CN104216288B - The gain self scheduling PID controller of thermal power plant's double-input double-output system - Google Patents
The gain self scheduling PID controller of thermal power plant's double-input double-output system Download PDFInfo
- Publication number
- CN104216288B CN104216288B CN201410432220.1A CN201410432220A CN104216288B CN 104216288 B CN104216288 B CN 104216288B CN 201410432220 A CN201410432220 A CN 201410432220A CN 104216288 B CN104216288 B CN 104216288B
- Authority
- CN
- China
- Prior art keywords
- loop
- slow
- fast
- control
- input
- 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
Landscapes
- Feedback Control In General (AREA)
Abstract
The present invention provides a kind of gain self scheduling PID controller of thermal power plant's double-input double-output system, tuning device is detected by the faster slower loop, faster slower circuit controls mode selector switch, hurry up, slow loop feedforward controller, and it is fast, slow loop feedforward and feedback synthesis arithmetic unit, and by the overlap joint of appropriate module, realize the Relay feedback test of broad sense double-input double-output system circuit, measure the critical concussion information of process, can be in the case of unknown controlled system Mathematical Modeling, automatically can just adjust out the P of double-input double-output system circuit in closed-loop control by required stability margin, I, D parameters；And according to the change of actual measurement operating condition signal and deviation signal, automatically, P, I, D parameter fast, that slow loop is optimal is inferred in real time, and the transit time of control system is shortened to greatest extent, reduce its dynamic deviation and static deviation.Make the parameter tuning more quick, science of multivariable control system, and make control system that there is preferable robustness.
Description
Technical field
The present invention relates to the multivariable control techniques field of flow production process, more particularly to a kind of thermal power plant's double-input double-output
The gain self scheduling PID controller of system.
Background technology
It is a complex process for multivariable in thermal power plant, there is very strong correlation between each variable, it is right
This need to employ the double-input double-output control strategy of broad sense, and the change of the operating condition of unit and fuel supply etc., such as load become
Change, coal varitation, will result in the dynamic characteristic change by a relatively large margin of production control object so that this kind of control object manifests
Go out strong coupling, time variation, non-linear.And modern multi-variant control method requirement has complete process model, in many situations
Under this model cannot accurately obtain so that many multi-variant control methods are difficult in power plant application.Modern power plants are big
The scattered control system (DCS) for all using carries out full factory's control, main using conventional PID (proportional-integral-differential) in DCS
Controller carries out closed-loop control, and is aided with various feedforward controllers and carries out uneoupled control, and this is built upon greatly control object dynamic
Characteristic it is constant or it is approximate it is constant on the basis of.Actually power plant production process has time variation, non-linear, when load, ature of coal
During Deng changing, plant characteristic can occur seriously, and if do not changed controller parameter, Control platform cannot ensure.Past each PID
The parameter tuning of controller is preset by experience value, then is manually adjusted according to the response wave shape for running, and this way is often
Experience is more than science, both time-consuming, is difficult to reach optimal setting valve again, because the bad adjustment of the controller differential action is often micro-
Point link is removed, and does not reach the effect of differential prediction control.Even more one consumption and the pid control parameter of double-input double-output system is adjusted
When, laborious work.
Ten thousand other PID controller setting methods are differed from there has been proposed dry.Although the circuit controls with PID self-setting functions
The existing commercially produced product of device, but it is general in how realizing still this area of adjusting of multivariable control loops pid parameter in dcs
All over the technical barrier for facing.
The content of the invention
The technical problem to be solved in the present invention be provide it is a kind of can be to fast, the slow loop of thermal power plant's double-input double-output system
P, I, D parameter carry out the gain self scheduling PID controller of thermal power plant's double-input double-output system adjusted in real time automatically.
A kind of gain self scheduling PID controller of thermal power plant's double-input double-output system, including：Fast control loop, the inspection of fast loop
Survey tuning device, fast loop feedforward controller, fast loop feedforward and feedback synthesis arithmetic unit, slow control loop, the detection of slow loop are adjusted
Device, slow loop feedforward controller, slow loop feedforward and feedback synthesis arithmetic unit, the first actuator and the second actuator；
The fast control loop includes that fast loop PID gain self scheduling controllers and fast loop control mode switch are switched,
The PID inputs of the fast loop control mode switch switch connect the output of the fast loop PID gains self scheduling controller
Detect the output end of tuning device, its output end connection fast loop feedforward in end, its ON-OFF input connection fast loop
The first input end of feedback integration arithmetic unit；
The fast loop detection tuning device receives the composite measurement signal in fast loop, and by the composite measurement in the fast loop
Signal compares with the setting signal in fast loop, and two control signals of first mode switching switch are exported according to comparative result computing
To the fast loop control mode switch switch；
The fast loop feedforward controller receives fast loop feed-forward signal, and it is anti-that its output end is connected to the fast loop feedforward
Second input of the comprehensive arithmetic unit of feedback；
The output end of the fast loop feedforward and feedback synthesis arithmetic unit is connected to first actuator；
The slow control loop includes slow loop PID gain self scheduling controllers and slow circuit controls mode selector switch,
The PID inputs of the slow circuit controls mode selector switch connect the output of the slow loop PID gains self scheduling controller
Detect the output end of tuning device, its output end connection slow loop feedforward in end, its ON-OFF input connection slow loop
The first input end of feedback integration arithmetic unit；
The slow loop detection tuning device receives the composite measurement signal in slow loop, and by the composite measurement in the slow loop
Signal compares with the setting signal in slow loop, and two control signals of second mode switching switch are exported according to comparative result computing
To the slow circuit controls mode selector switch；
The slow loop feedforward controller receives slow loop feed-forward signal, and it is anti-that its output end is connected to the slow loop feedforward
Second input of the comprehensive arithmetic unit of feedback；
The output end of the slow loop feedforward and feedback synthesis arithmetic unit is connected to second actuator；
The fast control loop also includes：Fast loop scale parameter gain self scheduling computing module, fast loop integral parameter
Gain self scheduling computing module and fast loop differential parameter gain self scheduling computing module；
The external scale parameter input connection fast loop ratio ginseng of the fast loop PID gains self scheduling controller
Number gain self scheduling computing module；Its external described fast loop integral parameter gain self scheduling of integral parameter input connection is calculated
Module；Its external differential parameter input connects the fast loop differential parameter gain self scheduling computing module.
Compared with prior art, the gain self scheduling PID controller of thermal power plant's double-input double-output system of the invention, passes through
The faster slower loop detection tuning device, faster slower circuit controls mode selector switch, fast, slow loop feedforward controller, and soon,
Slow loop feedforward and feedback synthesis arithmetic unit, and by the overlap joint of appropriate module, realize the relay of broad sense double-input double-output system circuit
Feedback control, measures the critical concussion information of process, can be in the case of unknown controlled system Mathematical Modeling, in closed-loop control
In automatically can just be adjusted out by required stability margin P, I, D parameter of double-input double-output system circuit；And transported according to actual measurement
The change of row working condition signal and deviation signal, automatically, infers P, I, D parameter fast, that slow loop is optimal, to greatest extent in real time
Shortening control system transit time, reduce its dynamic deviation and static deviation.Make the parameter tuning of multivariable control system
More fast, science, and make control system that there is preferable robustness (Robust).
Brief description of the drawings
Fig. 1 is the structure diagram of the gain self scheduling PID controller of thermal power plant's double-input double-output system of the present invention；
Fig. 2 is the detailed construction schematic diagram of the gain self scheduling PID controller of thermal power plant's double-input double-output system of the present invention；
Fig. 3 be thermal power plant's double-input double-output system of the present invention gain self scheduling PID controller in each PID controller computing
Timing diagram；
Fig. 4 is the gain self scheduling PID controller adjustment feedwater control parameter of thermal power plant's double-input double-output system of the present invention
Gain scheduling membership function schematic diagram.
Specific embodiment
Fig. 1 is referred to, Fig. 1 is the structure letter of the gain self scheduling PID controller of thermal power plant's double-input double-output system of the present invention
Figure.
A kind of gain self scheduling PID controller of thermal power plant's double-input double-output system, including：Fast control loop 10, fast loop
Detection tuning device 20, slow control loop 30, slow loop detection tuning device 40, fast loop feedforward controller 17, the feedforward of fast loop are anti-
The comprehensive arithmetic unit 16 of feedback, slow loop feedforward controller 37, slow loop feedforward and feedback synthesis arithmetic unit 36, the first actuator and second
Actuator；
The fast control loop 10 includes that fast loop PID gain self schedulings controller 11 and fast loop control mode switch are opened
15 are closed, the PID inputs of the fast loop control mode switch switch 15 connect the fast loop PID gains self scheduling controller
The output end of tuning device 20 is detected in 11 output end, its ON-OFF input connection fast loop, and the connection of its output end is described
The first input end of fast loop feedforward and feedback synthesis arithmetic unit 16；
The fast loop detection tuning device 20 receives the composite measurement signal in fast loop, and the synthesis in the fast loop is surveyed
Amount signal compares with the setting signal in fast loop, and first mode switch-over control signal to the fast loop is exported according to comparative result
Control mode switch switch 15；
The fast loop feedforward controller 17 receives fast loop feed-forward signal, and its output end is connected to the fast loop feedforward
Second input of feedback integration arithmetic unit 16；
The output end of the fast loop feedforward and feedback synthesis arithmetic unit 16 is connected to first actuator；
The slow control loop 30 includes that slow loop PID gain self schedulings controller 31 and slow circuit controls pattern switching are opened
35 are closed, the PID inputs of the slow circuit controls mode selector switch 35 connect the slow loop PID gains self scheduling controller
The output end of tuning device 40 is detected in 31 output end, its ON-OFF input connection slow loop, and the connection of its output end is described
The first input end of slow loop feedforward and feedback synthesis arithmetic unit 36；
The slow loop detection tuning device 40 receives the composite measurement signal in slow loop, and the synthesis in the slow loop is surveyed
Amount signal compares with the setting signal in slow loop, and second mode switch-over control signal to the slow loop is exported according to comparative result
Control mode switch switch 35；
The slow loop feedforward controller 37 receives slow loop feed-forward signal, and its output end is connected to the slow loop feedforward
Second input of feedback integration arithmetic unit 36；
The output end of the slow loop feedforward and feedback synthesis arithmetic unit 36 is connected to second actuator.
The gain self scheduling PID controller of thermal power plant's double-input double-output system of the invention, is detected by the faster slower loop
Tuning device, faster slower circuit controls mode selector switch, fast, slow loop feedforward controller, and soon, slow loop feedforward and feedback it is comprehensive
Arithmetic unit is closed, and by the overlap joint of appropriate module, realizes the Relay feedback test of broad sense double-input double-output system circuit, measure process
Critical concussion information, institute just can be automatically pressed in closed-loop control in the case of unknown controlled system Mathematical Modeling
It is required that stability margin adjust out P, I, D parameter of double-input double-output system circuit；And according to actual measurement operating condition signal and deviation
The change of signal, automatically, infers P, I, D parameter fast, that slow loop is optimal in real time, and control system is shortened to greatest extent
Transit time, reduce its dynamic deviation and static deviation.Make the parameter tuning more quick, science of multivariable control system, and
Make control system that there is preferable robustness (Robust).The fast control loop and the slow control loop can pass through respectively
The setting methods such as SPAM or Z-N calculate ratio (P), integration (I), differential (D) parameter of required stability margin, or by mould
Type discrimination method calculates the model parameter of circuit controls object, is calculated with the setting method of internal mode controller and meets nominal property
The parameter such as energy and P, I, D of stability requirement, respectively by multigroup P, I, D parameter and its actual measurement operating condition signal of input, from
Dynamic allotment P, I, D parameter, via controller controls non-linear controlled system, so that system is in whole working range, obtains preferable
Sound state Control platform.
Fig. 2 is referred to, Fig. 2 is the detailed knot of the gain self scheduling PID controller of thermal power plant's double-input double-output system of the present invention
Structure schematic diagram.
In the present embodiment, the fast loop detection tuning device 20 includes：First deviation alarm module 201, a RS
Trigger 202 and the first two output valve switching switches 203；
The input of the first deviation alarm module 201 is input into fast loop composite measurement signal and fast loop settings respectively
Signal, two output end is connected to the input of first rest-set flip-flop 202；
The output end of first rest-set flip-flop 202 is connected to the control of the first two output valves switching switch 203
End；
Two inputs of the first two output valves switching switch 203 are input into high potential signal and low potential letter respectively
Number；
The first deviation alarm module 201 compares the composite measurement signal in fast loop with setting signal, is obtained comparing
When the deviation for obtaining reaches alarming value, alarm signal is sent to first rest-set flip-flop 202；First rest-set flip-flop 202
According to the alarm signal to the first two output valves switching, 203 output switch control signals of switch；Described the first two
Under the control of the switch controlling signal, selection exports the high potential signal or the low electricity to output valve switching switch 203
Position signal is used as the first mode switch-over control signal.
When Relay feedback test is carried out, it is inclined that composite measurement signal and the setting signal in the fast loop are sent to described first
Difference alarming module 201 carries out computing, once deviation has exceeded dead band set in advance, the first deviation alarm module 201
High or low alarm is issued by, switching signal is sent by first rest-set flip-flop 202, described the first two of switching signal control
The high-order output of output of output valve switching switch 203 or low level output.So, the fast loop detection tuning device 20 just can be according to
According to measured value and the change of error situation of setting value, continuously send out alternate high-order or low signal output and realize relay feedback control
Function processed.
The slow loop detection tuning device 40 includes：Second deviation alarm module 401, the second rest-set flip-flop 402 and second
Two output valve switching switches 403；
The input of the second deviation alarm module 401 is input into slow loop composite measurement signal and slow loop settings respectively
Signal, two output end is connected to the input of second rest-set flip-flop 402；
The output end of second rest-set flip-flop 402 is connected to the control of the second two output valves switching switch 403
End；
Two inputs of the second two output valves switching switch 403 are input into high potential signal and low potential letter respectively
Number；
The second deviation alarm module 401 compares the composite measurement signal in slow loop with setting signal, is obtained comparing
When the deviation for obtaining reaches alarming value, alarm signal is sent to second rest-set flip-flop；The basis of second rest-set flip-flop 402
The alarm signal is to the second two output valves switching, 403 output switch control signals of switch；The second two output
Under the control of the switch controlling signal, selection exports the high potential signal or low potential letter to value switching switch 403
Number as the second mode switch-over control signal.
When Relay feedback test is carried out, measured value and the setting value in the slow loop are sent to the second deviation alarm mould
Block 401 carries out computing, once deviation has exceeded dead band set in advance, the second deviation alarm module 401 is issued by height
Or low alarm, switching signal is sent by second rest-set flip-flop 402, switching signal control the second two output valves are cut
Change the high-order output of the output of switch 403 or low level output.So, the slow loop detection tuning device 40 just can be according to measured value
With the change of error situation of setting value, continuously send out alternate high-order or low signal output and realize Relay feedback test function.
In the present embodiment, further, the fast control loop 10 may also include：The gain of fast loop scale parameter is certainly
Scheduling computing module 12, fast loop integral parameter gain self scheduling computing module 13 and fast loop differential parameter gain self scheduling meter
Calculate module 14；The external scale parameter input of the fast loop PID gain self schedulings controller 11 connects the fast loop ratio
Example parametric gain self scheduling computing module 12；Its external integral parameter input connects the fast loop integral parameter gain self-regulated
Degree computing module 13；Its external differential parameter input connects the fast loop differential parameter gain self scheduling computing module 14.
Each gain self scheduling computing module, the degree of each typical condition is deviateed according to the operating mode of current operation, using intelligence
Can inference rule from multigroup P, I, D parameter of input, extrapolate and meet under current working control stability requirement P, I, D control
Parameter.Extrapolate the outer given side of control parameter that the parameters such as P, I, D of correspondence current working are sent to each loop control unit respectively.It is logical
Cross the fast loop scale parameter gain self scheduling computing module 12, the fast and of loop integral parameter gain self scheduling computing module 13
Fast loop differential parameter gain self scheduling computing module 14, makes control loop 10 in PID gain self schedulings increase with P, I, D parameter
Beneficial self scheduling function.
In actual motion, the composite measurement signal in fast loop is connected to the fast loop PID gain self schedulings controller 11
Measurement end PV, the setting value in fast loop is connected to the setting end SP of the fast loop PID gain self schedulings controller 11.Return to
Road operating condition signal is coupled with the fast loop scale parameter gain self scheduling computing module 12, the fast contour integration ginseng
Number gain self scheduling computing module 13 and the fast loop differential parameter gain self scheduling computing module 14.The fast loop PID
The output of gain self scheduling controller 11 terminates to the A ends (PID) of the fast loop control mode switch switch 15, described to return to
The output of two output valves switching switch 203 of road detection tuning device 20 is connected to the fast loop control mode switch switch 15
B ends (ON-OFF).The output of the fast loop control mode switch switch 15 is connected to fast loop feedforward and feedback synthesis arithmetic unit 17
1 end, the output of fast loop feedforward controller 16 is connected to 2 ends of fast loop feedforward and feedback synthesis arithmetic unit 17, fast loop feedforward
The output of feedback integration arithmetic unit 17 is connected to first actuator in fast loop.
In the present embodiment, the slow control loop 30 may also include：Slow loop scale parameter gain self scheduling is calculated
Module 32, slow contour integration parametric gain self scheduling computing module 33 and slow loop differential parameter gain self scheduling computing module
33；The external scale parameter input of the slow loop PID gain self schedulings controller 31 connects the slow loop scale parameter
Gain self scheduling computing module 32；Its external described slow contour integration parametric gain self scheduling of integral parameter input connection is calculated
Module 33；Its external differential parameter input connects the slow loop differential parameter gain self scheduling computing module 34.
Each gain self scheduling computing module, the degree of each typical condition is deviateed according to the operating mode of current operation, using intelligence
Can inference rule from multigroup P, I, D parameter of input, extrapolate and meet under current working control stability requirement P, I, D control
Parameter.Extrapolate the outer given side of control parameter that the parameters such as P, I, D of correspondence current working are sent to each loop control unit respectively.It is logical
Cross the slow loop scale parameter gain self scheduling computing module 32, the slow and of contour integration parametric gain self scheduling computing module 33
Slow loop differential parameter gain self scheduling computing module 33, makes the PID gains self scheduling outer control loop 30 join with P, I, D
Number gain self scheduling function.
In actual motion, the composite measurement signal in slow loop is connected to the slow loop PID gain self schedulings controller 31
Measurement end PV, the setting value sent by slow loop is connected to the setting end of the slow loop PID gain self schedulings controller 31
SP.Slow loop operating condition signal be coupled with the slow loop scale parameter gain self scheduling computing module 32, described slow time
Road integral parameter gain self scheduling computing module 33 and the slow loop differential parameter gain self scheduling computing module 33.It is described slow
The output end OT1 of loop PID gain self schedulings controller 31 is connected to the A ends of the slow circuit controls mode selector switch 35
(PID), the output of the second two output valve switching switches 403 in the slow loop detection tuning device 40 is connected to the slow loop
The B ends (ON-OFF) of control mode switch switch 35.The output of the slow circuit controls mode selector switch 35 is connected to described slow
1 end of loop feedforward and feedback synthesis arithmetic unit 37, the output of slow loop feedforward controller 36 is connected to slow loop feedforward and feedback synthesis
2 ends of arithmetic unit 37, the output of slow loop feedforward and feedback synthesis arithmetic unit 37 is connected to second actuator in slow loop.
Explanation below carries out the pid parameter setting method of double-input double-output system using the present invention：
For complex industrial process control system, when certain typical condition needs to detect broad sense double-input double-output system controlled pair
During as control characteristic, following operation is performed：
1. the control mode switch switch 15 of the fast control loop 10 is switched under detection pattern：I.e. by suitable
Dead band, two the high-order of switching switch when adjustment deviation alarm module export (Oh) and low level output valve (O1), make described fast
Loop detection tuning device 20 realizes Relay feedback test function.Meanwhile, slow control loop 30 remains at pid control mode, right
The exporting change scope of slow control loop is limited, and holds it at the operation interval of stabilization.
2. the switching control state of the PID controller and the measured value of control system are obtained.
3. by detecting fast loop threshold oscillation to the time T of peakingPAnd reach the time T of valleyV；Upper hemiwave time
TMAXAnd lower hemiwave time TMIN, threshold oscillation peak value VMAXAnd valley VMIN, it is possible to calculate control system threshold oscillation
Cycle time T (T=TMAX+TMIN) and threshold oscillation amplitude V (V=VMAX-VMIN).Measure out critical period of the oscillation T and amplitude
V, time to peak TPAnd valley time TV.As shown in Figure 3.
4. using the critical period of the oscillation T and amplitude V for measuring, the stagnant ring of its corresponding stagnant ring relay-type switch control
The Relay feedback test parameter such as width (H) and hysteresis band (e), it is possible to the ratio in the loop is calculated using Z-N setting methods
(P), integration (I), differential (D) numerical value；Required phase angle stabilization, nargin (Q can also be setm) and amplitude stability nargin (Am),
P, I, D parameter of required stability margin are calculated using SPAM setting methods；Can also be calculated by identification Method
The model parameter of circuit controls object, is calculated with the setting method of internal mode controller and meets nominal performance and stability requirement
The parameter such as P, I, D.
5. the numerical value of the parameters such as P, I, the D that will then calculate and this operating mode boot parameter η etc. is respectively filled in each intelligence
Gain self scheduling computing module to should in the control law of operating mode so that formed fuzzy P, I, D parameter of complete full working scope from
Adjustment control law.And fast loop is switched back into PID control pattern.
Then, by the described slow circuit controls mode selector switch 35 of the slow loop PID gain self schedulings controller 30
Detection pattern end is switched to, i.e., using the adjustable Relay feedback test function of stagnant ring.Returned slowly by being detected the step of 1~4, being calculated
The ratio (P) of road PID gain self scheduling controllers, integration (I), differential (D) numerical value.Then ratio (P), the integration that will be calculated
(I), numerical value of differential (D) numerical value and this operating mode boot parameter η etc., is filled into the corresponding slow loop ratio of this operating mode, integration, micro-
Point parameter is from scheduling module.And slow loop is switched back into PID control pattern.
Under each typical condition, the step of using the above, by starting double-input double-output system gain self scheduling PID of the present invention
The detection pattern of controller, detects, adjusts, calculating, inserting ratio (P), integration (I), differential (D) and its guiding under different operating modes
The numerical value of parameter η, can just form correspondence nonlinear system improves full working scope intelligence P, I, D parameters self-tuning control law.
By multigroup P, I, D parameter setting adjusted out in each typical condition in parameters gain self scheduling computing device,
Multigroup P, I, D parameter and typical condition boot parameter of input are η, and each PID controller is accessed using outer given way, to
Cause controlled system control characteristic to change during working conditions change, P, I, D parameter are adaptively allocated automatically, by double-input double-output of the present invention
System gain self scheduling PID controller controls non-linear controlled system, so that system is in whole working range, obtains preferably dynamic
Static cost control quality.
In the present invention, parameters gain self scheduling computing module uses Takagi-Sugeno-Kang types intelligent inference side
Method, control law forms as follows：
Parameters gain self scheduling computing module, the degree of each typical condition is deviateed according to the operating mode of current operation, is adopted
With intelligent inference rule from multigroup P, I, D parameter of input, extrapolate and meet under current working control stability requirement P, I, D
Control parameter.Extrapolate correspondence current working the parameter such as P, I, D be sent to respectively outside the control parameter of each loop PID controller to
Fixed end.
If the calculating formula of PID controller is as follows：
In formula：U (s) controllers are exported, KprIt is ratio, TirIt is the time of integration, TdrIt is derivative time, FF feed-forward signals.
The boot parameter of control domain is η (setting parameter is S), is the PID that η is determined by boot parameter when control domain is in Zone R
Control law is as follows：
Control parameter has its membership function mui in P, I, D parameter of control point ηA k(α) is determined, generality is not lost any
The control parameter of control point η is determined by two groups of P, I, D parameters of adjacent control domain, it is assumed that common factor of this in control domain k and k-1
On, then P, I, D parameter of the point are drawn by following formula：
Kpr=(1- μA k(α))Kp K-1+μA K(α)Kp K
Tir=(1- μA k(α))Ti k-1+μA k(α)Ti k
Tdr=(1- μA k(α))Td k-1+μA k(α)Td k
According to the operating mode of current operation, P, I, D the control ginseng for meeting stability margin requirement, correspondence current working are extrapolated
Number.Correspondingly the ratio P of current working, integration I, differential D parameters is extrapolated to be sent to outside the control parameter of the PID controller in each loop
Given side.
The present invention, by increasing module, and appropriate module overlap joint newly, is realized each using module intrinsic in each DCS
The gain self scheduling control of the Relay feedback test, double-input double-output system in loop, and it is applied to various broad sense double-input double-output systems
Control loop includes：The boiler-turbine coordinated control loop of various fired power generating units, the pressure load of thermal power plant unit are coordinated control and are controlled back
Road, the Control System of Load Distribution loop of Combined Cycle Unit etc..
The fortune of double-input double-output system gain self scheduling PID controller of the present invention is illustrated with concrete engineering application example below
Make：
Controlled using double-input double-output system gain self scheduling PID of the present invention in the control of 300MW circulating fluidized bed unit cooperatives
Device processed.Unit cooperative control is typical double-input double-output system, and its two inputs are respectively the positions of boiler oil and pitch,
Two outputs are generated output and steam pressure.There are fast, slow two PID controllers in coordinated control system.Ramping controller PID1
It is boiler implosion loop, it mainly according to steam pressure deviation and is aided with the decoupling signal that turbine control loop comes and controls to coal
Amount；Fast controller PID2 is turbine control loop, and it mainly according to generation load deviation and is aided with what boiler implosion loop came
Decoupling signal controls the position of pitch；Thus what is constituted is the control system of double-input double-output.
Because coordinated control system has close coupling, serious non-linear, very big difficulty is brought to unit allocation, by
Gain self scheduling PID controller is used in turbine control, this double-input double-output system of boiler implosion loop, is preferably overcome negative
The influence of lotus, coal varitation etc. interference, controls the parameters such as generation load, steam pressure of unit to control well.Realize
The load rapid track and control of Properties of CFB.
Respectively in high and low two loads section, setting method is fed back using relay, measured boiler implosion loop, steam turbine
The characterisitic parameter of each comfortable two loads section control object of control loop.The control parameter that different load section is adjusted out see the table below.
Setting parameter in the coordinated control system of table 1 under controller different load
Then, gain scheduling calculating is carried out using following inference rule in each parametric gain self scheduling module.
Boiler, steam turbine loop control parameters gain scheduling rule it is as follows：
Wherein boot parameter η is main quantity of steam S-FLOW, so obtains boiler, the gain scheduling of steam turbine loop control parameters
Membership function is as shown in Figure 4.
The pid parameter calculated with Tuning under being filled in different load in speed gain scheduling loop, just forms and assists
Adjust the double-input double-output gain self scheduling PID controller of control system.
Invention described above implementation method, is not intended to limit the scope of the present invention..It is any in the present invention
Spirit and principle within modification, equivalent and the improvement made etc., should be included in claim protection model of the invention
Within enclosing.
Claims (6)
1. the gain self scheduling PID controller of a kind of thermal power plant's double-input double-output system, it is characterised in that including：Fast control loop,
Fast loop detection tuning device, fast loop feedforward controller, fast loop feedforward and feedback synthesis arithmetic unit, slow control loop, slow loop
Detection tuning device, slow loop feedforward controller, slow loop feedforward and feedback synthesis arithmetic unit, the first actuator and the second actuator；
The fast control loop includes that fast loop PID gain self scheduling controllers and fast loop control mode switch are switched, described
The PID inputs of fast loop control mode switch switch connect the output end of the fast loop PID gains self scheduling controller, its
The output end of tuning device is detected in the ON-OFF inputs connection fast loop, and it is comprehensive that its output end connects the fast loop feedforward and feedback
Close the first input end of arithmetic unit；
The fast loop detection tuning device receives the composite measurement signal in fast loop, and by the composite measurement signal in the fast loop
Setting signal with fast loop compares, two control signals according to comparative result computing output first mode switching switch to institute
State fast loop control mode switch switch；
The fast loop feedforward controller receives fast loop feed-forward signal, and it is comprehensive that its output end is connected to the fast loop feedforward and feedback
Close the second input of arithmetic unit；
The output end of the fast loop feedforward and feedback synthesis arithmetic unit is connected to first actuator；
The slow control loop includes slow loop PID gain self scheduling controllers and slow circuit controls mode selector switch, described
The PID inputs of slow circuit controls mode selector switch connect the output end of the slow loop PID gains self scheduling controller, its
The output end of tuning device is detected in the ON-OFF inputs connection slow loop, and it is comprehensive that its output end connects the slow loop feedforward and feedback
Close the first input end of arithmetic unit；
The slow loop detection tuning device receives the composite measurement signal in slow loop, and by the composite measurement signal in the slow loop
Setting signal with slow loop compares, two control signals according to comparative result computing output second mode switching switch to institute
State slow circuit controls mode selector switch；
The slow loop feedforward controller receives slow loop feed-forward signal, and it is comprehensive that its output end is connected to the slow loop feedforward and feedback
Close the second input of arithmetic unit；
The output end of the slow loop feedforward and feedback synthesis arithmetic unit is connected to second actuator；
The fast control loop also includes：Fast loop scale parameter gain self scheduling computing module, fast loop integral parameter gain
Self scheduling computing module and fast loop differential parameter gain self scheduling computing module；
The external scale parameter input connection fast loop scale parameter of the fast loop PID gains self scheduling controller increases
Beneficial self scheduling computing module；Its external described fast loop integral parameter gain self scheduling of integral parameter input connection calculates mould
Block；Its external differential parameter input connects the fast loop differential parameter gain self scheduling computing module.
2. the gain self scheduling PID controller of thermal power plant's double-input double-output system as claimed in claim 1, it is characterised in that institute
Stating fast loop detection tuning device includes：First deviation alarm module, the first rest-set flip-flop and the first two output valve switching switches；
The input of the first deviation alarm module is input into fast loop composite measurement signal and fast loop setting signal respectively, its
Two output ends are connected to the input of first rest-set flip-flop；
The output end of first rest-set flip-flop is connected to the control end of the first two output valves switching switch；
Two inputs of the first two output valves switching switch are input into high potential signal and low-potential signal respectively；
The first deviation alarm module compares the composite measurement signal in fast loop with setting signal, is comparing the deviation of acquisition
When value reaches alarming value, alarm signal is sent to first rest-set flip-flop；First rest-set flip-flop is according to the alarm signal
Number to the first two output valves switching switch output switch control signal；The first two output valves switching switch is in institute
State under the control of switch controlling signal, selection exports the high potential signal or the low-potential signal as the first mode
Switch two control signals of switch.
3. the gain self scheduling PID controller of thermal power plant's double-input double-output system as described in claim 1 or 2, its feature exists
In the fast control loop also includes：The fast loop scale parameter gain self scheduling computing module, fast loop integral parameter increase
The input of beneficial self scheduling computing module and fast loop differential parameter gain self scheduling computing module connects the fortune in fast loop respectively
Row working condition signal, the degree of each typical condition is deviateed according to the operating mode of current operation, using intelligent inference rule, from many of input
Extrapolated in group P, I, D parameter and meet under current working control stability requirement P, I, D control parameter, extrapolate the current work of correspondence
P, I, D control parameter of condition are sent to the outer given side of control parameter of the fast loop PID gains self scheduling controller respectively.
4. the gain self scheduling PID controller of thermal power plant's double-input double-output system as claimed in claim 1, it is characterised in that institute
Stating slow control loop also includes：Slow loop scale parameter gain self scheduling computing module, slow contour integration parametric gain self scheduling
Computing module and slow loop differential parameter gain self scheduling computing module；
The external scale parameter input connection slow loop scale parameter of the slow loop PID gains self scheduling controller increases
Beneficial self scheduling computing module；Its external described slow contour integration parametric gain self scheduling of integral parameter input connection calculates mould
Block；Its external differential parameter input connects the slow loop differential parameter gain self scheduling computing module.
5. the gain self scheduling PID controller of thermal power plant's double-input double-output system as claimed in claim 4, it is characterised in that institute
Stating slow loop detection tuning device includes：Second deviation alarm module, the second rest-set flip-flop and the second two output valve switching switches；
The input of the second deviation alarm module is input into slow loop composite measurement signal and slow loop settings signal respectively, its
Two output ends are connected to the input of second rest-set flip-flop；
The output end of second rest-set flip-flop is connected to the control end of the second two output valves switching switch；
Two inputs of the second two output valves switching switch are input into high potential signal and low-potential signal respectively；
The second deviation alarm module compares the composite measurement signal in slow loop with setting signal, is comparing the deviation of acquisition
When value reaches alarming value, alarm signal is sent to second rest-set flip-flop；Second rest-set flip-flop is according to the alarm signal
Number to the second two output valves switching switch output switch control signal；The second two output valves switching switch is in institute
State under the control of switch controlling signal, selection exports the high potential signal or the low-potential signal as the second mode
Switch two control signals of switch.
6. the gain self scheduling PID controller of the thermal power plant's double-input double-output system as described in claim 4 or 5, its feature exists
In the slow control loop also includes：The slow loop scale parameter gain self scheduling computing module, slow contour integration parameter increase
The input of beneficial self scheduling computing module and slow loop differential parameter gain self scheduling computing module connects the fortune in slow loop respectively
Row working condition signal, the degree of each typical condition is deviateed according to the operating mode of current operation, using intelligent inference rule, from many of input
Extrapolated in group P, I, D parameter and meet under current working control stability requirement P, I, D control parameter, extrapolate the current work of correspondence
P, I, D control parameter of condition are sent to the outer given side of control parameter of the slow loop PID gains self scheduling controller respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410432220.1A CN104216288B (en) | 2014-08-28 | 2014-08-28 | The gain self scheduling PID controller of thermal power plant's double-input double-output system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410432220.1A CN104216288B (en) | 2014-08-28 | 2014-08-28 | The gain self scheduling PID controller of thermal power plant's double-input double-output system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104216288A CN104216288A (en) | 2014-12-17 |
| CN104216288B true CN104216288B (en) | 2017-06-13 |
Family
ID=52097908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410432220.1A Active CN104216288B (en) | 2014-08-28 | 2014-08-28 | The gain self scheduling PID controller of thermal power plant's double-input double-output system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104216288B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10001764B2 (en) | 2015-09-11 | 2018-06-19 | Woodward, Inc. | Adaptive multiple input multiple output PID control system for industrial turbines |
| CN105485674B (en) * | 2016-01-21 | 2017-12-05 | 山西大学 | The feedwater instruction construction method of overcritical low heat value CFBB |
| CN105974800B (en) * | 2016-07-26 | 2019-05-28 | 广东电网有限责任公司电力科学研究院 | A kind of feedback feedforward control system |
| US10606254B2 (en) * | 2016-09-14 | 2020-03-31 | Emerson Process Management Power & Water Solutions, Inc. | Method for improving process/equipment fault diagnosis |
| CN113093521A (en) * | 2021-03-19 | 2021-07-09 | 江苏固德威电源科技股份有限公司 | Dynamic PID control method and dynamic PID controller |
| CN113267996A (en) * | 2021-06-11 | 2021-08-17 | 福州大学 | Novel digital PID circuit and control method thereof |
| CN113467331B (en) * | 2021-07-28 | 2024-02-09 | 华电莱州发电有限公司 | Method for analyzing influence of controller parameters on automatic power generation control regulation performance |
| CN115167142A (en) * | 2022-07-29 | 2022-10-11 | 华能伊敏煤电有限责任公司 | Multi-heat-source heat supply unit combined control method, system, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6055524A (en) * | 1997-10-06 | 2000-04-25 | General Cybernation Group, Inc. | Model-free adaptive process control |
| US6095426A (en) * | 1997-11-07 | 2000-08-01 | Siemens Building Technologies | Room temperature control apparatus having feedforward and feedback control and method |
| CN102540878A (en) * | 2012-03-16 | 2012-07-04 | 广东电网公司电力科学研究院 | Cascading gain self-scheduling PID (Proportion Integration Differentiation) controller |
| CN204155077U (en) * | 2014-08-28 | 2015-02-11 | 广东电网公司电力科学研究院 | The gain self scheduling PID controller of thermal power plant's double-input double-output system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0822306A (en) * | 1994-07-08 | 1996-01-23 | Mitsubishi Heavy Ind Ltd | Automatic adjusting device for arithmetic control parameter |
-
2014
- 2014-08-28 CN CN201410432220.1A patent/CN104216288B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6055524A (en) * | 1997-10-06 | 2000-04-25 | General Cybernation Group, Inc. | Model-free adaptive process control |
| US6095426A (en) * | 1997-11-07 | 2000-08-01 | Siemens Building Technologies | Room temperature control apparatus having feedforward and feedback control and method |
| CN102540878A (en) * | 2012-03-16 | 2012-07-04 | 广东电网公司电力科学研究院 | Cascading gain self-scheduling PID (Proportion Integration Differentiation) controller |
| CN204155077U (en) * | 2014-08-28 | 2015-02-11 | 广东电网公司电力科学研究院 | The gain self scheduling PID controller of thermal power plant's double-input double-output system |
Non-Patent Citations (1)
| Title |
|---|
| 基于串级PID的过热汽温多模型切换控制;程光坤等;《电站系统工程》;20081130;第24卷(第6期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104216288A (en) | 2014-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104216288B (en) | The gain self scheduling PID controller of thermal power plant's double-input double-output system | |
| CN102540878B (en) | Cascading gain self-scheduling PID (Proportion Integration Differentiation) controller | |
| Li et al. | A new coordinated control strategy for boiler-turbine system of coal-fired power plant | |
| CN102841539B (en) | Based on the subcritical control method for coordinating of multi-model PREDICTIVE CONTROL | |
| CN106873381A (en) | Spray ammonia control system | |
| CN105180135A (en) | Unit plant coordination control method and system | |
| CN102841540A (en) | MMPC-based supercritical unit coordination and control method | |
| CN105373065A (en) | Dry fog dedusting control system and method | |
| CN110309585A (en) | A kind of implementation method of flexibility tuning controller | |
| CN107831652A (en) | A kind of unit load intelligent optimized control method based on cold end system energy storage | |
| CN106406101A (en) | Intelligent calculating prediction control method of thermal power generating unit coordination control system | |
| CN111102559A (en) | Screen type superheater steam temperature control method based on double neural network inverse model | |
| CN204155077U (en) | The gain self scheduling PID controller of thermal power plant's double-input double-output system | |
| CN113390099A (en) | Coordination control method under coal-fired cogeneration unit self-adaptive BTU | |
| CN102854797B (en) | Fired power generating unit Dynamic matrix control multi-model switching method | |
| CN202661791U (en) | Cascade gain-self-scheduling PID (proportion integration differentiation) controller | |
| CN113485499B (en) | Coal feeding regulation and control method for coal quality working condition change | |
| Hu et al. | Feedforward DMC-PID cascade strategy for main steam temperature control system in fossil-fired power plant | |
| Ma et al. | Intelligent Compensation for the Set Values of PID Controllers to Improve Boiler Superheated Steam Temperature Control | |
| CN112686538B (en) | Data-driven thermal process adjustment quality calculation method and device | |
| Ahooghalandari et al. | $\mathcal {H} _ {\infty} $ Robust Control Design for Three-Element Industrial Boiler Supervisory System | |
| Ma et al. | Neural network inverse control for the coordinated system of a 600MW supercritical boiler unit | |
| Xing et al. | Fuzzy-PID controller with variable integral parameter for temperature control in variable air volume air conditioning systems | |
| Liu et al. | Fuzzy PID Control Strategy Applied in Boiler Combustion System | |
| Meng et al. | The Application of Intelligent Control to Combustion Control System of CFB Boiler |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 510080 water Donggang 8, Dongfeng East Road, Yuexiu District, Guangzhou, Guangdong. Patentee after: ELECTRIC POWER RESEARCH INSTITUTE, GUANGDONG POWER GRID CO., LTD. Address before: 510080 water Donggang 8, Dongfeng East Road, Yuexiu District, Guangzhou, Guangdong. Patentee before: Electrical Power Research Institute of Guangdong Power Grid Corporation |