CN103345160A - Humidity optimization control method for waste heat drying system - Google Patents

Humidity optimization control method for waste heat drying system Download PDF

Info

Publication number
CN103345160A
CN103345160A CN2013102811637A CN201310281163A CN103345160A CN 103345160 A CN103345160 A CN 103345160A CN 2013102811637 A CN2013102811637 A CN 2013102811637A CN 201310281163 A CN201310281163 A CN 201310281163A CN 103345160 A CN103345160 A CN 103345160A
Authority
CN
China
Prior art keywords
control
value
humidity
expression
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.)
Granted
Application number
CN2013102811637A
Other languages
Chinese (zh)
Other versions
CN103345160B (en
Inventor
薛安克
陈云
周绍生
孔亚广
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Xinghai Energy Conservation and Environmental Protection Technology Co.,Ltd.
Original Assignee
Hangzhou Dianzi University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Dianzi University filed Critical Hangzhou Dianzi University
Priority to CN201310281163.7A priority Critical patent/CN103345160B/en
Publication of CN103345160A publication Critical patent/CN103345160A/en
Application granted granted Critical
Publication of CN103345160B publication Critical patent/CN103345160B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Feedback Control In General (AREA)
  • Drying Of Solid Materials (AREA)

Abstract

The invention discloses a humidity optimization control method for a waste heat drying system. At first, a model prediction control method is adopted to control the humidity value of gas of an outlet of the waste heat drying system, the humidity value of the gas is made to be stably transited to a set value according to a set reference track, and the situation that due to the fact that the control input amount is overlarge, dry goods spontaneously combust and potential safety hazards of the system are caused is avoided; in consideration of the influence of disturbance of the input control amount of the system, feed-forward compensation is added to form a feed-forward and feedback composite control, a prediction model is rectified, system disturbance input response is quickened, and the prediction error of the system is reduced at the same time; after the humidity value reaches the set value, an LG optimization control method is adopted to perform optimization control over the system, the capacity of resisting disturbance by the system is increased, and the gas consumption of the system is made to be least. The humidity optimization control method reduces the influence on the system by the input disturbance, improves the control precision, enables the energy consumption of the system to be least, and meets the requirements of industrial production for energy conservation.

Description

The method for optimally controlling humidity of waste heat drying system
Technical field
The invention belongs to areas of information technology, relate to the model predictive control method (MAC in the automatic technology, model algorithm control), feedforward and Feedback control method and LQ method for optimally controlling, be based on above several method the humidity of waste heat drying system be optimized control.
Background technology
Waste plastic oil-refining be will life and industrial processes in the waste plastics that produces carry out the new technology that cracking is refined oil.For guaranteeing efficient and the safety of cracking process, before being carried out cracking, waste plastics must dry processing to waste plastics.Because waste plastics is combustible and wherein also is being mingled with inflammable object such as waste paper, avoids the drying article spontaneous combustion for improving drying efficiency, adopts hot-air to mix the mode of drying with waste plastics when oven dry as far as possible.The efficient of this mixing oven dry and the utilization factor of combustion gas depended on the humidity value of oven dry back outlet flue gas, in this drying system when exporting the smoke moisture value and reach 50% drying efficiency and combustion gas utilization factor reach the highest.Yet the control algolithm in the similar drying system control in the past reckons without the influence of smoke moisture, thereby often makes the drying effect of system bad, and cracking waste plastics efficient is not high, and causes the combustion gas waste, does not meet energy-saving and cost-reducing requirement.
Summary of the invention
Purpose of the present invention is exactly at the influence of not considering drying system middle outlet smoke moisture value in the drying system in the past, makes drying efficiency not high, causes combustion gas waste and method for optimally controlling humidity in a kind of drying system of proposing.
The inventive method is the mixture model method for handover control that a kind of model predictive control method (MAC, model algorithm control), feedforward and Feedback control method and LQ method for optimally controlling combine.At the inflammable characteristic of drying article, at first adopt model predictive control method (MAC, model algorithm control) the outlet smoke moisture value of waste heat drying system is controlled, make humidity value reach 50% stably according to given reference locus, avoid that input quantity is excessive causes the drying article spontaneous combustion to cause security of system hidden danger because of control; Consider the influence of system's input controlled quentity controlled variable disturbance, add the compound control that feedforward compensation constitutes feedforward and Feedback, forecast model is proofreaied and correct, reduce the predicated error of system when accelerating system disturbance input response; After humidity value reaches 50%, adopt the LQ method for optimally controlling that system is carried out optimization control, make system's gas consumption minimum when strengthening system's antijamming capability.
The concrete steps of the inventive method comprise:
Step 1: set up the forecast model of system, with model predictive control method (MAC, model algorithm control) and feedforward and Feedback control method, the outlet smoke moisture of control waste heat drying system makes humidity value reach 50% stably.
A. set up forecast model
Concrete grammar is: with the charging aperture flow, and gas valve aperture value, cold air distribution valve opening value is input quantity, the drying system outlet smoke moisture value that collects with humidity sensor is output quantity, sets up the discrete differential model based on least square method;
Figure 2013102811637100002DEST_PATH_IMAGE002
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE004
Expression drying system gas outlet air humidity value,
Figure 2013102811637100002DEST_PATH_IMAGE006
Expression control input vector,
Figure 2013102811637100002DEST_PATH_IMAGE008
The input disturbance vector of expression system; Control input vector and disturbance input vector are
Figure 2013102811637100002DEST_PATH_IMAGE010
，
Figure 2013102811637100002DEST_PATH_IMAGE012
Wherein Expression combustion gas two-port valve valve opening value,
Figure 2013102811637100002DEST_PATH_IMAGE016
Represent cold air distribution two-port valve valve opening value,
Figure 2013102811637100002DEST_PATH_IMAGE018
Expression charging aperture flow can be by regulating in drying course
Figure 780566DEST_PATH_IMAGE014
,
Figure 859380DEST_PATH_IMAGE016
,
Figure 427197DEST_PATH_IMAGE018
Three control input quantities are controlled air humidity in the drying system.
Figure 2013102811637100002DEST_PATH_IMAGE020
,
Figure 2013102811637100002DEST_PATH_IMAGE022
, Be respectively the disturbance input of three input quantities.
Figure 2013102811637100002DEST_PATH_IMAGE026
,
Figure 2013102811637100002DEST_PATH_IMAGE028
With
Figure 2013102811637100002DEST_PATH_IMAGE030
The systematic parameter matrix that expression obtains by the least square identification
Figure 2013102811637100002DEST_PATH_IMAGE032
， ，
Figure 2013102811637100002DEST_PATH_IMAGE036
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE038
Be scalar parameter to be identified,
Figure 2013102811637100002DEST_PATH_IMAGE040
,
Figure 2013102811637100002DEST_PATH_IMAGE042
For to be identified
Figure 2013102811637100002DEST_PATH_IMAGE044
The dimension matrix, nBe sampling number.
By
Figure 2013102811637100002DEST_PATH_IMAGE046
Inverse transformation, the non-parametric model of above-mentioned model conversation one-tenth based on the impulse response transport function, i.e. drying system outlet smoke moisture value prediction model:
Figure 2013102811637100002DEST_PATH_IMAGE048
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE050
Expression the
Figure 2013102811637100002DEST_PATH_IMAGE052
The humidity predicted value of individual sampling instant outlet flue gas,
Figure 2013102811637100002DEST_PATH_IMAGE054
Be the modeling time domain,
Figure 2013102811637100002DEST_PATH_IMAGE056
Expression the
Figure 2013102811637100002DEST_PATH_IMAGE058
Control input variable constantly,
Figure 2013102811637100002DEST_PATH_IMAGE060
Expression the Input disturbance constantly.Wherein
Figure 2013102811637100002DEST_PATH_IMAGE062
With
Figure 2013102811637100002DEST_PATH_IMAGE064
Computing formula as follows
Figure 2013102811637100002DEST_PATH_IMAGE066
，
Figure 2013102811637100002DEST_PATH_IMAGE070
,?
Figure 70854DEST_PATH_IMAGE068
Expression is done the formula in the bracket
Figure 34917DEST_PATH_IMAGE046
Inverse transformation.
B. Feedforward Controller Design
By forecast model
Figure 184139DEST_PATH_IMAGE002
Have after the arrangement
Figure 2013102811637100002DEST_PATH_IMAGE074
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE076
Transport function when not having disturbance for controlled device,
Figure 2013102811637100002DEST_PATH_IMAGE078
Be the transport function of controlled device disturbance passage, and have
Figure 2013102811637100002DEST_PATH_IMAGE080
，
The output of system after the adding feedforward control link
Figure 2013102811637100002DEST_PATH_IMAGE084
Become
Figure 2013102811637100002DEST_PATH_IMAGE086
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE088
Transport function for feedforward controller.When
Figure 2013102811637100002DEST_PATH_IMAGE090
In time, have
Order
Figure 2013102811637100002DEST_PATH_IMAGE094
, the transport function that can get feedforward controller is
Figure 2013102811637100002DEST_PATH_IMAGE096
Thus, feedforward controller is output as
Figure 2013102811637100002DEST_PATH_IMAGE098
C. adopt the feedforward-feedback control method that system is proofreaied and correct, realize the closed loop prediction
Concrete grammar is as follows: with Walk out of the actual value of mouthful smoke moisture
Figure 625222DEST_PATH_IMAGE084
With this humidity predicted value constantly
Figure 2013102811637100002DEST_PATH_IMAGE102
It is poor to do, and can be somebody's turn to do error amount constantly
Figure 2013102811637100002DEST_PATH_IMAGE104
Utilize this error to
Figure 2013102811637100002DEST_PATH_IMAGE106
The prediction output of individual sampling instant
Figure 938523DEST_PATH_IMAGE050
Carry out feedback modifiers, obtain proofreading and correct back the
Figure 359140DEST_PATH_IMAGE106
The prediction of output value of individual sampling instant
Figure 2013102811637100002DEST_PATH_IMAGE108
For
Figure 2013102811637100002DEST_PATH_IMAGE110
，
Figure 2013102811637100002DEST_PATH_IMAGE112
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE114
Be the error correction coefficient,
Figure 2013102811637100002DEST_PATH_IMAGE116
Be the prediction time domain.
In order to make the setting value that reaches of humidity value safety and steady, with the designated reference track in the model predictive control method (MAC, model algorithm control)
Figure 2013102811637100002DEST_PATH_IMAGE118
The value of individual sampling instant is taken as
Figure 2013102811637100002DEST_PATH_IMAGE120
，
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE124
Be the output setting value;
Figure 2013102811637100002DEST_PATH_IMAGE126
Be given constant of reference time;
Figure 2013102811637100002DEST_PATH_IMAGE128
Be the sampling period.The time constant of reference locus
Figure 398956DEST_PATH_IMAGE126
Be worth more greatly, then the flexibility of system is more strong, and robustness is more strong, but the rapidity variation of control.
D. the optimal control law of model predictive control method (MAC, model algorithm control) is calculated
Select the quadratic performance index of output error and controlled quentity controlled variable weighting, it is expressed as follows:
Figure 2013102811637100002DEST_PATH_IMAGE130
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE132
, Be respectively the weighting coefficient of prediction output error and controlled quentity controlled variable,
Figure 2013102811637100002DEST_PATH_IMAGE136
For
Figure 2013102811637100002DEST_PATH_IMAGE138
Constantly total control input,
Figure 2013102811637100002DEST_PATH_IMAGE140
For
Figure 527449DEST_PATH_IMAGE138
Feedforward controller output constantly,
Figure 299227DEST_PATH_IMAGE116
Be the prediction time domain,
Figure 2013102811637100002DEST_PATH_IMAGE142
Be the control time domain, and have
Being write quadratic performance index as vector form has
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE148
With
Figure 2013102811637100002DEST_PATH_IMAGE150
Be the weighting battle array, generally can choose the diagonal matrix form
Figure 2013102811637100002DEST_PATH_IMAGE152
,
Figure 2013102811637100002DEST_PATH_IMAGE154
, the element on the diagonal line
Figure 2013102811637100002DEST_PATH_IMAGE156
,
Figure 2013102811637100002DEST_PATH_IMAGE158
Can choose suitable value according to the system of reality. Be total control input vector;
Figure 2013102811637100002DEST_PATH_IMAGE162
Be the reference input vector;
Figure 2013102811637100002DEST_PATH_IMAGE164
Be prediction of output value vector.
To the unknown control vector
Figure 2013102811637100002DEST_PATH_IMAGE166
Differentiate, even
Figure 2013102811637100002DEST_PATH_IMAGE168
, just can draw optimal control law
Figure 2013102811637100002DEST_PATH_IMAGE170
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE172
Be the error correction vector,
Figure 2013102811637100002DEST_PATH_IMAGE174
Value general desirable 1.
,
Figure 2013102811637100002DEST_PATH_IMAGE178
Wherein , For The dimension matrix, can once calculate simultaneously from
Figure 799872DEST_PATH_IMAGE100
Arrive All controlled quentity controlled variables constantly,
Figure 897272DEST_PATH_IMAGE062
Calculating in step a, provide.
In actual applications, in order to reduce error as far as possible, the present invention adopts closed loop control algorithm, namely only carries out the control action of current time
Figure 2013102811637100002DEST_PATH_IMAGE188
, and next moment controlled quentity controlled variable
Figure 2013102811637100002DEST_PATH_IMAGE190
Press again
Figure 185165DEST_PATH_IMAGE188
Calculating formula recursion one step rerun.At this moment, the instant controlled quentity controlled variable of optimum control can be written as
Figure 2013102811637100002DEST_PATH_IMAGE192
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE194
Step 2: after drying system outlet smoke moisture value reaches 50%, adopt the LQ method for optimally controlling that system is carried out optimization control.
Concrete grammar is as follows:
Initial time serves as the control input quantity with charging aperture flow, gas valve aperture value, cold air distribution valve opening value, utilizes the forecast Control Algorithm of step 1, and the smoke moisture in the waste heat drying system is controlled.In order to make the amount of consumed gas minimum, after outlet smoke moisture value reaches 50%, serve as single control input quantity with the burnt gas valve aperture, realize the optimum control of system by regulating the burnt gas valve opening value, reach purpose of energy saving.
By the forecast model of setting up previously through the state-space expression that can get system after the model conversion be:
Figure 2013102811637100002DEST_PATH_IMAGE196
，
Wherein
Figure 2013102811637100002DEST_PATH_IMAGE200
,
Figure 2013102811637100002DEST_PATH_IMAGE202
Be the outlet smoke moisture,
Figure 2013102811637100002DEST_PATH_IMAGE204
Be the charging aperture flow,
Figure 2013102811637100002DEST_PATH_IMAGE206
Be cold air distribution valve opening value,
Figure 2013102811637100002DEST_PATH_IMAGE208
Be the burnt gas valve opening value,
Figure 2013102811637100002DEST_PATH_IMAGE210
Be the state value of initial time,
Figure 2013102811637100002DEST_PATH_IMAGE212
With
Figure 2013102811637100002DEST_PATH_IMAGE214
Be the constant matrices that obtains by identification,
Figure 2013102811637100002DEST_PATH_IMAGE216
The expression theorem in Euclid space.
Choose quadratic performance index
Figure 2013102811637100002DEST_PATH_IMAGE218
Wherein
Figure 711699DEST_PATH_IMAGE116
Be the prediction time domain, S, Q d , R d For
Figure 2013102811637100002DEST_PATH_IMAGE220
The real symmetrical weighting matrix of the positive definite of dimension is generally chosen diagonal matrix ,
Figure 2013102811637100002DEST_PATH_IMAGE224
,
Figure 2013102811637100002DEST_PATH_IMAGE226
, the element on the diagonal line ,
Figure 2013102811637100002DEST_PATH_IMAGE230
,
Figure 2013102811637100002DEST_PATH_IMAGE232
Can suitably choose according to system's actual motion characteristic.
According to matrix in the state equation AWith BValue, the Riccati matrix equation of finding the solution discrete-time system
Figure 2013102811637100002DEST_PATH_IMAGE234
Wherein IExpression The unit matrix of dimension can solve positive definite symmetric matrices W( k).Can design discrete time LQ optimal controller thus, the optimum humidity control law of corresponding feedback of status is
Figure 2013102811637100002DEST_PATH_IMAGE236
This method for optimally controlling will make that moisture control system has
Figure 2013102811637100002DEST_PATH_IMAGE238
Gain margin and
Figure 2013102811637100002DEST_PATH_IMAGE240
Phase margin.
The present invention be directed in this mixing drying system that drying article contains fire goods and traditional oven dry control algolithm is not considered the drying system smoke moisture, and make system's drying efficiency and fuel availability not high, and the drying system method for optimally controlling humidity that proposes.The present invention adopts feedforward and Feedback control and Model Predictive Control (MAC, model algorithm control) closed loop control method that combines, can be according to the control requirement, constantly carry out feedback compensation, finally obtain optimum valve opening value, the feedforward compensation link can be accelerated system the response of input disturbance is reduced error.When the oven dry beginning, adopt model predictive control method (MAC, model algorithm control), can make the output of system reach designated value gradually according to intended trajectory, eliminate the potential safety hazard that causes the drying article spontaneous combustion because valve opening is excessive, make system's transitional processes steady simultaneously.After system's humidity reaches designated value, adopt the LQ method for optimally controlling to control, when guaranteeing system stability, can make amount of consumed gas minimum.Adopt feedforward and Feedback composite control method, model predictive control method (MAC, model algorithm control) and the mixing control method of LQ method for optimally controlling combination, can eliminate the potential safety hazard of crossing the spontaneous combustion of ambassador's drying article because of valve opening, reduced the influence of input disturbance to system, improved control accuracy, make system's power consumption minimum simultaneously, satisfy industrial energy-saving safe requirement.
Embodiment
Waste heat drying system method for optimally controlling humidity, adopt feedforward and Feedback control method and model predictive control method (MAC, model algorithm control) to combine valve opening and the feed rate of combustion gas two-port valve, cold air distribution two-port valve are controlled at the oven dry initial stage.Specific implementation method is as follows:
1) determines forecast model In the value of each parameter matrix
Adopt the least square discrimination method of many input-multiple output systems that systematic parameter is carried out identification, systematic parameter is
Figure 662404DEST_PATH_IMAGE032
，
Figure DEST_PATH_IMAGE242
，
Figure DEST_PATH_IMAGE244
Wherein
Figure DEST_PATH_IMAGE246
, ,
Figure DEST_PATH_IMAGE250
Be sampling number, ,
Figure 280915DEST_PATH_IMAGE040
,
Figure 641489DEST_PATH_IMAGE042
Estimated value can be by formula
Figure DEST_PATH_IMAGE252
Nothing partially estimate to obtain the parameter matrix of Ce Lianging wherein
Figure DEST_PATH_IMAGE254
With the output valve matrix of measuring
Figure DEST_PATH_IMAGE256
Specifically be calculated as follows
Figure DEST_PATH_IMAGE258
Figure DEST_PATH_IMAGE262
Wherein Be the sampling time domain
Figure 776411DEST_PATH_IMAGE054
One group of sampled value of inner control input quantity,
Figure DEST_PATH_IMAGE266
Be the sampling time domain
Figure 287158DEST_PATH_IMAGE054
One group of sampled value of interior input disturbance amount,
Figure DEST_PATH_IMAGE268
Be the sampled value that a group system of correspondence control input quantity and disturbance input quantity is exported, such sampled value is total
Figure 725223DEST_PATH_IMAGE250
Group. In a 1 a n , b 01 b 02 b 03 b n1 b n2 b n3 , c 01 c 02 c 03 c n1 c n2 c n3 Three groups of data are respectively in the corresponding forecast model
Figure 584595DEST_PATH_IMAGE026
,
Figure 193431DEST_PATH_IMAGE028
,
Figure 448482DEST_PATH_IMAGE030
The coefficient of every fraction.
2) calculating of determining and proofreading and correct back prediction output valve of feedforward controller parameter.Because have model mismatch and error when the modeling of reality and parameter estimation, the output valve of forecast model may depart from actual value, so need revise output.Consider the influence of input quantity disturbance, be the response speed of enhancing system to the disturbance input simultaneously, the present invention adopts feedforward-feedback control that prediction output is proofreaied and correct.
In aforesaid Feedforward Controller Design, obtained the transport function of feedforward controller
Figure 459163DEST_PATH_IMAGE096
The transport function of controlled device wherein
Figure DEST_PATH_IMAGE272
Transport function with the disturbance passage
Figure 223857DEST_PATH_IMAGE078
Be respectively
Figure 133038DEST_PATH_IMAGE080
，
Figure 844642DEST_PATH_IMAGE082
With its substitution controller transfer function, can get the transport function of feedforward controller
Figure DEST_PATH_IMAGE274
Wherein
Figure 772147DEST_PATH_IMAGE042
,
Figure 724053DEST_PATH_IMAGE040
Obtain during by System Discrimination.
Prediction output valve behind each time correction is
Figure DEST_PATH_IMAGE276
Wherein
Figure DEST_PATH_IMAGE278
Be error prediction model,
Figure DEST_PATH_IMAGE280
For
Figure 182848DEST_PATH_IMAGE100
Constantly humidity sensor is to the sampled value of outlet flue gas,
Figure 494880DEST_PATH_IMAGE102
Expression
Figure 480154DEST_PATH_IMAGE100
Constantly export the predicted value of smoke moisture,
Figure 789912DEST_PATH_IMAGE172
Be the error correction vector, generally desirable
Figure 936335DEST_PATH_IMAGE174
Value be 1.
Given reference locus
Figure 989742DEST_PATH_IMAGE120
,
Figure 626260DEST_PATH_IMAGE122
, wherein
Figure 106920DEST_PATH_IMAGE124
Be the output setting value;
Figure 212410DEST_PATH_IMAGE126
Be given constant of reference time;
Figure 69508DEST_PATH_IMAGE128
Be the sampling period.Prediction output valve and reference locus behind the feedback compensation are compared, draw the predicated error scope.If predicated error is bigger, then readjust combustion gas two-port valve, cold air distribution two-port valve valve opening value and charging aperture flow, up to predicated error is controlled in certain scope.
3) given weighting matrix
Figure 826111DEST_PATH_IMAGE152
,
Figure 290722DEST_PATH_IMAGE154
, the element on the diagonal line
Figure DEST_PATH_IMAGE284
,
Figure DEST_PATH_IMAGE286
Figure DEST_PATH_IMAGE288
Can require to choose suitable value according to reality, work as predicated error When being controlled in certain scope, more easy when making the by-pass valve control aperture, the present invention adopts the mode of output control increment that valve opening is controlled.
Set up the quadratic performance index of prediction of output sum of errors controlled quentity controlled variable weighting, be described below:
Figure DEST_PATH_IMAGE292
Wherein
Figure DEST_PATH_IMAGE294
,
Figure DEST_PATH_IMAGE296
Be control input increment, and
Figure DEST_PATH_IMAGE298
Wherein
Figure DEST_PATH_IMAGE300
Order
Figure DEST_PATH_IMAGE302
, have behind the abbreviation
Figure DEST_PATH_IMAGE304
With following formula launch to try to achieve from
Figure 852151DEST_PATH_IMAGE100
Arrive
Figure DEST_PATH_IMAGE306
Constantly carry out the increment of order open loop control
Figure DEST_PATH_IMAGE308
,
Figure DEST_PATH_IMAGE310
...,
Figure DEST_PATH_IMAGE312
, namely
Figure DEST_PATH_IMAGE314
In the formula
Above computation process can be realized that all to input, output data sampling, sampled data is kept in the database of having built computing machine by the AD modular converter by computer program.Calculate the optimum control increment by computing machine, by the DA modular converter working control input quantity is regulated, thus the humidity of control waste heat drying system.
Set up performance index and find the solution optimal control law, can make the error minimum of forecast model, calculate by optimal control law, the control input quantity that can obtain minimum is the minimal valve aperture in the native system, thereby reaches purpose of energy saving.
4) after the outlet smoke moisture value of drying system reaches 50%, system is switched to the LQ method for optimally controlling system is carried out optimum control.
Initial time serves as the control input quantity with charging aperture flow, gas valve aperture value, cold air distribution valve opening value, utilizes the forecast Control Algorithm of step 1, and the smoke moisture in the waste heat drying system is controlled.In order to make the amount of consumed gas minimum, after outlet smoke moisture value reaches 50%, serve as single control input quantity with the burnt gas valve aperture, realize the optimum control of system by regulating the burnt gas valve opening value, realize purpose of energy saving.By the forecast model of setting up previously through the state-space expression that can get system after the model conversion be:
Figure DEST_PATH_IMAGE318
，
Wherein
Figure DEST_PATH_IMAGE320
, Be the outlet smoke moisture,
Figure DEST_PATH_IMAGE322
Be the charging aperture flow, Be cold air distribution valve opening value,
Figure DEST_PATH_IMAGE324
Be the burnt gas valve opening value,
Figure 745896DEST_PATH_IMAGE210
Be the state value of initial time,
Figure DEST_PATH_IMAGE325
With
Figure DEST_PATH_IMAGE326
Be the constant matrices that obtains by identification.
Choose quadratic performance index
Figure DEST_PATH_IMAGE327
Wherein
Figure 170055DEST_PATH_IMAGE116
Be the prediction time domain, S, Q d , R d For
Figure 320413DEST_PATH_IMAGE220
The real symmetrical weighting matrix of the positive definite of dimension is generally chosen diagonal matrix
Figure DEST_PATH_IMAGE328
,
Figure DEST_PATH_IMAGE329
,
Figure DEST_PATH_IMAGE330
, the element on the diagonal line
Figure DEST_PATH_IMAGE332
Figure DEST_PATH_IMAGE334
,
Figure DEST_PATH_IMAGE336
Figure DEST_PATH_IMAGE338
,
Figure DEST_PATH_IMAGE340
Figure DEST_PATH_IMAGE342
Can suitably choose according to system's actual motion characteristic.
According to matrix in the state equation AWith BValue, find the solution following discrete-time system Riccati matrix equation
Figure DEST_PATH_IMAGE343
Wherein IExpression The unit matrix of dimension can solve positive definite symmetric matrices W( k).Thus can discrete time LQ optimal controller, corresponding state feeds back optimum humidity control law and is
Figure DEST_PATH_IMAGE344
By the input and output value sampling of each AD module to system, sampled value is given LQ optimal controller, finds the solution the optimum state FEEDBACK CONTROL of output waste heat drying system humidity value through the LQ controller
Adorn the control input value that the die change piece just can be regulated real system by DA again.Utilize the LQ optimum control can strengthen the robustness of system, simultaneously optimization control is carried out in the combustion gas input of waste heat drying system, make gas consumption minimum, meet industrial energy-conservation requirement.
The present invention adopts model predictive control method (MAC, model algorithm control), and the mixture model method for handover control that feedforward and Feedback control method and LQ method for optimally controlling combine is optimized control to the humidity of waste heat drying system.This method has been eliminated the potential safety hazard that may cause the drying article spontaneous combustion because valve opening is excessive, realizes the control of humidity in the waste heat drying system fast and accurately.By the designated model performance index, calculate optimal control law and can make the control energy resource consumption of system and firing rate all reach minimum, realize the optimum control of the humidity in the waste heat drying system, reach the energy-saving and cost-reducing of enterprise.

Claims (1)

1. the method for optimally controlling humidity of waste heat drying system is characterized in that the concrete steps of this method are:
Step 1: set up the forecast model of system, with model predictive control method and feedforward and Feedback control method, the outlet smoke moisture of control waste heat drying system makes humidity value reach 50% stably;
A. set up forecast model
Concrete grammar is: with the charging aperture flow, and gas valve aperture value, cold air distribution valve opening value is input quantity, the drying system outlet smoke moisture value that collects with humidity sensor is output quantity, sets up the discrete differential model based on least square method;
Figure 2013102811637100001DEST_PATH_IMAGE002
Wherein
Figure 2013102811637100001DEST_PATH_IMAGE004
Expression drying system gas outlet air humidity value, Expression control input vector,
Figure 2013102811637100001DEST_PATH_IMAGE008
The input disturbance vector of expression system; Control input vector and disturbance input vector are
，
Figure DEST_PATH_IMAGE012
Wherein
Figure DEST_PATH_IMAGE014
Expression combustion gas two-port valve valve opening value, Represent cold air distribution two-port valve valve opening value,
Figure DEST_PATH_IMAGE018
Expression charging aperture flow can be by regulating in drying course
Figure 497538DEST_PATH_IMAGE014
,
Figure 756481DEST_PATH_IMAGE016
,
Figure 573128DEST_PATH_IMAGE018
Three control input quantities are controlled air humidity in the drying system;
Figure DEST_PATH_IMAGE020
,
Figure DEST_PATH_IMAGE022
,
Figure DEST_PATH_IMAGE024
Be respectively the disturbance input of three input quantities;
Figure DEST_PATH_IMAGE026
,
Figure DEST_PATH_IMAGE028
With
Figure DEST_PATH_IMAGE030
The systematic parameter matrix that expression obtains by the least square identification
Figure DEST_PATH_IMAGE032
，
Figure DEST_PATH_IMAGE034
，
Figure DEST_PATH_IMAGE036
Wherein Be scalar parameter to be identified,
Figure DEST_PATH_IMAGE040
,
Figure DEST_PATH_IMAGE042
For to be identified
Figure DEST_PATH_IMAGE044
The dimension matrix, nBe sampling number;
By Inverse transformation, the non-parametric model of above-mentioned model conversation one-tenth based on the impulse response transport function, i.e. drying system outlet smoke moisture value prediction model:
Figure DEST_PATH_IMAGE048
Wherein
Figure DEST_PATH_IMAGE050
Expression the
Figure DEST_PATH_IMAGE052
The humidity predicted value of individual sampling instant outlet flue gas,
Figure DEST_PATH_IMAGE054
Be the modeling time domain,
Figure DEST_PATH_IMAGE056
Expression the Control input variable constantly,
Figure DEST_PATH_IMAGE060
Expression the
Figure 363667DEST_PATH_IMAGE058
Input disturbance constantly; Wherein
Figure DEST_PATH_IMAGE062
With
Figure DEST_PATH_IMAGE064
Computing formula as follows
Figure DEST_PATH_IMAGE066
，
Figure DEST_PATH_IMAGE068
,?
Figure 824736DEST_PATH_IMAGE068
Figure DEST_PATH_IMAGE072
Expression is done the formula in the bracket
Figure 336138DEST_PATH_IMAGE046
Inverse transformation;
B. Feedforward Controller Design
By forecast model
Figure 312184DEST_PATH_IMAGE002
Have after the arrangement
Figure DEST_PATH_IMAGE074
Wherein
Figure DEST_PATH_IMAGE076
Transport function when not having disturbance for controlled device,
Figure DEST_PATH_IMAGE078
Be the transport function of controlled device disturbance passage, and have
Figure DEST_PATH_IMAGE080
，
Figure DEST_PATH_IMAGE082
The output of system after the adding feedforward control link
Figure DEST_PATH_IMAGE084
Become
Figure DEST_PATH_IMAGE086
Wherein
Figure DEST_PATH_IMAGE088
Transport function for feedforward controller; When
Figure DEST_PATH_IMAGE090
In time, have
Figure DEST_PATH_IMAGE092
Order
Figure DEST_PATH_IMAGE094
, the transport function that can get feedforward controller is
Figure DEST_PATH_IMAGE096
Thus, feedforward controller is output as
Figure DEST_PATH_IMAGE098
C. adopt the feedforward-feedback control method that system is proofreaied and correct, realize the closed loop prediction
Concrete grammar is as follows: with
Figure DEST_PATH_IMAGE100
Walk out of the actual value of mouthful smoke moisture With this humidity predicted value constantly It is poor to do, and can be somebody's turn to do error amount constantly
Figure DEST_PATH_IMAGE104
Utilize this error to
Figure DEST_PATH_IMAGE106
The prediction output of individual sampling instant
Figure 604680DEST_PATH_IMAGE050
Carry out feedback modifiers, obtain proofreading and correct back the
Figure 2164DEST_PATH_IMAGE106
The prediction of output value of individual sampling instant
Figure DEST_PATH_IMAGE108
For
，
Wherein
Figure DEST_PATH_IMAGE114
Be the error correction coefficient, Be the prediction time domain;
In order to make the setting value that reaches of humidity value safety and steady, with the designated reference track in the model predictive control method
Figure DEST_PATH_IMAGE118
Figure 575095DEST_PATH_IMAGE106
The value of individual sampling instant is taken as
Figure DEST_PATH_IMAGE120
，
Figure DEST_PATH_IMAGE122
Wherein
Figure DEST_PATH_IMAGE124
Be the output setting value;
Figure DEST_PATH_IMAGE126
Be given constant of reference time;
Figure DEST_PATH_IMAGE128
Be the sampling period; The time constant of reference locus Be worth more greatly, then the flexibility of system is more strong, and robustness is more strong, but the rapidity variation of control;
D. the optimal control law of model predictive control method is calculated
Select the quadratic performance index of output error and controlled quentity controlled variable weighting, it is expressed as follows:
Figure DEST_PATH_IMAGE130
Wherein
Figure DEST_PATH_IMAGE132
,
Figure DEST_PATH_IMAGE134
Be respectively the weighting coefficient of prediction output error and controlled quentity controlled variable,
Figure DEST_PATH_IMAGE136
For
Figure DEST_PATH_IMAGE138
Constantly total control input,
Figure DEST_PATH_IMAGE140
For Feedforward controller output constantly,
Figure 156489DEST_PATH_IMAGE116
Be the prediction time domain,
Figure DEST_PATH_IMAGE142
Be the control time domain, and have
Being write quadratic performance index as vector form has
Figure DEST_PATH_IMAGE146
Wherein
Figure DEST_PATH_IMAGE148
With Be the weighting battle array, choose the diagonal matrix form
Figure DEST_PATH_IMAGE152
,
Figure DEST_PATH_IMAGE154
, the element on the diagonal line
Figure DEST_PATH_IMAGE156
, Can choose suitable value according to the system of reality; Be total control input vector;
Figure DEST_PATH_IMAGE162
Be the reference input vector;
Figure DEST_PATH_IMAGE164
Be prediction of output value vector;
To the unknown control vector
Figure DEST_PATH_IMAGE166
Differentiate, even
Figure DEST_PATH_IMAGE168
, just can draw optimal control law
Figure DEST_PATH_IMAGE170
Wherein
Figure DEST_PATH_IMAGE172
Be the error correction vector, Value desirable 1;
Figure DEST_PATH_IMAGE176
,
Figure DEST_PATH_IMAGE178
Wherein
Figure DEST_PATH_IMAGE180
,
Figure DEST_PATH_IMAGE182
For
Figure DEST_PATH_IMAGE184
The dimension matrix, can once calculate simultaneously from
Figure 362255DEST_PATH_IMAGE100
Arrive All controlled quentity controlled variables constantly,
Figure 572656DEST_PATH_IMAGE062
Calculating in step a, provide;
In order to reduce error as far as possible, adopt closed loop control algorithm, namely only carry out the control action of current time
Figure DEST_PATH_IMAGE188
, and next moment controlled quentity controlled variable
Figure DEST_PATH_IMAGE190
Press again
Figure 534927DEST_PATH_IMAGE188
Calculating formula recursion one step rerun; At this moment, the instant controlled quentity controlled variable of optimum control can be written as
Figure DEST_PATH_IMAGE192
Wherein
Figure DEST_PATH_IMAGE194
Step 2: after drying system outlet smoke moisture value reaches 50%, adopt the LQ method for optimally controlling that system is carried out optimization control;
Concrete grammar is as follows:
Initial time serves as the control input quantity with charging aperture flow, gas valve aperture value, cold air distribution valve opening value, utilizes the forecast Control Algorithm of step 1, and the smoke moisture in the waste heat drying system is controlled; In order to make the amount of consumed gas minimum, after outlet smoke moisture value reaches 50%, serve as single control input quantity with the burnt gas valve aperture, realize the optimum control of system by regulating the burnt gas valve opening value, reach purpose of energy saving;
By the forecast model of setting up previously through the state-space expression that can get system after the model conversion be:
Figure DEST_PATH_IMAGE196
，
Figure DEST_PATH_IMAGE198
Wherein
Figure DEST_PATH_IMAGE200
,
Figure DEST_PATH_IMAGE202
Be the outlet smoke moisture,
Figure DEST_PATH_IMAGE204
Be the charging aperture flow, Be cold air distribution valve opening value,
Figure DEST_PATH_IMAGE208
Be the burnt gas valve opening value, Be the state value of initial time,
Figure DEST_PATH_IMAGE212
With
Figure DEST_PATH_IMAGE214
Be the constant matrices that obtains by identification,
Figure DEST_PATH_IMAGE216
The expression theorem in Euclid space;
Choose quadratic performance index
Wherein Be the prediction time domain, S, Q d , R d For
Figure DEST_PATH_IMAGE220
The real symmetrical weighting matrix of the positive definite of dimension is chosen diagonal matrix
Figure DEST_PATH_IMAGE222
,
Figure DEST_PATH_IMAGE224
,
Figure DEST_PATH_IMAGE226
, the element on the diagonal line
Figure DEST_PATH_IMAGE228
,
Figure DEST_PATH_IMAGE230
,
Figure DEST_PATH_IMAGE232
Can suitably choose according to system's actual motion characteristic;
According to matrix in the state equation AWith BValue, the Riccati matrix equation of finding the solution discrete-time system
Figure DEST_PATH_IMAGE234
Wherein IExpression
Figure 968938DEST_PATH_IMAGE220
The unit matrix of dimension can solve positive definite symmetric matrices W( k); Can design discrete time LQ optimal controller thus, the optimum humidity control law of corresponding feedback of status is
Figure DEST_PATH_IMAGE236
This method for optimally controlling will make that moisture control system has Gain margin and Phase margin.
CN201310281163.7A 2013-07-05 2013-07-05 The method for optimally controlling humidity of residual heat drying system Active CN103345160B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310281163.7A CN103345160B (en) 2013-07-05 2013-07-05 The method for optimally controlling humidity of residual heat drying system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310281163.7A CN103345160B (en) 2013-07-05 2013-07-05 The method for optimally controlling humidity of residual heat drying system

Publications (2)

Publication Number Publication Date
CN103345160A true CN103345160A (en) 2013-10-09
CN103345160B CN103345160B (en) 2016-03-02

Family

ID=49279964

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310281163.7A Active CN103345160B (en) 2013-07-05 2013-07-05 The method for optimally controlling humidity of residual heat drying system

Country Status (1)

Country Link
CN (1) CN103345160B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1567107A (en) * 2003-06-09 2005-01-19 石油大学(北京) Advanced control method and system for complex lagged process
JP2010266318A (en) * 2009-05-14 2010-11-25 Shizuoka Prefecture Prediction method of wet bulb temperature and wbgt, wbgt meter, and heat stroke risk determination device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1567107A (en) * 2003-06-09 2005-01-19 石油大学(北京) Advanced control method and system for complex lagged process
JP2010266318A (en) * 2009-05-14 2010-11-25 Shizuoka Prefecture Prediction method of wet bulb temperature and wbgt, wbgt meter, and heat stroke risk determination device

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
张振涛等: "木材热泵干燥窑内的湿度模糊控制模型初探", 《华北电力大学学报》, vol. 31, no. 6, 30 November 2004 (2004-11-30) *
李国昉等: "粮食干燥过程控制", 《中国粮油学报》, vol. 21, no. 2, 30 April 2006 (2006-04-30) *
翟文鹏: "基于负荷预测和设定点优化的制冷***模型预测控制方法研究", 《中国博士学位论文全文数据库信息科技辑》, 15 May 2013 (2013-05-15), pages 140 - 10 *
赵晓东等: "废塑料炼油过程控制反应釜温度控制***设计", 《杭州电子科技大学学报》, vol. 32, no. 6, 31 December 2012 (2012-12-31) *

Also Published As

Publication number Publication date
CN103345160B (en) 2016-03-02

Similar Documents

Publication Publication Date Title
CN108227488B (en) Sliding mode prediction control-based ultra-supercritical thermal power generating unit coordination control method
Fang et al. Backstepping-based nonlinear adaptive control for coal-fired utility boiler–turbine units
CN102734795B (en) Self-adapting coal quality change coordinated control method of single-bed circulating fluidized bed without external bed
CN103399486A (en) Temperature optical energy-saving control method for plastic dryer
CN104699064B (en) A kind of ature of coal self-tuning model method of fired power generating unit coordinated control system
CN106019939B (en) Supercritical unit reheat steam temperature Optimal Control System based on Nonlinear Intelligent Prediction and Control Technology
CN102401371A (en) Reheated gas temperature optimization control method based on multi-variable predictive control
CN105137760A (en) Denitration spraying ammonia automation control method and system
CN104834215A (en) Variation particle swarm optimized BP neural network proportion integration differentiation (PID) control algorithm
CN101509656A (en) Supercritical DC furnace synthesis type coordinating control method
CN111013370A (en) Wet desulphurization slurry supply amount prediction method based on deep neural network
CN107908106B (en) Double reheat power generation sets reheat steam temperature concentrates Prediction Control system from depression of order multiloop
CN107270283B (en) Multivariable constraint predictive control method based on circulating fluidized bed unit
CN111562744A (en) Boiler combustion implicit generalized predictive control method based on PSO algorithm
CN101788789A (en) Nonlinear predictive control method of unit plant based on chaos and hybrid optimization algorithm
CN110515304A (en) Overheating steam temperature PID forecast Control Algorithm based on ARX-Laguerre function model
CN114330869A (en) Multi-energy day-ahead scheduling optimization method for iron and steel enterprise considering multi-level indexes
CN105700357B (en) Method of Boiler Combustion Control System based on multivariable PID-PFC
CN114884063A (en) Distributed power supply and energy storage voltage control method and device considering interval coordination
CN109539359B (en) Phase-change electric heat storage heating system and method based on different working conditions PID + adaptive feedforward compensation
CN105786034B (en) System and method for controlling content of nitrogen oxides at denitration outlet of boiler
CN103605284A (en) Dynamic matrix control optimization-based waste plastic cracking furnace pressure controlling method
CN103345160A (en) Humidity optimization control method for waste heat drying system
Liu et al. Design for fuzzy decoupling control system of temperature and humidity
CN102902201A (en) Decoupling state spatial prediction control method of chemical multivariate processes

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20231114

Address after: 276800 Menlou Village, Zhongzhi Town, Wulian County, Rizhao City, Shandong Province

Patentee after: Shandong Xinghai Energy Conservation and Environmental Protection Technology Co.,Ltd.

Address before: 310018 No. 2 street, Xiasha Higher Education Zone, Hangzhou, Zhejiang

Patentee before: HANGZHOU DIANZI University

TR01 Transfer of patent right