CN103760931B  The oil gas water horizontal threephase separator compress control method that dynamic matrix control optimizes  Google Patents
The oil gas water horizontal threephase separator compress control method that dynamic matrix control optimizes Download PDFInfo
 Publication number
 CN103760931B CN103760931B CN201410029644.3A CN201410029644A CN103760931B CN 103760931 B CN103760931 B CN 103760931B CN 201410029644 A CN201410029644 A CN 201410029644A CN 103760931 B CN103760931 B CN 103760931B
 Authority
 CN
 China
 Prior art keywords
 phase separator
 oil gas
 gas water
 water horizontal
 pressure
 Prior art date
Links
 239000007789 gases Substances 0.000 title claims abstract description 65
 230000000875 corresponding Effects 0.000 claims abstract description 7
 230000001276 controlling effects Effects 0.000 claims description 18
 238000005070 sampling Methods 0.000 claims description 12
 230000017105 transposition Effects 0.000 claims description 6
 206010018987 Haemorrhages Diseases 0.000 claims description 5
 230000000740 bleeding Effects 0.000 claims description 5
 231100000319 bleeding Toxicity 0.000 claims description 5
 238000004062 sedimentation Methods 0.000 claims description 4
 238000000034 methods Methods 0.000 description 4
 238000005516 engineering processes Methods 0.000 description 2
 238000010276 construction Methods 0.000 description 1
 238000004519 manufacturing process Methods 0.000 description 1
 238000004886 process control Methods 0.000 description 1
 238000007670 refining Methods 0.000 description 1
Abstract
Description
Technical field
The invention belongs to technical field of automation, relate to a kind of oil gas water optimized based on dynamic matrix control (DMC) and crouch Pressure proportional integrationproportion differential (PIPD) control method in formula three phase separator.
Background technology
PID controller simple in construction, easy to control, it is widely used in various industrial control system.But, for longpending Dividing, vibrate or the control object of instability, PID is sometimes difficult to meet higher control requirement.Such as, at Stepped Impedance Resonators Time, often producing bigger hyperharmonic vibration, this may bring potential safety hazard to production.At present, the horizontal three phase separation of oil gas water The control of device pressure is to use PID to control mostly, if can control plus PD at internal ring, first suppresses its overshoot, and outer shroud uses PI Control, more preferable production performance will be obtained.Dynamic array control algorithm is as the one of advanced control algorithm, to model needs The lowest, control performance is good simultaneously, if can dynamic matrix control and PIPD technology be combined, can improve oil refining further With the efficiency collecting natural gas.
Summary of the invention
It is an object of the invention to the weak point for existing PID controller, it is provided that a kind of excellent based on dynamic matrix control The PIPD control method of pressure in the oil gas water horizontal threephase separator changed, for Reducing overshoot, in order to obtain actual Control performance.The method controls by combining dynamic matrix control and PIPD, has obtained a kind of with dynamic matrix control performance PIPD control method.The method not only inherits the premium properties of dynamic matrix control, and Simultaneous Forms is simple and can meet real The needs of border industrial process.
The inventive method is primarily based on the step response data of the pressure object in oil gas water horizontal threephase separator and sets up In oil gas water horizontal threephase separator, the model of pressure object, excavates basic plant characteristic；Then according to dynamic matrix control The characteristic of system is gone to adjust the parameter of corresponding PIPD controller；Finally real to the pressure object in oil gas water horizontal threephase separator Execute PIPD control.
Technical scheme passes through data acquisition, sets up dynamic matrix, sets up forecast model, prediction mechanism, optimization Etc. means, establish a kind of PIPD control method optimized based on dynamic matrix control, utilize the method can effective Reducing overshoot And improve the stability of system.
The step of the inventive method includes:
Step (1). set up the model of controlled device by the realtime step response data of process object, concrete grammar is:
A. one step input signal of controlled device, the step response curve of record controlled device are given.
B. it is filtered the step response curve that a step obtains processing, then fits to a smooth curve, recording light The step response data that on sliding curve, each sampling instant is corresponding, first sampling instant is T_{s}, between adjacent two sampling instants Every time be T_{s}, sampling instant order is T_{s}、2T_{s}、3T_{s}……；The step response of controlled device will be at some moment t_{N}= Tend to be steady after NT, work as a_{i}(i ＞ N) and a_{N}Error and measurement error when having the identical order of magnitude, i.e. it is believed that a_{N}It is approximately equal to The steadystate value of step response.Set up the model vector a of object:
A=[a_{1},a_{2},…a_{N}]^{Τ}
Wherein Τ is the transposition symbol of matrix, a_{i}Being the data of process object step response, N is modeling time domain.
Step (2). the PIPD controller of design controlled device, concrete grammar is:
A. the dynamic matrix of controlled device is set up
The model vector a utilizing step (1) b to obtain, that sets up controlled device dynamically controls matrix, and its form is as follows:
Wherein, A is P × M rank dynamic matrix of controlled device, and P is the optimization time domain of Dynamic array control algorithm, and M is The control time domain of state matrix control algorithm, M ＜ P ＜ N.
B. model prediction initial communication value y in controlled device current k moment is calculated_{M}(k)
1.. the calculating k1 moment adds the model predication value y after controlling increment Δ u (k1)_{p}(k1):
y_{P}(k1)=y_{M}(k1)+A_{0}Δu(k1)
Wherein,
y_{1}(kk1),y_{1}(k+1k1),…,y_{1}(k+N1  k1) represent respectively controlled device in the k1 moment to k, k+ 1 ..., the k+N1 moment adds the model predication value after controlling increment Δ u (k1), y_{0}(kk1),y_{0}(kk1),…y_{0}(k+N 1  k1) represent the k1 moment to k, k+1 ..., the initial prediction in k+N1 moment, A_{0}The matrix set up for step response data, Δ u (k1) is the input controlling increment in k1 moment.
2.. model predictive error value e (k) of calculating k moment controlled device:
Ess (k)=y (k)y_{1}(kk1)
Wherein, the real output value of the controlled device that y (k) the expression k moment records, y_{1}(k  k1) represents and adds control After increment Delta u (k1), controlled device is model predication value to the k moment in the k1 moment.
3.. calculate correction value y of k moment model output_{cor}(k):
y_{cor}(k)=y_{M}(k1)+h*ess(k)
Wherein,
y_{cor}(kk),y_{cor}(k+1k),…y_{cor}(k+N1  k) represent controlled device repairing at k moment forecast model respectively On the occasion of, h is the weight matrix of error compensation, and α is error correction coefficient.
4.. calculate initial communication value y of the model prediction in k moment_{M}(k):
y_{M}(k)=Sy_{cor}(k)
Wherein, S is the statetransition matrix on N × N rank,
C. calculating controlled device is M continuous print controlling increment Δ u (k) ..., the prediction output valve under Δ u (k+M1) y_{PM}, concrete grammar is:
y_{PM}(k)=y_{p0}(k)+AΔu_{M}(k)
Wherein,
y_{M}(k+1k),y_{M}(k+2k),…,y_{M}(k+P  k) be the k moment to k+1, k+2 ..., the model prediction in k+P moment Output valve, y_{0}(k+1k),y_{0}(k+2k1),…y_{0}(k+N  k) represent the k moment to k+1, k+2 ... the initial predicted in k+P moment Value.
D. making control time domain M=1 of controlled device, choose object function J (k) of controlled device, J (k) form is as follows:
Ref (k)=[ref_{1}(k),ref_{2}(k),…,ref_{P}(k)]^{Τ}
Q=diag (q_{1},q_{2},…q_{P})
R=diag (r_{1},r_{2},…r_{M})
ref_{i}(k)=β^{i}y(k)+(1β^{i})c(k)
Wherein, Q is error weighting matrix, q_{1},q_{2},…,q_{P}Weight coefficient for weighting matrix；β is softening coefficient, c (k) Setting value for process object；R is for controlling weighting matrix, r_{1},r_{2},…r_{M}For controlling the weight coefficient of weighting matrix, ref (k) is The reference locus of system, ref_{i}K () is the value of ith reference point in reference locus.
E. controlled quentity controlled variable u (k) is converted:
E (k)=c (k)y (k)
U (k)=u (k1)+K_{p}(k)(e(k)e(k1))+K_{i}(k)e(k)K_{f}(k)(y(k)y(k1)Kd(y(k)2y (k1)+y (k2))=u (k1)+K_{p}(k)(e(k)e(k1))+K_{i}(k)e(k)K_{f}(k)(y(k)y(k1)Kd(y(k)y (k1))+Kd(y(k1)y(k2))
U (k) is processed further, can obtain
U (k)=u (k1)+w (k)^{Τ}E(k)
Wherein,
W (:, k)=[K_{p}(k)+K_{i}(k),K_{p}(k),K_{f}(k)K_{d}(k),K_{d}(k)]^{Τ}
E (k)=(e (k), e (k1), y (k)y (k1), y (k1)y (k2))^{Τ}
Kp(k)、K_{i}(k)、K_{f}(k)、K_{d}K () is respectively ratio of k moment PIPD controller outer shroud, the integration of outer shroud, interior The ratio of ring, the differential parameter of internal ring, e (k) is the error between k moment reference locus value and real output value, and Τ is matrix Transposition symbol, w (:, k) be four row k column matrix.
F. the object function being updated in step d by u (k) solves the parameter in PIPD controller, can obtain:
Can obtain further:
K_{p}(k)=w (1, k)+w (2, k)
K_{i}(k)=w (2, k)
K_{f}(k)=w (3, k)w (4, k)
K_{d}(k)=w (4, k)
G. parameter K of PIPD controller is obtained_{p}(k)、K_{i}(k)、K_{f}(k)、K_{d}K () constitutes controlled quentity controlled variable u (k) later and acts on Controlled device
U (k)=u (k1)+K_{p}(k)(e(k)e(k1))+K_{i}(k)e(k)K_{f}(k)(y(k)y(k1)K_{d}(y(k)2y (k1)+y(k2))。
H. at subsequent time, continue to solve parameter k that PIPD controller is new according to the step in b to g_{P}(k+1)、k_{i}(k+ 1)、k_{f}(k+1)、k_{d}(k+1) value, circulates successively.
The present invention proposes the PIof pressure in a kind of oil gas water horizontal threephase separator optimized based on dynamic matrix control PD control method, the method combines PIPD and controls and the good control performance of dynamic matrix control, is effectively improved biography The deficiency of system control method, also promotes development and the application of advanced control algorithm simultaneously.
Detailed description of the invention
In oil gas water horizontal threephase separator as a example by the process control of pressure:
In oil gas water horizontal threephase separator, the control of pressure is a Delay Process, and regulating measure uses control settlement indoor The aperture of air bleeding valve valve.
Step (1). set up controlled by the realtime step response data of pressure object in oil gas water horizontal threephase separator The model of object, concrete grammar is:
A. one step input signal of oil supply air water horizontal threephase separator, records its step response curve.
B. it is filtered processing by corresponding step response curve, then fits to a smooth curve, record smooth song The step response data that on line, each sampling instant is corresponding, first sampling instant is T_{s}, adjacent two sampling instants interval Time is T_{s}, sampling instant order is T_{s}、2T_{s}、3T_{s}……；Response will be at some moment t_{N}Tend to be steady after=NT, work as a_{i}(i ＞ N) and a_{N}Error and measurement error when having the identical order of magnitude, i.e. it is believed that a_{N}It is approximately equal to the steadystate value of step response.Build The model vector a of pressure object in vertical oil gas water horizontal threephase separator:
A=[a_{1},a_{2},…a_{N}]^{Τ}
Wherein Τ is the transposition symbol of matrix, a_{i}It is the step response of oil gas water horizontal threephase separator sedimentation room pressure Data, N for modeling time domain.
Step (2). the PIPD controller of pressure in design oil gas water horizontal threephase separator, concrete grammar is:
A. the model vector a utilizing step (1) b to obtain sets up the dynamic square of pressure in oil gas water horizontal threephase separator Battle array, its form is as follows:
Wherein, A is P × M rank dynamic matrix of pressure in oil gas water horizontal threephase separator, and P is that dynamic matrix control is calculated The optimization time domain of method, M is the control time domain of Dynamic array control algorithm, M ＜ P ＜ N.
B. pressure is set up in oil gas water horizontal threephase separator at the initial model predictive value y in current k moment_{M}(k)
1.. calculating the k1 moment adds the model of pressure in controlling increment Δ u (k1) oil gas water horizontal threephase separator afterwards Predictive value y_{p}(k1):
y_{P}(k1)=y_{M}(k1)+A_{0}Δu(k1)
Wherein,
y_{1}(kk1),y_{1}(k+1k1),…,y_{1}(k+N1  k1) represent that oil gas water horizontal threephase separator is intrinsic pressure respectively Power in the k1 moment to k, k+1 ..., the k+N1 moment adds the model predication value after Δ u (k1), y_{0}(kk1),y_{0}(kk 1),…y_{0}(k+N1  k1) represent in oil gas water horizontal threephase separator pressure in the k1 moment to k, k+1 ..., the k+N1 moment Initial prediction, A_{0}For the matrix set up by oil gas water horizontal threephase separator sedimentation room pressure step response data, Δ u (k1) it is the controlling increment of air bleeding valve valve opening in k1 moment oil gas water horizontal threephase separator.
2.. model predictive error value ess (k) of pressure in calculating k moment oil gas water horizontal threephase separator:
Ess (k)=y (k)y_{1}(kk1)
Wherein, the real output value of pressure, y in the oil gas water horizontal threephase separator that y (k) the expression k moment records_{1}(k K1) represent add controlling increment Δ u (k1) after, in oil gas water horizontal threephase separator pressure in the k1 moment to the k moment Model predication value.
3.. calculate correction value y that the pressure model in k moment oil gas water horizontal threephase separator exports_{cor}(k):
y_{cor}(k)=y_{M}(k1)+h*ess(k)
Wherein,
y_{cor}(kk),y_{cor}(k+1k),…y_{cor}(k+N1  k) represents the pressure in oil gas water horizontal threephase separator respectively Power is in the correction value of k moment model, and h is the weight matrix of error compensation, and α is error correction coefficient.
4.. calculate model prediction initial communication value y in the k moment of the pressure in oil gas water horizontal threephase separator_{M}(k):
y_{M}(k)=Sy_{cor}(k)
Wherein, S is the statetransition matrix on N × N rank,
C. the pressure in calculating oil gas water horizontal threephase separator is M continuous print controlling increment Δ u (k) ..., Δ u (k+ M1) prediction output valve y under_{PM}, concrete grammar is:
y_{PM}(k)=y_{P0}(k)+AΔu_{M}(k)
Wherein,
y_{P0}K () is y_{M}The front P item of (k), y_{M}(k+1k),y_{M}(k+2k),…,y_{M}(k+P  k) it is the horizontal threephase separate of oil gas water Pressure in device in the k moment to k+1, k+2 ..., the model prediction output valve in k+P moment.
D. order controls time domain M=1, and chooses object function J (k) of pressure, J (k) in oil gas water horizontal threephase separator Form is as follows:
Ref (k)=[ref_{1}(k),ref_{2}(k),…,ref_{P}(k)]^{Τ}
Q=diag (q_{1},q_{2},…q_{P})
R=diag (r_{1},r_{2},…r_{M})
ref_{i}(k)=β^{i}y(k)+(1β^{i})c(k)
Wherein, Q is error weighting matrix, q_{1},q_{2},…,q_{P}Weight coefficient for error weighting matrix；β is softening coefficient, c K () is the setting value of pressure in oil gas water horizontal threephase separator；R=diag (r_{1},r_{2},…r_{M}) for controlling weighting matrix, r_{1}, r_{2},…r_{M}For controlling the weight coefficient of weighting matrix;Ref (k) is the reference of pressure in k moment oil gas water horizontal threephase separator Track, ref_{i}K () is the value of ith reference point in reference locus.
E. controlled quentity controlled variable u (k) of air bleeding valve valve opening in oil gas water horizontal threephase separator is converted:
E (k)=c (k)y (k)
U (k)=u (k1)+K_{p}(k)(e(k)e(k1))+K_{i}(k)e(k)K_{f}(k)(y(k)y(k1)Kd(y(k)2y (k1)+y (k2))=u (k1)+K_{p}(k)(e(k)e(k1))+K_{i}(k)e(k)K_{f}(k)(y(k)y(k1)Kd(y(k)y (k1))+Kd(y(k1)y(k2))
U (k) is processed further, can obtain
U (k)=u (k1)+w (k)^{Τ}E(k)
Wherein,
W (:, k)=[K_{p}(k)+K_{i}(k),K_{p}(k),K_{f}(k)K_{d}(k),K_{d}(k)]
E (k)=(e (k), e (k1), y (k)y (k1), y (k1)y (k2))^{Τ}
Kp(k)、K_{i}(k)、K_{f}(k)、K_{d}K () is respectively the ratio of PIPD controller outer shroud, the integration of outer shroud, the ratio of internal ring Example, the differential parameter of internal ring, e (k) is the error between k moment reference locus value and real output value, and Τ is the transposition of matrix Symbol, w (:, k) it is four row k column matrix.
F. u (k) is updated in the object function in step d, solves the parameter in PIPD controller, can obtain:
Can obtain further:
K_{p}(k)=w (1, k)+w (2, k)
K_{i}(k)=w (2, k)
K_{f}(k)=w (3, k)w (4, k)
K_{d}(k)=w (4, k)
G. parameter K of PIPD controller is obtained_{p}(k)、K_{i}(k)、K_{f}(k)、K_{d}K () constitutes controlled quentity controlled variable u (k) later and acts on Oil gas water horizontal threephase separator
U (k)=u (k1)+K_{p}(k)(e(k)e(k1))+K_{i}(k)e(k)K_{f}(k)(y(k)y(k1)Kd(y(k)2y (k1)+y(k2))
H. at subsequent time, continue to solve parameter k that PIPD controller is new according to the step in b to g_{P}(k+1)、k_{i}(k+ 1)、k_{f}(k+1)、k_{d}(k+1) value, acts on controlled device, and circulates successively.
Claims (1)
Priority Applications (1)
Application Number  Priority Date  Filing Date  Title 

CN201410029644.3A CN103760931B (en)  20140122  20140122  The oil gas water horizontal threephase separator compress control method that dynamic matrix control optimizes 
Applications Claiming Priority (1)
Application Number  Priority Date  Filing Date  Title 

CN201410029644.3A CN103760931B (en)  20140122  20140122  The oil gas water horizontal threephase separator compress control method that dynamic matrix control optimizes 
Publications (2)
Publication Number  Publication Date 

CN103760931A CN103760931A (en)  20140430 
CN103760931B true CN103760931B (en)  20160914 
Family
ID=50528185
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

CN201410029644.3A CN103760931B (en)  20140122  20140122  The oil gas water horizontal threephase separator compress control method that dynamic matrix control optimizes 
Country Status (1)
Country  Link 

CN (1)  CN103760931B (en) 
Families Citing this family (2)
Publication number  Priority date  Publication date  Assignee  Title 

CN105955014A (en) *  20160511  20160921  杭州电子科技大学  Method for controlling coke furnace chamber pressure based on distributed dynamic matrix control optimization 
CN106444388A (en) *  20161206  20170222  杭州电子科技大学  Distributed PID type dynamic matrix control method for furnace pressure of coke furnace 
Citations (7)
Publication number  Priority date  Publication date  Assignee  Title 

US5460196A (en) *  19920609  19951024  Technolog Limited  Fluid supply pressure control method and apparatus 
CN2455329Y (en) *  20001123  20011024  中国石化集团河南石油勘探局勘察设计研究院  Three phase separator for oil, gas and water 
CN201143393Y (en) *  20071120  20081105  中国石油天然气集团公司  Lowmultiple hypergravity oil, gas and water threephase separator 
CN101457264A (en) *  20081229  20090617  杭州电子科技大学  Blast furnace temperature optimization control method 
CN101495712A (en) *  20060525  20090729  霍尼韦尔国际公司  System and method for multivariable control in threephase separation oil and gas production 
CN103116283A (en) *  20130118  20130522  杭州电子科技大学  Method for controlling dynamic matrix of nonselfbalance object 
CN103389746A (en) *  20130719  20131113  杭州电子科技大学  Prediction function control optimized control method for furnace pressure of waste plastic oil refining cracking furnace 

2014
 20140122 CN CN201410029644.3A patent/CN103760931B/en active IP Right Grant
Patent Citations (7)
Publication number  Priority date  Publication date  Assignee  Title 

US5460196A (en) *  19920609  19951024  Technolog Limited  Fluid supply pressure control method and apparatus 
CN2455329Y (en) *  20001123  20011024  中国石化集团河南石油勘探局勘察设计研究院  Three phase separator for oil, gas and water 
CN101495712A (en) *  20060525  20090729  霍尼韦尔国际公司  System and method for multivariable control in threephase separation oil and gas production 
CN201143393Y (en) *  20071120  20081105  中国石油天然气集团公司  Lowmultiple hypergravity oil, gas and water threephase separator 
CN101457264A (en) *  20081229  20090617  杭州电子科技大学  Blast furnace temperature optimization control method 
CN103116283A (en) *  20130118  20130522  杭州电子科技大学  Method for controlling dynamic matrix of nonselfbalance object 
CN103389746A (en) *  20130719  20131113  杭州电子科技大学  Prediction function control optimized control method for furnace pressure of waste plastic oil refining cracking furnace 
Also Published As
Publication number  Publication date 

CN103760931A (en)  20140430 
Similar Documents
Publication  Publication Date  Title 

Mohanty et al.  Differential evolution algorithm based automatic generation control for interconnected power systems with nonlinearity  
CN107023825B (en)  Fluidizedbed combustion boiler control and combustion optimizing system  
King et al.  The application of fuzzy control systems to industrial processes  
Munteanu et al.  Optimization of variable speed wind power systems based on a LQG approach  
CN102777879B (en)  Main steam temperature predictive control method of ultra supercritical unit of thermal power plant  
CN102494336B (en)  Combustion process multivariable control method for CFBB (circulating fluidized bed boiler)  
CN101446807B (en)  Realization method for heatengine plant speed regulating system model in power system simulation  
CN103529869B (en)  Pressure control device, volume control device, compress control method and flow control methods  
CN104317348B (en)  Based on photovoltaic battery panel maximum power tracking method and the system of particle cluster algorithm  
CN104279713B (en)  A kind of airconditioner control method, system and airconditioner controller  
CN103972908A (en)  Multitarget reactive power optimization method based on adaptive chaos particle swarm algorithm  
AU2012200304A1 (en)  Reactive voltage control system and method for wind power field of doublefed wind powergenerating unit  
CN103280821B (en)  Multiperiod dynamic reactive power optimization method of intelligent power distribution system  
Georgiou et al.  Multicriteria decision making on the energy supply configuration of autonomous desalination units  
CN103838920B (en)  A kind of turboshaft engine dynamic realtime model modelling approach and system  
CN102588211A (en)  Model prediction control method and model prediction control system for all working conditions of wind generating set  
CN101206754A (en)  Method for optimizing distribution of thermal power station load based on a plurality of restriction rules  
CN1060915A (en)  The universal process control of using artificial neural networks  
CN103968367B (en)  Based on the method for the general predictive control that Fuzzy Neural PID controls  
CN102562446A (en)  Method for operating a pitchcontrolled wind turbine  
CN103309371B (en)  Intermittent reactor control system based on control variable parameterization method  
CN103104931B (en)  A kind of primary frequency modulation method based on boiler primary air pressure dynamic compensation and system  
CN102444784B (en)  Pressure control system for steel enterprise steam pipe network based on dynamic matrix control  
CastilloGarcia et al.  Frequency specifications regions of fractionalorder PI controllers for first order plus time delay processes  
CN102562561B (en)  The pump group operational energy efficiency analytical procedure of industrial circulating water system 
Legal Events
Date  Code  Title  Description 

PB01  Publication  
C06  Publication  
SE01  Entry into force of request for substantive examination  
C10  Entry into substantive examination  
GR01  Patent grant  
C14  Grant of patent or utility model 