CN105259756B - Power plant's control loop identification Method and system - Google Patents
Power plant's control loop identification Method and system Download PDFInfo
- Publication number
- CN105259756B CN105259756B CN201510686150.7A CN201510686150A CN105259756B CN 105259756 B CN105259756 B CN 105259756B CN 201510686150 A CN201510686150 A CN 201510686150A CN 105259756 B CN105259756 B CN 105259756B
- Authority
- CN
- China
- Prior art keywords
- data
- control loop
- power plant
- model
- input 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.)
- Active
Links
Abstract
The present invention, which provides a kind of power plant's control loop identification Method and system, this method, to be included：Obtain the operation data of Power Plant；Based on revolving door algorithm operation data are carried out with the key point that quick Linear Quasi merging determines to reflect plots changes after being fitted on curve；Successively calculate four adjacent key points between data segment slope and respectively compared with corresponding threshold value with judge whether occur step or whether there are slope be more than the first preset value data segment；If so, least square method is then used to carry out Model Distinguish with Revised genetic algorithum to obtain the transfer function model of the control loop.This method provides modeling tool for the optimization design of the control system of Power Plant with debugging, without obtaining subtangent, the intermediate quantities such as 2 points again, simplifies the identification process of power plant's control loop model, has practicality.
Description
Technical field
The present invention relates to power plant's technical field, more particularly to a kind of power plant's control loop identification Method and system.
Background technology
In recent decades, process control technology develops rapidly, and conventional control process can be realized in industrial control computer
Control under carry out automatically, and as modern control system progressively tends to maximize, complicates, have in system substantial amounts of variable and
Circuit needs to monitor.For power industry, as unit capacity is continuously increased, the automatic control system structure of large-sized unit is more
Add complexity, the safety and economic operation of unit is even more important, and the requirement to automation is higher and higher.In order to which control mesh is better achieved
Mark the dynamic spy for it is necessary to grasp the dynamic characteristic of various process equipments and automatic control equipment, only having grasped their each several parts
Property, it just can guarantee that the reliability and automatization level of higher.In traditional experiment modeling, using open loop upset test data and correlation
Discrimination method asks for controlled device transmission function, although realizing simply, exists and modeling conditions are required to build when height, system complex
The shortcomings of mould cycle is long, model is poor to operating mode and time availability.So Model Distinguish of the research based on modern optimization technology is built
Mould method has great significance in the hope of improving Control platform.
Using conventional method, the requirement usually to field condition and test parameter is calculated such as tangential method, two-point method, area-method
Harsher, practicality is poor.In fact, System Discrimination can not necessarily carry out under open loop situations.Because some systems cannot
The backfeed loop of cutting-off process easily, it is out of control to be otherwise likely to result in process, seriously affects production, and this requires identification must be
Carried out under closed loop states.Scattered control system (DCS) has been achieved with the digitlization of production process with supervisory information system (SIS),
Save magnanimity operation data.So how to obtain machine unit characteristic information from these data, volume is not being increased to control system
In the case of outer disturbance, the unit model of certain precision is obtained, establishes the mathematical model of controlled device, it has also become System Discrimination is led
One of hot issue of domain research and development.
The content of the invention
Based on this, it is necessary to a kind of simple and practical power plant's control loop identification Method and system of offer.
A kind of power plant's control loop identification Method, including：
Obtain the operation data of Power Plant；
Based on revolving door algorithm operation data are carried out with quick Linear Quasi merging to determine to reflect curve on curve after being fitted
The key point of variation tendency；
Successively calculate four adjacent key points between data segment slope and respectively compared with corresponding threshold value with
Judge whether that step occurs or whether there are the data segment that slope is more than the first preset value；
If so, least square method is then used to carry out Model Distinguish with Revised genetic algorithum to obtain the biography of the control loop
Delivery function model.
In wherein a kind of embodiment, further include：
When judging step occurs or there are slope and be more than the data segment of the first preset value, corresponding four adjacent passes are obtained
The input data and output data of data segment between key point, and the reliability of input data and output data is examined, if examining
Reliably, then the step of Model Distinguish is performed.
In wherein a kind of embodiment, the step of reliability for examining input data and output data, includes：
Calculate the related coefficient of input data and output data and judge whether related coefficient is more than the second preset value；
If so, the first amplitude spectrum of input data and output data is then calculated respectively, and respectively to input data and output
The amplitude spectrum of data is normalized to obtain the second amplitude spectrum of input data and output data；
Compare the similarity of the second amplitude spectrum of input data and output data；
If similarity is within a preset range, it is determined that input data and output data are reliable.
In wherein a kind of embodiment, after the step of determining key point, between any four key point is calculated
Three data segments slope the step of before, further include step：False key point is rejected according to plots changes.
In wherein a kind of embodiment, least square method is used to carry out Model Distinguish with Revised genetic algorithum to obtain
The step of transfer function model of the control loop, includes：
Obtain the transmission function structure of the control loop of input；Transmission function includes first order inertial loop structure, single order is used to
Property delay link structure and second-order inertia delay link；
The model parameter of control loop is recognized using least square method of recursion；
The result obtained by least square method of recursion is optimized using Revised genetic algorithum, is exported by computation model
White noise verification is carried out to judge the attribute of control loop model with the residual error of the output of process and to residual sequence.
A kind of power plant's control loop model identification system, including：
Acquisition module, for obtaining the operation data of Power Plant；
Fitting module, it is bent after being fitted for being determined based on revolving door algorithm to the quick Linear Quasi merging of operation data progress
Reflect the key point of plots changes on line；
Judgment module, for calculate successively between four adjacent key points the slope of data segment and respectively with corresponding thresholding
Value is compared to judge whether that step occurs or whether there are data segment of the slope more than the first preset value；
Module is recognized, for when the judging result of judgment module is is, being calculated using least square method and improved heredity
Method carries out Model Distinguish to obtain the transfer function model of the control loop.
In wherein a kind of embodiment, further include：
Inspection module, for when the judging result of judgment module is is, acquisition to be corresponded between four adjacent key points
The input data and output data of data segment, and examine the reliability of input data and output data.
In wherein a kind of embodiment, inspection module includes：
First judging unit, for calculating the related coefficient of input data and output data and judging whether related coefficient is big
In the second preset value；
Processing unit, for when the judging result of the first judging unit is is, calculating input data and output number respectively
According to first amplitude compose, and respectively the amplitude spectrum of input data and output data is normalized to obtain input data and
Second amplitude spectrum of output data；
Comparing unit, the similarity of the second amplitude spectrum for comparing input data and output data；
Reliability determination unit, for comparing unit comparative result for similarity within a preset range when, determine defeated
Enter data and output data is reliable.
In wherein a kind of embodiment, further include：Module is rejected, it is false crucial for being rejected according to plots changes
Point.
In wherein a kind of embodiment, identification module includes：
Input unit, the transmission function structure of the control loop for obtaining input；Transmission function includes one order inertia ring
Section structure, one order inertia delay link structure and second-order inertia delay link；
Identification unit, for being recognized using least square method of recursion to the model parameter of control loop；
Optimize unit, for being optimized using Revised genetic algorithum to the result obtained by least square method of recursion, lead to
Cross the residual error of computation model output and the output of process and white noise verification is carried out to residual sequence to judge control loop model
Attribute.
Power plant's control loop identification Method, by obtaining the history and real-time running data of Power Plant, is based on
Revolving door algorithm to operation data be fitted to obtain curve, by the key point extracted on curve judge whether occur step or
It is being more than the first preset value in the slope for having data segment, if so, then carrying out mould using least square method and Revised genetic algorithum
Type identification adjusts to be controlled circuit pid parameter to obtain the transfer function model of the control loop, is Power Plant
The optimization design of control system provides modeling tool with debugging, without obtaining subtangent, the intermediate quantities such as 2 points again, simplifies electricity
The identification process of factory's control loop model, has practicality.
Brief description of the drawings
Fig. 1 is a kind of flow chart of power plant's control loop identification Method of embodiment；
Fig. 2 is a kind of step schematic diagram of embodiment；
Fig. 3 is the flow chart of power plant's control loop identification Method of another embodiment；
Fig. 4 is a kind of flow chart of the method for the judgement data reliability of embodiment；
Fig. 5 is a kind of flow chart of the identification Method of embodiment；
Fig. 6 is a kind of module diagram of power plant's control loop model identification system of embodiment；
Fig. 7 is the module diagram of power plant's control loop model identification system of another embodiment.
Embodiment
As shown in Figure 1, a kind of power plant's control loop identification Method, comprises the following steps：
S10：Obtain the operation data of Power Plant.
The operation data of Power Plant include input data and output data, can pass through scattered control system (DCS) and prison
Information system (SIS) is controlled to obtain.
S30：Based on revolving door algorithm operation data are carried out with quick Linear Quasi merging to determine on curve to reflect after being fitted
The key point of plots changes.
Revolving door algorithm is used for the mass data that the operation to Power Plant produces and carries out quick linear fit.Key point
To reflect the point of plots changes on curve after fitting, for example, changing the flex point in curve direction up or down.Specific
In embodiment, since the data variance of each control loop of power plant is larger, fluctuation range is inconsistent, and each control loop
Requirement to disturbance is also different.Therefore, after reading in input data and output data from field data file, difference can be directed to
Data under the different load in circuit can set corresponding compression accuracy, accurately to find out the key of reflection plots changes
Point.
S50：Successively calculate four adjacent key points between data segment slope and compared respectively with corresponding threshold value
Compared with judge whether occur step or whether there are slope be more than the first preset value data segment.
Key point has reacted the trend of curve.Curve between two key points can approximately regard straight line as, count successively
Calculate the slope of three data segments between four adjacent key points.
As shown in Fig. 2, the present invention is by calculating three slopes：The sampled value at the i-th -2 moment and the sampled value at the i-th -1 moment
Between between slope, the sampled value and the sampled value at the i-th moment at the i-th -1 moment when slope, the sampled value and i+1 at the i-th moment
Slope between the sampled value at quarter, by the slope of calculating compared with corresponding threshold value with judge whether occur step or whether
It there are the data segment that slope is more than the first preset value.
Step or the slope I -1 section as shown in Figure 2 more than the data segment of the first preset value occurs, which usually produces
Raw step signal, is step data segment, the last data section of the data segment and the slope of latter data section are more steadily steady
Data segment.
If so, then perform step S70：Least square method is used to carry out Model Distinguish with Revised genetic algorithum to obtain this
The transfer function model of control loop.
Power plant's control loop identification Method, by obtaining the history and real-time running data of Power Plant, is based on
Revolving door algorithm to operation data be fitted to obtain curve, by the key point extracted on curve judge whether occur step or
It is being more than the first preset value in the slope for having data segment, if so, then carrying out mould using least square method and Revised genetic algorithum
Type identification adjusts to be controlled circuit pid parameter to obtain the transfer function model of the control loop, is Power Plant
The optimization design of control system provides modeling tool with debugging, without obtaining subtangent, the intermediate quantities such as 2 points again, simplifies electricity
The identification process of factory's control loop model, has practicality.
In another embodiment, as shown in figure 3, being more than the judging to occur step or there are the slope of data segment
During one preset value, step S60 is further included：Obtain the input data and output number of the data segment between corresponding four adjacent key points
According to, and the reliability of input data and output data is examined, to ensure the identifiability of data.
If detection is reliable, step S70 is performed.
When the slope that step occurs for judgement or there are data segment is more than the first preset value, corresponding four of acquisition is adjacent
The input data and output data of data segment between key point, including：Input data and the output of the data segment of step occurs
Data, the input data and output data and input data and output number that balancing segment after step occurs that steady section before step occurs
According to.
Specifically, as shown in figure 4, step S60 is specifically included：
S601：Calculate the related coefficient of input data and output data and judge whether related coefficient is more than second and presets
Value.In a particular embodiment, using 0.9 as second preset value, when the related coefficient of input data and output data is more than
When 0.9, then step is performed：
S602：The first amplitude spectrum of input data and output data is calculated respectively, and respectively to input data and output number
According to amplitude spectrum be normalized to obtain the second amplitude spectrum of input data and output data.
S603：Compare the similarity of the second amplitude spectrum of input data and output data.
If similarity is within a preset range, step S604 is performed：Determine that input data and output data are reliable.
For stable Linear Time-Invariant System, the output steady-state component produced by harmonic wave input is still and input same frequency
Harmonic function, and the change of amplitude and phase is the function of frequency, and related to system mathematic model.So handle inputs, is defeated
Go out in data the harmonic wave for containing identical frequency as one of foundation for choosing data.Secondly, by the analysis of a large amount of field datas,
Input signal mainly includes the high-frequency harmonic of the larger low-frequency harmonics of amplitude and amplitude very little, the frequency of input signal low-frequency harmonics
For scope generally all in the lower passband of system object, low frequency signal can be by, and inputs, exports low frequency signal in signal
Amplitude Ration is a constant.High-frequency signal is then filtered after by system object.Analyzed more than, frequency in lower passband
The ratio between interior input, output signal amplitude are approximately a constant as another foundation for choosing data.
In another embodiment, as shown in figure 3, before step S30 and step S50, step is further included：
S40：False key point is rejected according to plots changes.
Virtual key point is usually to influence little key point to the variation tendency of curve.For example, pass through more a certain pass
The previous line segment of key point and latter line segment slope, if the slope of two lines section is within a preset range, can assert that the key point is
The variation tendency of curve influences less, which is false key point.After rejecting the key point, before and after the key point and the key point
Trend term between key point merges, and the data segment between adjacent key point is a trend term.
In another particular embodiment of the invention, as shown in figure 5, step S70 is specifically included：
S701：Obtain the transmission function structure of the control loop of input.
The transmission function structure of input different circuit, including first order inertial loop structure, single order are selected according to priori
Inertial delay link structure and second-order inertia delay link.
S702：The model parameter of control loop is recognized using least square method of recursion.
S703：The result obtained by least square method of recursion is optimized using Revised genetic algorithum, by calculating mould
Type exports with the residual error of the output of process and carries out white noise verification to residual sequence to judge the attribute of control loop model.
Due to reasons such as disturbance, noises, identification result may not be optimal, therefore, using Revised genetic algorithum pair
Result obtained by least square method of recursion optimizes.Also, by computation model output with the residual error of the output of process and to residual
Difference sequence carries out white noise verification to judge the quality of control loop model.
Power plant's control loop identification Method, by obtaining the history and real-time running data of Power Plant, is based on
Revolving door algorithm to operation data be fitted to obtain curve, by the key point extracted on curve judge whether occur step or
It is being more than the first preset value in the slope for having data segment, and is obtaining and the data segment of step or slope more than the first preset value occurs,
Examine the reliability of input data and output data, if examine it is reliable, using least square method and Revised genetic algorithum into
Row Model Distinguish is adjusted with obtaining the transfer function model of the control loop to be controlled circuit pid parameter.The power plant is controlled
The identification Method in circuit processed, by testing to the input data and output data that produce step signal, so that it is guaranteed that
The identifiability of data, to improve the accuracy of Model Distinguish.
The present invention also provides a kind of power plant's control loop model identification system, as shown in fig. 6, including：
Acquisition module 10, for obtaining the operation data of Power Plant.
The operation data of Power Plant include input data and output data, can pass through scattered control system (DCS) and prison
Information system (SIS) is controlled to obtain.
Fitting module 20, merges definite intend for carrying out quick Linear Quasi to the operation data based on revolving door algorithm
Reflect the key point of plots changes after conjunction on curve.
Revolving door algorithm is used for the mass data that the operation to Power Plant produces and carries out quick linear fit.Key point
To reflect the point of plots changes on curve after fitting, for example, changing the flex point in curve direction up or down.Specific
In embodiment, since the data variance of each control loop of power plant is larger, fluctuation range is inconsistent, and each control loop
Requirement to disturbance is also different.Therefore, after reading in input data and output data from field data file, difference can be directed to
Data under the different load in circuit can set corresponding compression accuracy, accurately to find out the key of reflection plots changes
Point.
Judgment module 30, for calculate successively between four adjacent key points the slope of data segment and respectively with it is corresponding
Threshold value be compared with judge whether occur step or whether there are slope be more than the first preset value data segment.
Key point has reacted the trend of curve.Curve between two key points can approximately regard straight line as, count successively
Calculate the slope of three data segments between four adjacent key points.
As shown in Fig. 2, the present invention is by calculating three slopes：The sampled value at the i-th -2 moment and the sampled value at the i-th -1 moment
Between between slope, the sampled value and the sampled value at the i-th moment at the i-th -1 moment when slope, the sampled value and i+1 at the i-th moment
Slope between the sampled value at quarter, by the slope of calculating compared with corresponding threshold value with judge whether occur step or whether
It there are the data segment that slope is more than the first preset value.
Step or the slope I -1 section as shown in Figure 2 more than the data segment of the first preset value occurs, which usually produces
Raw step signal, is step data segment, the last data section of the data segment and the slope of latter data section are more steadily steady
Data segment.
Recognize module 40, for the judgment module judging result for be when, using least square method with it is improved
Genetic algorithm carries out Model Distinguish to obtain the transfer function model of the control loop.
Power plant's control loop Model Distinguish method, system, by obtaining the history and real-time running data of Power Plant, base
Operation data are fitted to obtain curve in revolving door algorithm, judge whether that step occurs by the key point extracted on curve
Or it is being more than the first preset value in the slope for having data segment, if so, then being carried out using least square method and Revised genetic algorithum
Model Distinguish is adjusted with obtaining the transfer function model of the control loop to be controlled circuit pid parameter, is Power Plant
Optimization design and the debugging of control system provide modeling tool, without obtaining subtangent, the intermediate quantities such as 2 points again, simplify
The identification process of power plant's control loop model, has practicality.
In another embodiment, as shown in fig. 7, further including：
Inspection module 50, for when the judging result of the judgment module is is, obtaining corresponding four adjacent passes
The input data and output data of data segment between key point, and examine the reliable of the input data and the output data
Property, to ensure the identifiability of data.
Specifically, the inspection module 50 includes：
First judging unit, for calculating the related coefficient of the input data and the output data and judging the phase
Whether relation number is more than the second preset value.In a particular embodiment, using 0.9 as second preset value.
Processing unit, for when the judging result of first judging unit is is, calculating the input data respectively
Composed with the first amplitude of the output data, and place is normalized to the amplitude spectrum of the input data and output data respectively
Reason obtains the second amplitude spectrum of the input data and output data.
Comparing unit, for the input data and the similarity of the second amplitude spectrum of the output data；
Reliability determination unit, for the comparing unit comparative result for similarity within a preset range when, really
The fixed input data and the output data are reliable.
For stable Linear Time-Invariant System, the output steady-state component produced by harmonic wave input is still and input same frequency
Harmonic function, and the change of amplitude and phase is the function of frequency, and related to system mathematic model.So handle inputs, is defeated
Go out in data the harmonic wave for containing identical frequency as one of foundation for choosing data.Secondly, by the analysis of a large amount of field datas,
Input signal mainly includes the high-frequency harmonic of the larger low-frequency harmonics of amplitude and amplitude very little, the frequency of input signal low-frequency harmonics
For scope generally all in the lower passband of system object, low frequency signal can be by, and inputs, exports low frequency signal in signal
Amplitude Ration is a constant.High-frequency signal is then filtered after by system object.Analyzed more than, frequency in lower passband
The ratio between interior input, output signal amplitude are approximately a constant as another foundation for choosing data.
In another particular embodiment of the invention, as shown in fig. 7, further including
Module 60 is rejected, for rejecting false key point according to the plots changes.
Virtual key point is usually to influence little key point to the variation tendency of curve.For example, pass through more a certain pass
The previous line segment of key point and latter line segment slope, if the slope of two lines section is within a preset range, can assert that the key point is
The variation tendency of curve influences less, which is false key point.After rejecting the key point, before and after the key point and the key point
Trend term between key point merges, and the data segment between adjacent key point is a trend term.
In another particular embodiment of the invention, identification module includes：
Input unit, the transmission function structure of the control loop for obtaining input.
The transmission function structure of input different circuit, including first order inertial loop structure, single order are selected according to priori
Inertial delay link structure and second-order inertia delay link.
Identification unit, for being recognized using least square method of recursion to the model parameter of the control loop；
Optimize unit, for being optimized using Revised genetic algorithum to the result obtained by least square method of recursion, lead to
Cross the residual error of computation model output and the output of process and white noise verification is carried out to residual sequence to judge the control loop mould
The attribute of type.
Due to reasons such as disturbance, noises, identification result may not be optimal, therefore, using Revised genetic algorithum pair
Result obtained by least square method of recursion optimizes.Also, by computation model output with the residual error of the output of process and to residual
Difference sequence carries out white noise verification to judge the quality of control loop model.
Power plant's control loop model identification system, by obtaining the history and real-time running data of Power Plant, is based on
Revolving door algorithm to operation data be fitted to obtain curve, by the key point extracted on curve judge whether occur step or
It is being more than the first preset value in the slope for having data segment, and is obtaining and the data segment of step or slope more than the first preset value occurs,
Examine the reliability of input data and output data, if examine it is reliable, using least square method and Revised genetic algorithum into
Row Model Distinguish is adjusted with obtaining the transfer function model of the control loop to be controlled circuit pid parameter.The power plant is controlled
The model identification system in circuit processed, by testing to the input data and output data that produce step signal, so that it is guaranteed that
The identifiability of data, to improve the accuracy of Model Distinguish.
Each technical characteristic of above example can be combined arbitrarily, to make description succinct, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, lance is not present in the combination of these technical characteristics
Shield, is all considered to be the scope of this specification record.
Above example only expresses the several embodiments of the present invention, its description is more specific and detailed, but can not
Therefore it is construed as limiting the scope of the patent.It should be pointed out that for those of ordinary skill in the art,
On the premise of not departing from present inventive concept, various modifications and improvements can be made, these belong to protection scope of the present invention.
Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (6)
1. a kind of power plant's control loop identification Method, including：
Obtain the operation data of Power Plant；
Based on revolving door algorithm the operation data are carried out with quick Linear Quasi merging to determine to reflect curve on curve after being fitted
The key point of variation tendency；
Successively calculate four adjacent key points between data segment slope and respectively compared with corresponding threshold value with
Judge whether that step occurs or whether there are the data segment that slope is more than the first preset value；
If so, then obtaining the input data and output data of the data segment between corresponding four adjacent key points, and examine
The reliability of the input data and the output data；
If examining reliably, least square method is used to carry out Model Distinguish with Revised genetic algorithum to obtain the biography of the control loop
Delivery function model；
The step of reliability of the inspection input data and the output data, includes：
Calculate the related coefficient of the input data and the output data and judge whether the related coefficient is pre- more than second
If value；
If so, the first amplitude spectrum of the input data and the output data is then calculated respectively, and respectively to the input number
It is normalized to obtain the second amplitude spectrum of the input data and output data according to the amplitude spectrum with output data；
Compare the similarity of the second amplitude spectrum of the input data and the output data；
If similarity is within a preset range, it is determined that the input data and the output data are reliable.
2. power plant's control loop identification Method according to claim 1, it is characterised in that determining the step of key point
After rapid, the step of the slope of three data segments between key point described in the calculating any four before, further include step
Suddenly：False key point is rejected according to the plots changes.
3. power plant's control loop identification Method according to claim 1, it is characterised in that described to use least square
Method carries out Model Distinguish with Revised genetic algorithum to be included with obtaining the step of the transfer function model of the control loop：
Obtain the transmission function structure of the control loop of input；The transmission function includes first order inertial loop structure, single order is used to
Property delay link structure and second-order inertia delay link；
The model parameter of the control loop is recognized using least square method of recursion；
The result obtained by least square method of recursion is optimized using Revised genetic algorithum, passes through computation model output and mistake
The residual error of journey output simultaneously carries out white noise verification to judge the attribute of the control loop model to residual sequence.
4. a kind of power plant's control loop model identification system, including：
Acquisition module, for obtaining the operation data of Power Plant；
Fitting module, it is bent after being fitted for being determined based on revolving door algorithm to the quick Linear Quasi merging of the operation data progress
Reflect the key point of plots changes on line；
Judgment module, for calculate successively between four adjacent key points the slope of data segment and respectively with corresponding thresholding
Value is compared to judge whether that step occurs or whether there are data segment of the slope more than the first preset value；
Inspection module, for when the judging result of the judgment module is is, obtain corresponding four adjacent key points it
Between data segment input data and output data, and examine the reliability of the input data and the output data；
Module is recognized, for when the judging result of the judgment module is is, being calculated using least square method and improved heredity
Method carries out Model Distinguish to obtain the transfer function model of the control loop；
The inspection module includes：
First judging unit, for calculating the related coefficient of the input data and the output data and judging the phase relation
Whether number is more than the second preset value；
Processing unit, for when the judging result of first judging unit is is, calculating the input data and institute respectively
The first amplitude spectrum of output data is stated, and the amplitude spectrum of the input data and output data is normalized respectively
To the second amplitude spectrum of the input data and output data；
Comparing unit, for the input data and the similarity of the second amplitude spectrum of the output data；
Reliability determination unit, for the comparing unit comparative result for similarity within a preset range when, determine institute
State input data and the output data is reliable.
5. power plant's control loop model identification system according to claim 4, it is characterised in that further include：Reject module,
For rejecting false key point according to the plots changes.
6. power plant's control loop model identification system according to claim 4, it is characterised in that the identification module bag
Include：
Input unit, the transmission function structure of the control loop for obtaining input；The transmission function includes one order inertia ring
Section structure, one order inertia delay link structure and second-order inertia delay link；
Identification unit, for being recognized using least square method of recursion to the model parameter of the control loop；
Optimize unit, for being optimized using Revised genetic algorithum to the result obtained by least square method of recursion, pass through meter
Calculate the residual error of model output and the output of process and white noise verification is carried out to residual sequence to judge the control loop model
Attribute.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510686150.7A CN105259756B (en) | 2015-10-20 | 2015-10-20 | Power plant's control loop identification Method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510686150.7A CN105259756B (en) | 2015-10-20 | 2015-10-20 | Power plant's control loop identification Method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105259756A CN105259756A (en) | 2016-01-20 |
| CN105259756B true CN105259756B (en) | 2018-05-04 |
Family
ID=55099501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510686150.7A Active CN105259756B (en) | 2015-10-20 | 2015-10-20 | Power plant's control loop identification Method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105259756B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106569406B (en) * | 2016-11-07 | 2023-05-30 | 华侨大学 | Method for rapidly identifying model parameters of brush direct-current motor based on scale transformation |
| CN107688296A (en) * | 2017-09-06 | 2018-02-13 | 广州华润热电有限公司 | Automatic adjustment circuit online evaluation method and system based on SIS platforms |
| CN107643322A (en) * | 2017-09-15 | 2018-01-30 | 重庆交通大学 | Quality determining method and device |
| CN107942677B (en) * | 2017-12-17 | 2021-04-30 | 北京世纪隆博科技有限责任公司 | Industrial data mining method applied to closed-loop intelligent modeling |
| CN108549346B (en) * | 2018-05-14 | 2019-04-02 | 山东科技大学 | A kind of historical data section automatic searching method suitable for System Discrimination |
| CN110071667B (en) * | 2019-05-10 | 2020-09-25 | 福建省宏闽电力工程监理有限公司 | Doubly-fed wind generator parameter identification method |
| CN110426953B (en) * | 2019-07-18 | 2022-06-10 | 国网山东省电力公司电力科学研究院 | AGC performance evaluation method based on thermal power generating unit power generation model |
| CN111290267B (en) * | 2019-12-27 | 2023-08-15 | 国网河北省电力有限公司电力科学研究院 | Thermal power model identification device and identification method based on LabVIEW |
| CN112462604A (en) * | 2020-10-28 | 2021-03-09 | 国家能源集团泰州发电有限公司 | Genetic algorithm-based ultra-supercritical unit dry-wet state modeling and coordination control method |
| CN112965365B (en) * | 2021-02-23 | 2023-03-31 | 浙江中智达科技有限公司 | Model identification method, device and system of PID control loop and storage medium |
| CN113010887B (en) * | 2021-02-24 | 2022-03-11 | 天津大学 | Linear regression-based direct-current micro-grid virtual and false data injection attack defense method |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08286707A (en) * | 1995-04-11 | 1996-11-01 | Toshiba Corp | Automatic adjustment method of controller for multiplex loop control |
| JP3095199B2 (en) * | 1993-06-16 | 2000-10-03 | ランディス アンド スティーファー インコーポレイティド | Process control equipment |
| CN101131571A (en) * | 2007-08-15 | 2008-02-27 | 华北电力大学 | PID parameter setting method for monobloc generator group coordination and control system |
| CN101795138A (en) * | 2010-01-19 | 2010-08-04 | 北京四方继保自动化股份有限公司 | Compressing method for high density time sequence data in WAMS (Wide Area Measurement System) of power system |
| CN101968628A (en) * | 2010-09-08 | 2011-02-09 | 华北电力大学 | Saturated self-adjusting controller for time-varying delay uncertain system |
| CN102103376A (en) * | 2010-12-16 | 2011-06-22 | 浙江大学 | Method for evaluating control performance of PID (Proportion Integration Differentiation) loop based on subspace identification |
| CN202421881U (en) * | 2011-09-14 | 2012-09-05 | 华北电力大学 | Real-time monitoring device for controlling loop performance |
| CN102662326A (en) * | 2012-05-04 | 2012-09-12 | 广东电网公司电力科学研究院 | Method and device for industrial control by determining system parameter model |
| CN102722101A (en) * | 2012-06-26 | 2012-10-10 | 广东电网公司电力科学研究院 | Identification method and identification system based on closed-loop frequency domain |
| CN103646349A (en) * | 2013-09-27 | 2014-03-19 | 华北电力大学 | Power load curve segmented identification method |
| CN103760768A (en) * | 2013-12-30 | 2014-04-30 | 广东电网公司电力科学研究院 | Generator set multivariable system identification method based on normal operating data |
| CN104517034A (en) * | 2014-12-18 | 2015-04-15 | 广东电网有限责任公司电力科学研究院 | Method and system for identifying fossil power generation unit return passage model |
-
2015
- 2015-10-20 CN CN201510686150.7A patent/CN105259756B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3095199B2 (en) * | 1993-06-16 | 2000-10-03 | ランディス アンド スティーファー インコーポレイティド | Process control equipment |
| JPH08286707A (en) * | 1995-04-11 | 1996-11-01 | Toshiba Corp | Automatic adjustment method of controller for multiplex loop control |
| CN101131571A (en) * | 2007-08-15 | 2008-02-27 | 华北电力大学 | PID parameter setting method for monobloc generator group coordination and control system |
| CN101795138A (en) * | 2010-01-19 | 2010-08-04 | 北京四方继保自动化股份有限公司 | Compressing method for high density time sequence data in WAMS (Wide Area Measurement System) of power system |
| CN101968628A (en) * | 2010-09-08 | 2011-02-09 | 华北电力大学 | Saturated self-adjusting controller for time-varying delay uncertain system |
| CN102103376A (en) * | 2010-12-16 | 2011-06-22 | 浙江大学 | Method for evaluating control performance of PID (Proportion Integration Differentiation) loop based on subspace identification |
| CN202421881U (en) * | 2011-09-14 | 2012-09-05 | 华北电力大学 | Real-time monitoring device for controlling loop performance |
| CN102662326A (en) * | 2012-05-04 | 2012-09-12 | 广东电网公司电力科学研究院 | Method and device for industrial control by determining system parameter model |
| CN102722101A (en) * | 2012-06-26 | 2012-10-10 | 广东电网公司电力科学研究院 | Identification method and identification system based on closed-loop frequency domain |
| CN103646349A (en) * | 2013-09-27 | 2014-03-19 | 华北电力大学 | Power load curve segmented identification method |
| CN103760768A (en) * | 2013-12-30 | 2014-04-30 | 广东电网公司电力科学研究院 | Generator set multivariable system identification method based on normal operating data |
| CN104517034A (en) * | 2014-12-18 | 2015-04-15 | 广东电网有限责任公司电力科学研究院 | Method and system for identifying fossil power generation unit return passage model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105259756A (en) | 2016-01-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105259756B (en) | Power plant's control loop identification Method and system | |
| CN112382582B (en) | Wafer test classification method and system | |
| CN102152172B (en) | Cutter abrasion monitoring method based on co-integration modeling | |
| CN102369491B (en) | Operation information output device, method for controlling operation information output device, monitoring device, method for controlling monitoring device, and control program | |
| CN104390657B (en) | A kind of Generator Unit Operating Parameters measurement sensor fault diagnosis method and system | |
| EP2992340B1 (en) | System and method for ups battery monitoring and data analysis | |
| US20170060664A1 (en) | Method for verifying bad pattern in time series sensing data and apparatus thereof | |
| CN109741927B (en) | Intelligent prediction system for equipment faults and potential defective products of miniature transformer production line | |
| CN111401584B (en) | Automatic equipment fault diagnosis method and system | |
| CN115630847B (en) | Transceiving assembly detection method and system based on data prediction and storage medium | |
| CN103884943A (en) | Method for comprehensively analyzing and diagnosing deformation of winding of transformer | |
| CN110632546B (en) | Electronic transformer credibility evaluation method and device based on whole-network-domain evidence set | |
| CN103698489A (en) | Verification method and device of test data | |
| Yao et al. | Real-time chatter detection and automatic suppression for intelligent spindles based on wavelet packet energy entropy and local outlier factor algorithm | |
| CN107103425B (en) | Intelligent energy evaluation system for power generation equipment running state computer | |
| CN110298369A (en) | A kind of discrimination method and system of electric system bad data | |
| CN111855219A (en) | Diesel engine lubricating oil entering security parameter prediction method based on grey theory | |
| CN113255222B (en) | Intelligent auxiliary debugging method and device for cavity filter | |
| CN112711237B (en) | Automatic control quality online evaluation method and system for thermal power generating unit | |
| CN106485049B (en) | A kind of detection method of the NIRS exceptional sample based on Monte Carlo cross validation | |
| CN112149296A (en) | Method for judging stability type of hydrological time sequence | |
| CN115469585B (en) | Hydropower unit state monitoring method and system based on big data | |
| CN111473975A (en) | Intelligent diagnosis method for secondary gearbox fault based on characteristic vector baseline method | |
| CN104078381A (en) | Method for setting specification limit of measurement machine monitoring chart | |
| CN103761432B (en) | Method and device for analyzing board-level test data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |