Perfect process fuzzy controller is carried out to the quantitative off-line analysis emulation of control device, and collection, analysis and sieve are got the input of process fuzzy controller, output data, design process fuzzy controller regression model and BP network model.
Described regression model adopts
Y=a
0+ a
1x
1+ a
2x
2+ a
3x
1 2+ a
4x
2 2+ a
5x
1x
2+ a
6x
1 3+ a
7x
2 3+ a
8x
1 2x
2+ ..., x
1represent system deviation, x
2represent system deviation differential, a
i(i=1,2 ...) be model parameter, utilize the model parameter of self-compiling program deterministic process fuzzy controller regression model, BP network model adopts traditional neural network S type function, utilizes the model parameter of BP network self-learning algorithm deterministic process fuzzy controller BP network model.
Described off-line carries out process fuzzy controller regression model, BP network model and process fuzzy controller prototype emulation comparison, improves regression model and BP network model, until its simulated effect and process fuzzy controller prototype are coincide.
Described according to regression model or BP network model, by process fuzzy controller, gather and output data establishment host computer fuzzy control program by control device being moved to control with the communication of process fuzzy controller.
The program of described establishment based on regression model or BP network model is downloaded to process fuzzy controller and control device moved to control.
Beneficial effect of the present invention:
The present invention is by the analysis to existing control device control method, the fuzzy control process control method of control device based on mathematical model proposed, improve the control effect of control device, and then raising product yield and product quality, reduce production costs, improve the performance of enterprises, the present invention is by utilizing the identification of fuzzy control simulation data to fuzzy controller, set up fuzzy controller mathematical model, establishment corresponding program migrates in process fuzzy controller, thereby on the basis that does not increase control chip cost, realize the fuzzy control of control system, the present invention all has practical guided significance to the production development of the control upgrading of existing Process Control System and intelligent instrument, the industrial process that fuzzy control simulation software based on this Patent exploitation can be realized based on host computer configuration is controlled, fill up existing industrial configuration and control the blank of software in fuzzy control function.
Accompanying drawing explanation
Fig. 1 is process fuzzy controller detail programming system chart of the present invention.
Fig. 2 is the composition frame chart of process fuzzy controller of the present invention.
Fig. 3 is new type of control method process flow diagram of the present invention.
Wherein: 1, host computer, 2, process fuzzy controller, 3, control device, 4, program composition module, 5, program debug module, 6, control index performance evaluation module, 7, communication module, 8, data acquisition module, 9, data outputting module, 10, microprocessor CPU chip, 11, memory RAM chip, 12, storer rom chip, 13, direct supply interface module, 14, communication interface modules, 15, A/D ALT-CH alternate channel, 16, D/A ALT-CH alternate channel.
Embodiment
Referring to accompanying drawing, a kind of process fuzzy controller system and programmed method based on mathematical model, comprise host computer 1, communication module 7, process fuzzy controller 2, control device 3, program composition module 4, program debug module 5, control index performance evaluation module 6, data outputting module 9, data acquisition module 8, described host computer 1 is connected by communication module 7 with process fuzzy controller 2, the input end of data acquisition module 8 is connected with control device 3, data acquisition module 8 is input to process fuzzy controller 2 by the data input quantity of collection, process fuzzy controller 2 receives data input quantity and is input to host computer 1 by communication module 7, host computer 1 is input to the data input quantity of reception to control index performance evaluation module 6, 6 pairs of data input quantities of control index performance evaluation module are analyzed and are exported analytical model and are input to host computer 1, host computer 1 is input to program composition module 4 by the analytical model of reception, program composition module 4 receiving and analyzing models output program to host computer 1, host computer 1 is input to program debug module by programming of reception and debugs 5, program debug module 5 is input to host computer 1 by tune-up data, host computer 1 receives tune-up data and is input to program composition module 4, program composition module 4 is input to host computer 1 by revised program, host computer 1 is input to process fuzzy controller 2 by revised program, process fuzzy controller 2 is controlled output quantity through data outputting module 9 outputs control device 3 is controlled.
Described process fuzzy controller comprises microprocessor CPU chip 10, memory RAM chip 11, storer rom chip 12, direct supply interface module 13, communication interface module 14, A/D ALT-CH alternate channel 15, D/A ALT-CH alternate channel 16, microprocessor CPU chip 10 and memory RAM chip 11, communication interface module 14 interconnection, the output terminal of storer rom chip 12 connects microprocessor CPU chip 10, direct supply interface module 13 is connected with microprocessor CPU chip 10, being connected of the input of the output terminal of A/D ALT-CH alternate channel 15 and microprocessor CPU chip 10, the output terminal of microprocessor CPU chip 10 is connected with the input end of D/A ALT-CH alternate channel 16, the process control amount of on-site control device 3 is converted to normal voltage or current signal by sensor, through A/D ALT-CH alternate channel 15, be converted to digital signal input process fuzzy controller 2, and the digital quantity after microprocessor CPU chip 11 is processed, by D/A ALT-CH alternate channel 16, be converted to normal voltage or current signal, on-site control device 3 is controlled.
According to foregoing invention content description, the Programming method of this invention comprises the steps:
Step 1: carry out qualitative analysis by controlling 6 pairs of control device of index performance evaluation module 3, gather and output data by process fuzzy controller 2, utilize host computer 1 to differentiate the closed-loop control effect regulating without traditional PI D (ratio, integration, differential);
Step 2: according to control device 3 is carried out to qualitative analysis, for the deviation x of setting value and measured value
1, deviation differential x
2carry out change of scale as 2 inputs of process fuzzy controller, the controlled quentity controlled variable that control device 3 is controlled is carried out change of scale as 2 outputs of process fuzzy controller, choose leg-of-mutton membership function input, output are carried out to fuzzy partition, formulate fuzzy control rule, the process fuzzy controller 2 that design is controlled control device 3;
Step 3: gather and output data by process fuzzy controller 2, utilize the control effect of 2 pairs of control device 3 of host computer 1 analytic process fuzzy controller;
Step 4: revise fuzzy control rule, until the control effect of 2 pairs of control device 3 of process fuzzy controller reaches equipment performance index request, improve process fuzzy controller 2;
Step 5: perfect process fuzzy controller 2 is carried out to control device quantitative test emulation, 2 inputs of gatherer process fuzzy controller, output data;
Step 6: get design process fuzzy controller regression model and BP network model according to input, output analysis of data collected and sieve;
Step 7: the model parameter of deterministic process fuzzy controller 2 regression models and BP network model;
Step 8: carry out respectively process fuzzy controller 2 regression models, BP network model and the 2 prototype emulation comparisons of process fuzzy controller, simulated effect is misfitted, and jumps to step 6, repeats above step, until simulated effect coincide;
Step 9: according to regression model or BP network model, gather and output data by process fuzzy controller 2, establishment host computer 1 fuzzy control program also moves debugging to control device 3;
Step 10: the control effect to control device 3 does not reach equipment performance index request, jumps to step 5, repeats above step, until reach equipment performance index request;
Step 11: work out the program based on regression model or BP network model and be downloaded to process fuzzy controller;
Step 12: finish.
2 pairs of on-site signals of process fuzzy controller gather, by communication module 7, data are sent to host computer 1, host computer 1 is programmed and is realized fuzzy control operation according to analytical model, result reaches process fuzzy controller 2 through communication module 7 again, output channel output by process fuzzy controller 2 realizes the fuzzy control to field apparatus 3, according to the analysis of debug results, program is modified with perfect, and then compilation process fuzzy controller 2 programs migrate to process fuzzy controller 2.For the Process Control System that adopts host computer 1 monitoring, directly at host computer 1, carry out regretional analysis or the program design of BP neural network model, and then implementation procedure fuzzy control.At this, host computer 1 finger table household computer, notebook computer, industrial computer etc., its communication interface should have a kind of in serial ports, bus or the Industrial Ethernet with process fuzzy controller 2 abilitys to communicate, and mounting software should have operating system as WINDOWSXP, the configuration software of monitoring in real time, the SIMULINK of simulation software of control procedure etc.Control device 3 finger production run opertaing devices, temperature, pressure, flow, position probing etc. sensor is installed, output signal is normal voltage or electric current, can be received by the data acquisition channel of process fuzzy controller 2, control valve, frequency converter etc. controller is installed, normal voltage or the current signal of 2 outputs of energy receiving course fuzzy controller, accurately control control device 3.
Process fuzzy controller 2 inside are by microprocessor CPU chip 10, memory RAM chip 11, storer rom chip 12, direct supply interface module 13, communication interface module 14, A/D ALT-CH alternate channel 15, D/A ALT-CH alternate channel 16, the process control amount of on-site control device 3 is converted to normal voltage or electric current by sensor, the process control amount of on-site control device 3 is converted to normal voltage or current signal by sensor, through A/D ALT-CH alternate channel 15, be converted to digital signal and send into process fuzzy controller 2, digital quantity after microprocessor CPU chip 11 is processed, by D/A ALT-CH alternate channel 16, be converted to normal voltage or current signal, on-site control device 3 is controlled.Its communication interface interface module 14 should comprise a kind of in serial ports, bus or Industrial Ethernet.
Fig. 3 is the Programming method flow diagram of the process fuzzy controller system based on mathematical model.In Fig. 3, step 101: control device 3 is carried out to qualitative analysis, as copying machine model is
consider that be time lag
by A/D ALT-CH alternate channel 15 image data, the D/A ALT-CH alternate channel 16 output data of process fuzzy controller 2, the closed-loop control effect of utilizing host computer 1 to differentiate without traditional PID adjustment.
Step 102: according to control device 3 is carried out to qualitative analysis, for the deviation x of setting value and measured value
1, deviation differential x
2carry out change of scale as 2 inputs of process fuzzy controller, the controlled quentity controlled variable that control device 3 is controlled is carried out change of scale as 2 outputs of process fuzzy controller, choose leg-of-mutton membership function input, output are carried out to fuzzy partition, formulate fuzzy control rule, the process fuzzy controller 2 that design is controlled control device 3.
Step 103: gather and output data by process fuzzy controller 2, process fuzzy controller 2 and host computer 1 data communication, utilize host computer 1 by the control effect of 2 pairs of control device 3 of on-line monitoring process analysis process fuzzy controller.
Step 104: revise fuzzy control rule, until the control effect of 2 pairs of control device 3 of process fuzzy controller reaches equipment performance index request, improve process fuzzy controller 2.
Step 105: perfect process fuzzy controller 2 is carried out to control device 3 quantitative test emulation, as copying machine quantitative model is
gatherer process fuzzy controller input x1 and x2 and corresponding output data u thereof, obtain 1 group of data according to time series;
Step 106: according to time series input, output analysis of data collected, sieve and take out the data that model had to certain influence, design process fuzzy controller 2 regression models and BP network model; Step 107: the model parameter of deterministic process fuzzy controller 2 regression models and BP network model, utilize the model parameter of self-compiling program deterministic process fuzzy controller 2 regression models, BP network model adopts traditional neural network S type function, utilizes the model parameter of the BP network model of BP network self-learning algorithm deterministic process fuzzy controller 2.
Step 108: carry out respectively process fuzzy controller 2 regression models, BP network model and the 2 prototype emulation comparisons of process fuzzy controller, simulated effect is misfitted, and jumps to step 106, repeats above step, until simulated effect coincide;
Step 109: according to regression model or BP network model, gather and output data by process fuzzy controller 2, establishment host computer 1 fuzzy control program also moves debugging to control device 3;
Step 110: the control effect to control device 3 does not reach equipment performance index request, jumps to step 105, as copying machine quantitative model increases delay component is
deng, repeat above step, until reach equipment performance index request;
Step 111: work out the program based on regression model or BP network model and be downloaded to process fuzzy controller, the floating-point operations such as regression model only relates to and adding, subtracts, multiplication and division, the microprocessor with this kind of calculation function all can be developed the process fuzzy controller based on regression model; The floating-point operations such as BP network model only relates to and adding, subtracts, multiplication and division, index, the Programmable Logic Controller with this kind of calculation function all can be developed the process fuzzy controller based on BP network model.
Step 112: finish.
Claims (7)
1. the process fuzzy controller system based on mathematical model, it is characterized in that: comprise host computer, communication module, process fuzzy controller, control device, program composition module, program debug module, control index performance evaluation module, data outputting module, data acquisition module, described host computer is connected by communication module with process fuzzy controller, the input end of data acquisition module is connected with control device, data acquisition module is input to process fuzzy controller by the data input quantity of collection, process fuzzy controller receives data input quantity and is input to host computer by communication module, host computer is input to the data input quantity of reception to control index performance evaluation module, control index performance evaluation module is analyzed and is exported analytical model to data input quantity and is input to host computer, host computer is input to program composition module by the analytical model of reception, program composition module receiving and analyzing model output program to host computer, host computer is input to program debug module by programming of reception and debugs, program debug module is input to host computer by tune-up data, host computer receives tune-up data and is input to program composition module, program composition module is input to host computer by revised program, host computer is input to process fuzzy controller by revised program, process fuzzy controller is controlled output quantity through data outputting module output control device is controlled.
2. a kind of process fuzzy controller system based on mathematical model according to claim 1, it is characterized in that: described process fuzzy controller comprises microprocessor CPU chip, memory RAM chip, storer rom chip, direct supply interface module, communication interface module, A/D ALT-CH alternate channel, D/A ALT-CH alternate channel, microprocessor CPU chip and memory RAM chip, communication interface module interconnection, the output terminal of storer rom chip connects microprocessor CPU chip, direct supply interface module is connected with microprocessor CPU chip, being connected of the input of the output terminal of A/D ALT-CH alternate channel and microprocessor CPU chip, the output terminal of microprocessor CPU chip is connected with the input end of D/A ALT-CH alternate channel.
3. the Programming method of a kind of process fuzzy controller system based on mathematical model as claimed in claim 1, it is characterized in that: according to the deviation of controlled system and differential variation tendency design process fuzzy controller thereof, and carry out modeling and simulation, perfect, on perfect process fuzzy controller basis, process fuzzy controller is carried out to regretional analysis or the simplification of BP neural network model, the mode that adopts host computer to communicate by letter with process fuzzy controller is carried out regretional analysis or BP neural network model programming and debugging, process fuzzy controller gathers on-site signal, by communication, data are sent to host computer, host computer is programmed and is realized fuzzy control operation according to analytical model, result reaches process fuzzy controller through communication again, output channel output by process fuzzy controller realizes on-the-spot fuzzy control, according to the analysis of debug results, program is modified with perfect, and then compilation process fuzzy controller program migrate to process fuzzy controller, comprise following concrete steps:
Step 1: by controlling index performance evaluation module, control device is carried out to qualitative analysis, gather and output data by process fuzzy controller, utilize host computer to differentiate the closed-loop control effect regulating without traditional PI D (ratio, integration, differential);
Step 2: according to control device is carried out to qualitative analysis, for the deviation x of setting value and measured value
1, deviation differential x
2carrying out change of scale inputs as process fuzzy controller, the controlled quentity controlled variable that control device is controlled is carried out change of scale and is exported as process fuzzy controller, choose leg-of-mutton membership function input, output are carried out to fuzzy partition, formulate fuzzy control rule, the process fuzzy controller that design is controlled control device;
Step 3: gather and output data by process fuzzy controller, utilize the control effect of host computer analytic process fuzzy controller to control device;
Step 4: revise fuzzy control rule, until process fuzzy controller reaches equipment performance index request to the control effect of control device, improve process fuzzy controller;
Step 5: perfect process fuzzy controller is carried out to control device quantitative test emulation, the input of gatherer process fuzzy controller, output data;
Step 6: get design process fuzzy controller regression model and BP network model according to input, output analysis of data collected and sieve;
Step 7: the model parameter of deterministic process fuzzy controller regression model and BP network model;
Step 8: carry out respectively process fuzzy controller regression model, BP network model and process fuzzy controller prototype emulation comparison, simulated effect is misfitted, and jumps to step 6, repeats above step, until simulated effect coincide;
Step 9: according to regression model or BP network model, gather and output data by process fuzzy controller, establishment host computer fuzzy control program also moves debugging to control device;
Step 10: the control effect to control device does not reach equipment performance index request, jumps to step 5, repeats above step, until reach equipment performance index request;
Step 11: the program of compilation process fuzzy controller based on regression model or BP network model is also downloaded to process fuzzy controller;
Step 12: finish.
4. the Programming method of a kind of process fuzzy controller system based on mathematical model according to claim 3, it is characterized in that: by process fuzzy controller image data, host computer carries out quantities conversion to the difference of setting data and process fuzzy controller image data, be resent to process fuzzy controller, by the output of process fuzzy controller data, control device is controlled, utilized host computer to differentiate the closed-loop control effect without traditional PID adjustment.
5. the Programming method of a kind of process fuzzy controller system based on mathematical model according to claim 3, it is characterized in that: by process fuzzy controller, gather and output data, utilize host computer establishment fuzzy control on-line monitoring program, the real-time control effect of analytic process fuzzy controller to control device.
6. the Programming method of a kind of process fuzzy controller system based on mathematical model according to claim 3, it is characterized in that: perfect process fuzzy controller is carried out to the quantitative off-line analysis emulation of control device, collection, analysis and sieve are got the input of process fuzzy controller, output data, design process fuzzy controller regression model and BP network model.
7. the Programming method of a kind of process fuzzy controller system based on mathematical model according to claim 3, is characterized in that: described regression model adopts y=a
0+ a
1x
1+ a
2x
2+ a
3x
1 2+ a
4x
2 2+ a
5x
1x
2+ a
6x
1 3+ a
7x
2 3+ a
8x
1 2x
2+ ..., x
1represent system deviation, x
2represent system deviation differential, a
i(i=1,2 ...) be model parameter, utilize the model parameter of self-compiling program deterministic process fuzzy controller regression model, BP network model adopts traditional neural network S type function, utilizes the model parameter of BP network self-learning algorithm deterministic process fuzzy controller BP network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410172061.6A CN103941590A (en) | 2014-04-25 | 2014-04-25 | Process fuzzy controller system and program designing method based on mathematic models |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410172061.6A CN103941590A (en) | 2014-04-25 | 2014-04-25 | Process fuzzy controller system and program designing method based on mathematic models |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103941590A true CN103941590A (en) | 2014-07-23 |
Family
ID=51189312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410172061.6A Pending CN103941590A (en) | 2014-04-25 | 2014-04-25 | Process fuzzy controller system and program designing method based on mathematic models |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103941590A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110377472A (en) * | 2019-07-25 | 2019-10-25 | 北京中星微电子有限公司 | The method and device of positioning chip run-time error |
| CN111884716A (en) * | 2020-06-30 | 2020-11-03 | 中国南方电网有限责任公司 | Optical fiber communication system performance evaluation method based on neural network |
| CN113743572A (en) * | 2020-05-27 | 2021-12-03 | 南京大学 | Artificial neural network testing method based on fuzzy |
-
2014
- 2014-04-25 CN CN201410172061.6A patent/CN103941590A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110377472A (en) * | 2019-07-25 | 2019-10-25 | 北京中星微电子有限公司 | The method and device of positioning chip run-time error |
| CN113743572A (en) * | 2020-05-27 | 2021-12-03 | 南京大学 | Artificial neural network testing method based on fuzzy |
| CN111884716A (en) * | 2020-06-30 | 2020-11-03 | 中国南方电网有限责任公司 | Optical fiber communication system performance evaluation method based on neural network |
| CN111884716B (en) * | 2020-06-30 | 2021-10-15 | 中国南方电网有限责任公司 | Optical fiber communication system performance evaluation method based on neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN201638081U (en) | Monitoring system for whole purification treatment process of mine water | |
| US20200401154A1 (en) | Method for controlling vehicle speed for autonomous driving, electronic device, and computer-readable storage medium | |
| CN204086979U (en) | A kind of electric vehicle motor controller hardware-in―the-loop test system | |
| CN110824923A (en) | Sewage treatment control method and system based on deep learning and cloud computing | |
| CN103019104B (en) | Visual process control system configuration simulation method | |
| WO2021007871A1 (en) | Alumina production operation optimization system and method employing cloud-side collaboration | |
| CN103941590A (en) | Process fuzzy controller system and program designing method based on mathematic models | |
| CN103970121A (en) | Distributed automatic test framework system for satellites | |
| CN103345546A (en) | Speed regulator parameter identification method combined with frequency locus and particle swarm optimization | |
| CN110847111B (en) | Method for acquiring hydropower station gate scheduling parameters based on semi-physical simulation | |
| CN101598927B (en) | Control system of soda carbonization technique based on neural network and control method thereof | |
| CN104263960B (en) | On-line prediction method for replacement rate and gold mud grade in gold hydrometallurgy replacement process | |
| CN104632416A (en) | Control method for rotating speed of gas turbine | |
| CN104267600A (en) | Ladle refining furnace electrode adjustment control system and control method thereof | |
| CN116859839A (en) | Industrial control method and device based on model training | |
| CN115562141A (en) | Industrial equipment remote control method and device based on PLC | |
| CN102354158A (en) | Separation device for online measurement of dynamic error based on Bayesian network and method thereof | |
| CN102555064A (en) | Proportioning system and method for concrete, and upper computer | |
| CN104571086A (en) | Temperature controller simulation testing method based on transfer function | |
| CN115826648A (en) | Heating furnace trusteeship temperature control method and system based on cloud edge cooperation | |
| CN106020172B (en) | A kind of real-time sine sweep driving method of rudder servo based on FPGA platform and device | |
| CN201933134U (en) | DSP (digital signal processor) control system of I type electric reduction furnace units of reduction-distillation workshop | |
| CN101710233A (en) | Design method of motion controller based on RCP | |
| CN201662724U (en) | Liquid level fuzzy control device of oil refining fractionating tower | |
| CN104467500A (en) | Model-based design device and method for double closed-loop three-level inverter system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140723 |
|
| WD01 | Invention patent application deemed withdrawn after publication |