CN102645315A - Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger - Google Patents

Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger Download PDF

Info

Publication number
CN102645315A
CN102645315A CN2012101295665A CN201210129566A CN102645315A CN 102645315 A CN102645315 A CN 102645315A CN 2012101295665 A CN2012101295665 A CN 2012101295665A CN 201210129566 A CN201210129566 A CN 201210129566A CN 102645315 A CN102645315 A CN 102645315A
Authority
CN
China
Prior art keywords
frequency
air quantity
heat exchanger
parameter
value
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
CN2012101295665A
Other languages
Chinese (zh)
Other versions
CN102645315B (en
Inventor
姜周曙
江爱朋
王剑
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.)
Hangzhou Dianzi University
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 CN201210129566.5A priority Critical patent/CN102645315B/en
Publication of CN102645315A publication Critical patent/CN102645315A/en
Application granted granted Critical
Publication of CN102645315B publication Critical patent/CN102645315B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses an automatic, fast and accurate detection method for air resistance characteristics of a large heat exchanger. The existing plate-fin heat exchanger air resistance characteristic measuring and controlling system cannot achieve automatic and fast measuring. The method includes: firstly performing data mining according to historical test data, building a general relationship between designed parameters and initial frequency of the heat exchanger through a support vector machine, obtaining control characteristics of heat exchanger channels through automatic identification of control characteristic parameters of different heat exchanger channels, setting pining and instruments diagram (PID) parameters of an automatic controller by adopting optimum performance indexes, regulating converted air flue volume standard values to a set standard value portion by adopting an incremental PID controller and obtaining the air resistance characteristics of the large heat exchanger by detecting the air resistance value at the moment. The automatic, fast and accurate detection method can achieve automatic and fast measuring of the air resistance characteristics of the large heat exchanger, is high in measuring accuracy and effectively improves production efficiency.

Description

A kind of large heat exchanger vapour lock characteristic quick and precisely detection method automatically
Technical field
The invention belongs to control technology fields, are related to data modeling and the process control of industrial process control field.It is mainly concerned with a kind of for during large-scale plate-fin heat exchanger vapour lock feature measurement, then the setting airflow value by heat exchanger Boiler pressure control after the conversion automatically and rapidly to obtain the vapour lock characteristic value of heat exchanger by detection.
Background technique
Vapour lock characteristic is the important indicator of characterization large heat exchanger mobile performance and one of the important indicator that large-scale plate-fin heat exchanger product must be tested before factory.Main includes two aspects: one be for certain heat exchanger channel, in normal temperature, normal pressure and normal flow, the pressure loss at the heat exchanger channel both ends；Another is the friction factor of heat exchanger channel in the case.Since test result is different under the conditions ofs different temperatures, pressure etc., lack comparativity.Therefore it is the status of criterion that required standard situation, which is normal atmosphere pressure, a standard zero degrees celsius, and the air quantity measured at this time is standard air quantity, and the air quantity measured under other pressure and temperatures needs to be converted into standard air quantity.Past, then conversion obtained the vapour lock characteristic in nominally heat exchanger generally using orifice flowmeter by the vapour lock parameter of heat exchanger when measurement certain air quantity in measurement process.Had the shortcomings that using above method obvious, one be orifice flowmeter measurement accuracy it is poor, in addition the control of air quantity is by the way of manual control valve, transformed air quantity cannot be accurately controlled in, third is using orifice plate measurement and manual control is not energy-efficient, inaccurate, subsequent artefacts are computationally intensive, since each heat exchanger testing time is long, working efficiency is relatively low.
Using the air quantity of Frequency Converter Control blower, energy saving of system problem can solve using nozzle measuring system, the precision of nozzle measurement air quantity is much higher than orifice flowmeter.It is in standard condition due to not can guarantee current working in test process, needs the temperature and pressure of heat exchanging device that can be converted according to the equation of gas state, conversion method are as follows:
Figure 2012101295665100002DEST_PATH_IMAGE002
, here
Figure 2012101295665100002DEST_PATH_IMAGE004
With
Figure 2012101295665100002DEST_PATH_IMAGE006
Indicate standard temperature and pressure (STP),
Figure 2012101295665100002DEST_PATH_IMAGE008
To survey air quantity,With
Figure 2012101295665100002DEST_PATH_IMAGE012
For observed pressure and temperature,
Figure 2012101295665100002DEST_PATH_IMAGE014
Then air quantity is surveyed for the standard after conversion.For the sake of accurate, air quantity is quickly controlled to the actual measurement air quantity of the standard after conversion during heat exchanger channel performance test, it namely requires this actual measurement standard air quantity then to be controlled and make it equal to established standards air quantity actual measurement air quantity by being converted into actual measurement standard air quantity.In view of the efficiency and power conservation requirement of test, measuring device carries out air quantity adjusting using converter technique.
Heat exchanger difference lane testing is that test has intermittence there are one feature: after testing a channel, before new channel is connected to measuring device, the requirement of measuring device air channel volume is zero.Due to the model and type difference of each heat exchanger, during the test, how the original frequency of blower is determined, and how according to the characteristic of system under test (SUT), air quantity fast and stable is controlled in setting value corresponding with standard condition using suitable control method, heat exchanging device measurement accuracy and rapidity have a major impact.Due to the hysteresis quality of measuring system, different channels part throttle characteristics is widely different and the nonlinear characteristic of frequency variation signal variation, and system is caused to be difficult to realize automatic rapid survey.It is therefore desirable to study a kind of method that can realize automatic quickly observing and controlling air quantity in the case, it is allowed to meet the industrial requirements of plate-fin heat exchanger vapour lock characteristic test.
Summary of the invention
The object of the present invention is to provide a kind of pair of various heat exchange device difference channel characteristic, the method for carrying out automatic rapid survey vapour lock characteristic.In the method, data mining is carried out according to historical test data first, general relationship between design of heat exchanger parameter and original frequency is established using support vector machines, then the control characteristic for obtaining heat exchanger channel is recognized automatically by the control characteristic parameter in various heat exchange device channel, using the pid parameter of optimal performance index adjusting automatic controller, then the standard value after being converted air channel volume using incremental timestamp device is controlled at established standards air quantity, obtains its vapour lock characteristic by detecting vapour lock value at this time.
Specific steps of the invention are as follows:
Step (1) acquires different type plate-fin heat exchanger design parameter and detection parameters, establishes the real-time data base comprising design of heat exchanger parameter and detection parameters；The design of heat exchanger parameter includes the tunnel name of heat exchanger, design standard air quantity, design vapour lock, friction factor, and detection parameters include practical vapour lock, environment temperature, air pressure and the blower frequency of heat exchanger.On this basis, based on historical test data, design standard air quantity, design vapour lock, the relational model between environment temperature and practical vapour lock, practical blower frequency are established using the strong support vector machine ensembles modeling method of generalization ability, it is predicted under various heat exchange device path setting standard air quantity and design vapour lock with this, blower is in order to reach the required frequency values of the established standards air quantity
Figure 2012101295665100002DEST_PATH_IMAGE016
.Specific modeling method is as follows:
Input parameter and output parameter for modeling sample can be expressed as
Figure 2012101295665100002DEST_PATH_IMAGE018
, wherein
Figure 2012101295665100002DEST_PATH_IMAGE020
Indicate the
Figure 2012101295665100002DEST_PATH_IMAGE022
Parameter vector of the group as input data, including design standard air quantity, design vapour lock and environment temperature,
Figure 2012101295665100002DEST_PATH_IMAGE024
Indicate theParameter vector of the group as output, including practical vapour lock and practical blower frequency,
Figure 2012101295665100002DEST_PATH_IMAGE026
For sample size.
For algorithm of support vector machine, kernel function is selected as radial basis function:
Figure 2012101295665100002DEST_PATH_IMAGE028
For radial basis function,For mapping function,Indicate the
Figure 2012101295665100002DEST_PATH_IMAGE036
Parameter vector of the group as input data,
Figure 2012101295665100002DEST_PATH_IMAGE038
,
Figure 2012101295665100002DEST_PATH_IMAGE040
For radial basis function nuclear parameter, if required objective function are as follows:
Figure 2012101295665100002DEST_PATH_IMAGE042
,
Figure 2012101295665100002DEST_PATH_IMAGE044
The predicted value of the practical vapour lock and blower frequency that are exported for model,
Figure 2012101295665100002DEST_PATH_IMAGE046
For weight coefficient vector,
Figure 2012101295665100002DEST_PATH_IMAGE048
For intercept, in order to calculate
Figure 205346DEST_PATH_IMAGE046
WithValue.Introduce relaxation factorWith
Figure 2012101295665100002DEST_PATH_IMAGE052
, and allow the error of fitting to be,
Figure 293617DEST_PATH_IMAGE046
With
Figure 341207DEST_PATH_IMAGE048
Value can be by constraining:
Figure 2012101295665100002DEST_PATH_IMAGE056
Figure 2012101295665100002DEST_PATH_IMAGE058
, under the conditions of, it minimizes:
Figure 2012101295665100002DEST_PATH_IMAGE060
It obtains, wherein
Figure 2012101295665100002DEST_PATH_IMAGE062
For structural risk minimization function, constant
Figure 2012101295665100002DEST_PATH_IMAGE064
For penalty coefficient,
Figure 2012101295665100002DEST_PATH_IMAGE066
With
Figure 2012101295665100002DEST_PATH_IMAGE068
For parametric variable.The minimization problem is a convex quadratic programming problem, introduces Lagrangian:
     
Wherein:
Figure 2012101295665100002DEST_PATH_IMAGE072
≥0, 
Figure 2012101295665100002DEST_PATH_IMAGE074
>=0, it is Lagrange's multiplier.
At saddle point, Lagrangian
Figure 2012101295665100002DEST_PATH_IMAGE076
Be about
Figure 2012101295665100002DEST_PATH_IMAGE078
Minimal point, andMaximal point, then the above minimization problem is converted into the maximization problems for seeking its dual problem.LagrangianAt saddle point be about
Figure 588267DEST_PATH_IMAGE078
Minimal point, then:
Figure 2012101295665100002DEST_PATH_IMAGE082
                    
The dual function of Lagrangian can be obtained
Figure 2012101295665100002DEST_PATH_IMAGE084
:
Figure 2012101295665100002DEST_PATH_IMAGE086
At this point,
According to Ku En-Plutarch (KKT) conditional theorem, there is following formula establishment in saddle point:
Figure 2012101295665100002DEST_PATH_IMAGE092
Figure 2012101295665100002DEST_PATH_IMAGE094
             
By above formula as it can be seen that , With
Figure 2012101295665100002DEST_PATH_IMAGE100
It all will not be simultaneously non-zero, can obtain:
Figure 2012101295665100002DEST_PATH_IMAGE102
  
It can be found out from above formula
Figure 436715DEST_PATH_IMAGE048
。
According to the above algorithm of support vector machine, the step of support vector machine ensembles modeling method, is as follows:
A. original training data initialization weight is
Figure 2012101295665100002DEST_PATH_IMAGE104
,
Figure 2012101295665100002DEST_PATH_IMAGE106
For weight update times, when initializing weight, set the number of iterations
Figure 2012101295665100002DEST_PATH_IMAGE110
。
B. it calls the above algorithm of support vector machine to model training sample, obtains a model
Figure 2012101295665100002DEST_PATH_IMAGE112
, calculate
Figure 228959DEST_PATH_IMAGE112
Average forecasting error square value:
Figure 2012101295665100002DEST_PATH_IMAGE114
。
C. original training data weight is updated:
Figure 2012101295665100002DEST_PATH_IMAGE116
。
D. it is distributed, is sampled in former training set, sampling condition according to the new weight of original training data are as follows:,Threshold values is sampled for the weight of setting, generates the training set of a Sub-SVM.
E. it repeats step b~d and obtains new model
Figure 2012101295665100002DEST_PATH_IMAGE122
With new sub- training set, until
Figure 770668DEST_PATH_IMAGE110
Secondary iteration is completed.
F. by acquisition
Figure 690083DEST_PATH_IMAGE110
A sub- supporting vector machine model is integrated, Model Weight are as follows:, the integrated model that finally obtains are as follows:
Figure 2012101295665100002DEST_PATH_IMAGE126
,For obtained support vector machine ensembles model.
Step (2) reaches the initial value of established standards air quantity leeward unit frequency using the model prediction heat exchanger channel that step (1) is established according to the design standard air quantity in real exchanger channel, design vapour lock and ambient temperature conditions
Figure 509003DEST_PATH_IMAGE016
, due to by test condition and input and output and be not present strong mechanism relevance, blower is in original frequency
Figure 222881DEST_PATH_IMAGE016
Although air quantity in the case of requires to be not much different with actual set standard air quantity, but is unable to satisfy requirement.It needs to correct again and by automatically controlling actual air volume control at established standards air quantity.
The frequency initial value that step (3) is obtained according to step (2)
Figure 330514DEST_PATH_IMAGE016
, blower frequency is adjusted to from 0
Figure 725724DEST_PATH_IMAGE016
80%, be denoted as
Figure 2012101295665100002DEST_PATH_IMAGE130
, when frequency reaches
Figure 405010DEST_PATH_IMAGE016
80% and stable operation 10-20s after, blower frequency will be become and adjusted to 100%
Figure 493051DEST_PATH_IMAGE016
, and stablize 10-20s, actual measurement standard air quantity at this time is denoted as
Figure 2012101295665100002DEST_PATH_IMAGE132
, blower frequency reaches
Figure 884718DEST_PATH_IMAGE016
80% and actual measurement standard air quantity when stable operation 10-20s be denoted as
Figure 2012101295665100002DEST_PATH_IMAGE134
.And in frequency from 80%
Figure 145936DEST_PATH_IMAGE016
To 100%
Figure 611552DEST_PATH_IMAGE016
And during stablizing 10-20s, with the actual measurement standard air quantity of period 0.5-1s record heat exchanger channel, actual measurement vapour lock and blower frequency.Then, by frequency from 80%
Figure 870495DEST_PATH_IMAGE016
To 100%And 10-20s records obtained actual measurement standard air quantity to stabilization in the process, blower frequency is individually subtracted
Figure 486470DEST_PATH_IMAGE134
With
Figure 9855DEST_PATH_IMAGE130
, air quantity changing value is obtained, is denoted as
Figure 2012101295665100002DEST_PATH_IMAGE136
And frequency change, it is denoted as
Figure 2012101295665100002DEST_PATH_IMAGE138
, wherein
Figure 2012101295665100002DEST_PATH_IMAGE140
With
Figure 2012101295665100002DEST_PATH_IMAGE142
Respectively indicate the of recordA air quantity variation and frequency change.
Step (4) flows Controlling model feature according to pipeline gas, establishes the transmission function between frequency and air quantity.Air quantity control system transmission function can be considered as one order inertia and add delay link, therefore can be set as the transmission function between frequency and standard air quantity
Figure 2012101295665100002DEST_PATH_IMAGE146
, wherein
Figure 2012101295665100002DEST_PATH_IMAGE148
Open loop gain cofficient, time constant and delay time are respectively indicated,For plural number.It is obtained according to step (3)
Figure 813601DEST_PATH_IMAGE138
, will
Figure 851964DEST_PATH_IMAGE148
Positive initial value is assigned respectively, passes through transmission function
Figure 392667DEST_PATH_IMAGE146
Calculate the control amount at sampled point
Figure 2012101295665100002DEST_PATH_IMAGE152
Output under effect
Figure 2012101295665100002DEST_PATH_IMAGE154
, then with
Figure 2012101295665100002DEST_PATH_IMAGE156
For target, three parameters in transmission function are fitted using least-squares algorithm
Figure 217229DEST_PATH_IMAGE148
, obtain the specific transmission function in the heat exchanger channel.
Step (5), in order to ensure the adjusting air quantity to preset standard air quantity of fast and stable, adjusts the parameter of PID after step (4) establishes the transmission function of heat exchanger channel control characteristic.Using IATE integral performance as optimal index, with parameter in PID controller
Figure 2012101295665100002DEST_PATH_IMAGE158
、、
Figure 2012101295665100002DEST_PATH_IMAGE162
For variable, with PID+
Figure 2012101295665100002DEST_PATH_IMAGE164
As open-loop transfer function, using closed loop transfer function, as constraint equation, with
Figure 739346DEST_PATH_IMAGE158
、
Figure 530585DEST_PATH_IMAGE160
、
Figure 671716DEST_PATH_IMAGE162
Be positive value be variable bound, optimized using nonlinear optimization solution technique, obtain optimal PID controller parameter
Figure 169693DEST_PATH_IMAGE158
、、
Figure 16613DEST_PATH_IMAGE162
Value, wherein
Figure 899118DEST_PATH_IMAGE158
、
Figure 313919DEST_PATH_IMAGE160
、
Figure 318784DEST_PATH_IMAGE162
Respectively indicate ratio, integral and differential parameter.
Step (6) obtains original frequency according to step (2)
Figure 2012101295665100002DEST_PATH_IMAGE166
And the frequency that step (3) obtains, approximate linear relationship parameter between air output and frequency is calculated, and determine the correction value of original frequency, the adjusting of blower frequency is arrived
Figure 301969DEST_PATH_IMAGE168
, and stablize
Figure 2012101295665100002DEST_PATH_IMAGE170
Second.According to the principle of similitude, frequency
Figure 2012101295665100002DEST_PATH_IMAGE172
There are approximation relations to be with air quantity。
Figure 2012101295665100002DEST_PATH_IMAGE176
With
Figure 2012101295665100002DEST_PATH_IMAGE178
The linear relationship parameter as needed.It will be in the frequency that step (3) obtain
Figure 23807DEST_PATH_IMAGE166
AndUnder air quantity, bring relation above formula into and find out parameter
Figure 2012101295665100002DEST_PATH_IMAGE180
With.It is obtaining
Figure 989718DEST_PATH_IMAGE180
With
Figure 214025DEST_PATH_IMAGE182
Afterwards, according to
Figure 2012101295665100002DEST_PATH_IMAGE184
Relationship, enable
Figure DEST_PATH_IMAGE186
For established standards air quantity, can find out in established standards air quantity leeward unit frequency forecast value revision value
Figure 665735DEST_PATH_IMAGE168
。
Blower frequency is set in by step (7)Frequency point, etc. frequencies reach setting value and stabilization
Figure 487247DEST_PATH_IMAGE170
After second.Then use incremental timestamp device that the actual measurement Boiler pressure control of heat exchanger channel at established standards air quantity, is obtained the vapour lock characteristic of the heat exchanger channel under established standards air quantity.The output form of PID are as follows:
Figure DEST_PATH_IMAGE188
；
Figure DEST_PATH_IMAGE190
Here three parameters of incremental timestamp device
Figure 374300DEST_PATH_IMAGE158
、、
Figure 73452DEST_PATH_IMAGE162
To obtain obtaining that three parameters in step (5).
Figure DEST_PATH_IMAGE192
Indicate the sampling period,
Figure DEST_PATH_IMAGE194
The error between the setting value and value of feedback of corresponding step number is represented,
Figure DEST_PATH_IMAGE196
、
Figure DEST_PATH_IMAGE198
、
Figure DEST_PATH_IMAGE200
Respectively indicate the frequency values of current frequency value, frequency change and next step.By incremental timestamp, actual measurement airflow value can be controlled automatically at established standards airflow value, control precision within 0.5%.Then it is at this time the vapour lock characteristic under established standards air quantity by the vapour lock characteristic that measurement obtains, by actual measurement vapour lock and designs the comparison between vapour lock, heat exchanger vapour lock characteristic performance index can be obtained.
Beneficial effects of the present invention: this method not only replaces manual inspection heat exchanger vapour lock Characterization method in the past completely, have the characteristics that detect automatically, automatic calculating and measuring accuracy it is high.The automation of this method and rapidity are very good, can adapt to different loads and air quantity requirement, while improving measuring accuracy, reducing labor workload, can greatly speed up heat exchanger test lot number.
Specific embodiment
Quick and precisely detection method, specific implementation use following steps automatically for a kind of large heat exchanger vapour lock characteristic:
Step (1) acquires different type plate-fin heat exchanger design parameter and detection parameters, establishes the real-time data base comprising design of heat exchanger parameter and detection parameters；The design of heat exchanger parameter includes the tunnel name of heat exchanger, design standard air quantity, design vapour lock, friction factor, and detection parameters include practical vapour lock, environment temperature, air pressure and the blower frequency of heat exchanger.On this basis, based on historical test data, design standard air quantity, design vapour lock, the relational model between environment temperature and practical vapour lock, practical blower frequency are established using the strong support vector machine ensembles modeling method of generalization ability, it is predicted under various heat exchange device path setting standard air quantity and design vapour lock with this, blower is in order to reach the required frequency values of the established standards air quantity.Specific modeling method is as follows:
Input parameter and output parameter for modeling sample can be expressed as, wherein
Figure 949376DEST_PATH_IMAGE020
Indicate the
Figure 575529DEST_PATH_IMAGE022
Parameter vector of the group as input data, including design standard air quantity, design vapour lock and environment temperature,
Figure 42543DEST_PATH_IMAGE024
Indicate the
Figure 209082DEST_PATH_IMAGE022
Parameter vector of the group as output, including practical vapour lock and practical blower frequency,
Figure 365257DEST_PATH_IMAGE026
For sample size.
For algorithm of support vector machine, kernel function is selected as radial basis function:
Figure 224629DEST_PATH_IMAGE028
Figure 895782DEST_PATH_IMAGE030
For radial basis function,For mapping function,
Figure 142272DEST_PATH_IMAGE034
Indicate the
Figure 172545DEST_PATH_IMAGE036
Parameter vector of the group as input data,
Figure 268677DEST_PATH_IMAGE038
,
Figure 777019DEST_PATH_IMAGE040
For radial basis function nuclear parameter, if required objective function are as follows:
Figure 970103DEST_PATH_IMAGE042
,The predicted value of the practical vapour lock and blower frequency that are exported for model,For weight coefficient vector,
Figure 66738DEST_PATH_IMAGE048
For intercept, in order to calculate
Figure 114328DEST_PATH_IMAGE046
With
Figure 424087DEST_PATH_IMAGE048
Value.Introduce relaxation factor
Figure 557128DEST_PATH_IMAGE050
With
Figure 610534DEST_PATH_IMAGE052
, and allow the error of fitting to be
Figure 512631DEST_PATH_IMAGE054
,
Figure 55608DEST_PATH_IMAGE046
With
Figure 348049DEST_PATH_IMAGE048
Value can be by constraining:
Figure 24067DEST_PATH_IMAGE058
, under the conditions of, it minimizes:
Figure 675628DEST_PATH_IMAGE060
It obtains, wherein
Figure 789121DEST_PATH_IMAGE062
For structural risk minimization function, constant
Figure 246647DEST_PATH_IMAGE064
For penalty coefficient,
Figure 795440DEST_PATH_IMAGE066
With
Figure 945799DEST_PATH_IMAGE068
For parametric variable.The minimization problem is a convex quadratic programming problem, introduces Lagrangian:
Figure 275149DEST_PATH_IMAGE070
     
Wherein:≥0, 
Figure 939665DEST_PATH_IMAGE074
>=0, it is Lagrange's multiplier.
At saddle point, Lagrangian
Figure 260925DEST_PATH_IMAGE076
Be about
Figure 749676DEST_PATH_IMAGE078
Minimal point, and
Figure 814584DEST_PATH_IMAGE080
Maximal point, then the above minimization problem is converted into the maximization problems for seeking its dual problem.Lagrangian
Figure 400286DEST_PATH_IMAGE076
At saddle point be about
Figure 564551DEST_PATH_IMAGE078
Minimal point, then:
                    
The dual function of Lagrangian can be obtained
Figure 471513DEST_PATH_IMAGE084
:
Figure 849405DEST_PATH_IMAGE086
At this point,
Figure 38126DEST_PATH_IMAGE090
According to Ku En-Plutarch (KKT) conditional theorem, there is following formula establishment in saddle point:
Figure 382520DEST_PATH_IMAGE092
Figure 942814DEST_PATH_IMAGE094
             
By above formula as it can be seen that
Figure 511199DEST_PATH_IMAGE096
,
Figure 727417DEST_PATH_IMAGE098
With
Figure 672239DEST_PATH_IMAGE100
It all will not be simultaneously non-zero, can obtain:
Figure 87040DEST_PATH_IMAGE102
  
Figure 764009DEST_PATH_IMAGE058
It can be found out from above formula
Figure 258401DEST_PATH_IMAGE048
。
According to the above algorithm of support vector machine, the step of support vector machine ensembles modeling method, is as follows:
A. original training data initialization weight is
Figure 6914DEST_PATH_IMAGE104
,
Figure 213904DEST_PATH_IMAGE106
For weight update times, when initializing weight
Figure 389671DEST_PATH_IMAGE108
, set the number of iterations
Figure 377218DEST_PATH_IMAGE110
。
B. it calls the above algorithm of support vector machine to model training sample, obtains a model
Figure 601526DEST_PATH_IMAGE112
, calculate
Figure 990919DEST_PATH_IMAGE112
Average forecasting error square value:
Figure 337587DEST_PATH_IMAGE114
。
C. original training data weight is updated:
Figure 484535DEST_PATH_IMAGE116
。
D. it is distributed, is sampled in former training set, sampling condition according to the new weight of original training data are as follows:
Figure 574850DEST_PATH_IMAGE118
,
Figure 756433DEST_PATH_IMAGE120
Threshold values is sampled for the weight of setting, generates the training set of a Sub-SVM.
E. it repeats step b~d and obtains new modelWith new sub- training set, untilSecondary iteration is completed.
F. by acquisition
Figure 802252DEST_PATH_IMAGE110
A sub- supporting vector machine model is integrated, Model Weight are as follows:
Figure 900658DEST_PATH_IMAGE124
, the integrated model that finally obtains are as follows:
Figure 589129DEST_PATH_IMAGE126
,
Figure 710668DEST_PATH_IMAGE128
For obtained support vector machine ensembles model.
Step (2) reaches the initial value of established standards air quantity leeward unit frequency using the model prediction heat exchanger channel that step (1) is established according to the design standard air quantity in real exchanger channel, design vapour lock and ambient temperature conditions, due to by test condition and input and output and be not present strong mechanism relevance, blower is in original frequencyAlthough requiring to be not much different with actual set standard air quantity, but it is unable to satisfy requirement.It needs to correct again and by automatically controlling actual air volume control at established standards air quantity.
The frequency initial value that step (3) is obtained according to step (2)
Figure 158333DEST_PATH_IMAGE016
, blower frequency is adjusted to from 0
Figure 829486DEST_PATH_IMAGE016
80%, be denoted as, when frequency reaches
Figure 810397DEST_PATH_IMAGE016
80% and stable operation 10-20s after, frequency converter frequency is adjusted to 100%
Figure 840670DEST_PATH_IMAGE016
, and stablize 10-20s, standard actual measurement air quantity at this time is denoted as, frequency converter frequency reaches
Figure 716583DEST_PATH_IMAGE016
80% and standard actual measurement air quantity when stable operation 10-20s be denoted as
Figure 909666DEST_PATH_IMAGE134
.And in frequency from 80%
Figure 110841DEST_PATH_IMAGE016
To 100%And during stablizing 10-20s, with standard actual measurement air quantity, actual measurement vapour lock and the blower frequency after period 0.5-1s record heat exchanger channel conversion.Then, by frequency from 80%
Figure 6301DEST_PATH_IMAGE016
To 100%And stable 10-20s records obtained standard actual measurement air quantity in the process, blower frequency is individually subtracted
Figure 363650DEST_PATH_IMAGE134
With
Figure 231112DEST_PATH_IMAGE130
, air quantity changing value is obtained, is denoted as
Figure 346836DEST_PATH_IMAGE136
And frequency change, it is denoted as
Figure 186616DEST_PATH_IMAGE138
, whereinWith
Figure 349930DEST_PATH_IMAGE142
Respectively indicate the of record
Figure 941448DEST_PATH_IMAGE144
A air quantity variation and frequency change.
Step (4) flows Controlling model feature according to pipeline gas, establishes the transmission function between frequency and air quantity.Air quantity control system transmission function can be considered as one order inertia and add delay link, therefore can be set as the transmission function between frequency and standard air quantity, wherein
Figure 677509DEST_PATH_IMAGE148
Open loop gain cofficient, time constant and delay time are respectively indicated,
Figure 457246DEST_PATH_IMAGE150
For plural number.It is obtained according to step (3)
Figure 914772DEST_PATH_IMAGE138
, will
Figure 791461DEST_PATH_IMAGE148
Positive initial value is assigned respectively, passes through transmission function
Figure 613924DEST_PATH_IMAGE146
Calculate the control amount at sampled point
Figure 943274DEST_PATH_IMAGE152
Output under effect, then with
Figure 935687DEST_PATH_IMAGE156
For target, three parameters in transmission function are fitted using least-squares algorithm
Figure 929051DEST_PATH_IMAGE148
, obtain the specific transmission function in the heat exchanger channel.
Step (5), in order to ensure the adjusting air quantity to preset standard air quantity of fast and stable, adjusts the parameter of PID after step (4) establishes the transmission function of heat exchanger channel control characteristic.Using IATE integral performance as optimal index, with parameter in PID controller
Figure 763275DEST_PATH_IMAGE158
、
Figure 828183DEST_PATH_IMAGE160
、
Figure 351568DEST_PATH_IMAGE162
For variable, with PID+
Figure 578150DEST_PATH_IMAGE164
As open-loop transfer function, using closed loop transfer function, as constraint equation, with
Figure 882093DEST_PATH_IMAGE158
、
Figure 422795DEST_PATH_IMAGE160
、
Figure 863004DEST_PATH_IMAGE162
Be positive value be variable bound, optimized using nonlinear optimization solution technique, obtain optimal PID controller parameter
Figure 526066DEST_PATH_IMAGE158
、、
Figure 396119DEST_PATH_IMAGE162
Value, wherein
Figure 628518DEST_PATH_IMAGE158
、
Figure 462481DEST_PATH_IMAGE160
、
Figure 475437DEST_PATH_IMAGE162
Respectively indicate ratio, integral and differential parameter.
Step (6) obtains original frequency according to step (2)
Figure 357942DEST_PATH_IMAGE166
And the frequency that step (3) obtains, approximate linear relationship parameter between air output and frequency is calculated, and determine the correction value of original frequency
Figure 777608DEST_PATH_IMAGE168
, the adjusting of blower frequency is arrived
Figure 215543DEST_PATH_IMAGE168
, and stablize
Figure 964056DEST_PATH_IMAGE170
Second.According to the principle of similitude, frequency
Figure 233363DEST_PATH_IMAGE172
There are approximation relations to be with air quantity
Figure 346813DEST_PATH_IMAGE174
。With
Figure 620985DEST_PATH_IMAGE178
The linear relationship parameter as needed.It will be in the frequency that step (3) obtain
Figure 948061DEST_PATH_IMAGE166
And
Figure 29150DEST_PATH_IMAGE130
Under air quantity, bring relation above formula into and find out parameter
Figure 509853DEST_PATH_IMAGE180
With.It is obtaining
Figure 781751DEST_PATH_IMAGE180
With
Figure 299320DEST_PATH_IMAGE182
Afterwards, according to
Figure 933564DEST_PATH_IMAGE184
Relationship, enable
Figure 561991DEST_PATH_IMAGE186
For established standards air quantity, can find out in established standards air quantity leeward unit frequency forecast value revision value
Figure 925976DEST_PATH_IMAGE168
。
Blower frequency is set in by step (7)
Figure 552130DEST_PATH_IMAGE168
Frequency point, etc. frequencies reach setting value and stabilization
Figure 735987DEST_PATH_IMAGE170
After second.Then use incremental timestamp device that the actual measurement Boiler pressure control of heat exchanger channel at established standards air quantity, is obtained the vapour lock characteristic of the heat exchanger channel under established standards air quantity.The output form of PID are as follows:
；
Figure 324280DEST_PATH_IMAGE190
Here three parameters of incremental timestamp device
Figure 183651DEST_PATH_IMAGE158
、
Figure 854804DEST_PATH_IMAGE160
、
Figure 762717DEST_PATH_IMAGE162
To obtain obtaining that three parameters in step (5).
Figure 835715DEST_PATH_IMAGE192
Indicate the sampling period,
Figure 865988DEST_PATH_IMAGE194
The error between the setting value and value of feedback of corresponding step number is represented,、、
Figure 663546DEST_PATH_IMAGE200
Respectively indicate the frequency values of current frequency value, frequency change and next step.By incremental timestamp, actual measurement airflow value can be controlled automatically at established standards airflow value, control precision within 0.5%.Then it is at this time the vapour lock characteristic under established standards air quantity by the vapour lock characteristic that measurement obtains, by actual measurement vapour lock and designs the comparison between vapour lock, heat exchanger vapour lock characteristic performance index can be obtained.

Claims (1)

1. a kind of large heat exchanger vapour lock characteristic quick and precisely detection method automatically, it is characterised in that the step of this method includes:
Step (1) acquires different type design of heat exchanger parameter and detection parameters, establishes the real-time data base comprising design of heat exchanger parameter and detection parameters；
The design of heat exchanger parameter includes the tunnel name of heat exchanger, design standard air quantity, design vapour lock and friction factor；Detection parameters include practical vapour lock, environment temperature, air pressure and the blower frequency of heat exchanger；
On this basis, based on historical test data, design standard air quantity, design vapour lock, the relationship between environment temperature and practical vapour lock, practical blower frequency are established using the strong support vector machine ensembles modeling method of generalization ability, it is predicted under various heat exchange device path setting standard air quantity and design vapour lock with this, blower is in order to reach the required frequency values of the established standards air quantity
Figure 2012101295665100001DEST_PATH_IMAGE002
；Specific modeling method is as follows:
Input parameter and output parameter for modeling sample are expressed as
Figure 2012101295665100001DEST_PATH_IMAGE004
, wherein
Figure 2012101295665100001DEST_PATH_IMAGE006
Indicate the
Figure 2012101295665100001DEST_PATH_IMAGE008
Parameter vector of the group as input data, including design standard air quantity, design vapour lock and environment temperature,
Figure 2012101295665100001DEST_PATH_IMAGE010
Indicate the
Figure 876446DEST_PATH_IMAGE008
Parameter vector of the group as output, including practical vapour lock and practical blower frequency,
Figure 2012101295665100001DEST_PATH_IMAGE012
For sample size；
For algorithm of support vector machine, kernel function is selected as radial basis function:
Figure 2012101295665100001DEST_PATH_IMAGE014
Figure 2012101295665100001DEST_PATH_IMAGE016
For radial basis function,
Figure 2012101295665100001DEST_PATH_IMAGE018
For mapping function,
Figure 2012101295665100001DEST_PATH_IMAGE020
Indicate the
Figure 2012101295665100001DEST_PATH_IMAGE022
Parameter vector of the group as input data,
Figure 2012101295665100001DEST_PATH_IMAGE024
,
Figure 2012101295665100001DEST_PATH_IMAGE026
For radial basis function nuclear parameter, if required objective function are as follows:
Figure 2012101295665100001DEST_PATH_IMAGE028
,
Figure 2012101295665100001DEST_PATH_IMAGE030
The predicted value of the practical vapour lock and blower frequency that are exported for model,
Figure 2012101295665100001DEST_PATH_IMAGE032
For weight coefficient vector,
Figure 2012101295665100001DEST_PATH_IMAGE034
For intercept, in order to calculate
Figure 354438DEST_PATH_IMAGE032
With
Figure 134176DEST_PATH_IMAGE034
Value；Introduce relaxation factor
Figure DEST_PATH_IMAGE036
With
Figure 2012101295665100001DEST_PATH_IMAGE038
, and allow the error of fitting to be
Figure 2012101295665100001DEST_PATH_IMAGE040
,
Figure 716336DEST_PATH_IMAGE032
With
Figure 593025DEST_PATH_IMAGE034
Value is by constraining:
Figure 2012101295665100001DEST_PATH_IMAGE044
, under the conditions of, it minimizes:
Figure 2012101295665100001DEST_PATH_IMAGE046
It obtains, wherein
Figure 2012101295665100001DEST_PATH_IMAGE048
For structural risk minimization function, constant
Figure 2012101295665100001DEST_PATH_IMAGE050
For penalty coefficient,WithFor parametric variable；The minimization problem is a convex quadratic programming problem, introduces Lagrangian:
Figure 2012101295665100001DEST_PATH_IMAGE056
     
Wherein:
Figure 2012101295665100001DEST_PATH_IMAGE058
≥0, 
Figure 2012101295665100001DEST_PATH_IMAGE060
>=0, it is Lagrange's multiplier；
At saddle point, Lagrangian
Figure 2012101295665100001DEST_PATH_IMAGE062
Be about
Figure 2012101295665100001DEST_PATH_IMAGE064
Minimal point, and
Figure DEST_PATH_IMAGE066
Maximal point, then the above minimization problem is converted into the maximization problems for seeking its dual problem；LagrangianAt saddle point be about
Figure 373866DEST_PATH_IMAGE064
Minimal point, then:
Figure DEST_PATH_IMAGE068
                    
The dual function of Lagrangian can be obtained
Figure DEST_PATH_IMAGE070
:
Figure DEST_PATH_IMAGE072
At this point,
Figure DEST_PATH_IMAGE074
Figure DEST_PATH_IMAGE076
According to Kuhn-Tucker condition theorem, there is following formula establishment in saddle point:
Figure DEST_PATH_IMAGE080
 
By above formula as it can be seen that ,
Figure DEST_PATH_IMAGE084
With
Figure DEST_PATH_IMAGE086
It all will not be simultaneously non-zero, can obtain:
Figure DEST_PATH_IMAGE088
  
Figure 88879DEST_PATH_IMAGE044
It can be found out from above formula
Figure 820075DEST_PATH_IMAGE034
；
According to the above algorithm of support vector machine, the step of support vector machine ensembles modeling method, is as follows:
A. original training data initialization weight is
Figure DEST_PATH_IMAGE090
,
Figure DEST_PATH_IMAGE092
For weight update times, when initializing weight, set the number of iterations；
B. it calls the above algorithm of support vector machine to model training sample, obtains a model
Figure DEST_PATH_IMAGE098
, calculate
Figure 125023DEST_PATH_IMAGE098
Average forecasting error square value:
Figure DEST_PATH_IMAGE100
；
C. original training data weight is updated:；
D. it is distributed, is sampled in former training set, sampling condition according to the new weight of original training data are as follows:
Figure DEST_PATH_IMAGE104
,
Figure DEST_PATH_IMAGE106
Threshold values is sampled for the weight of setting, generates the training set of a Sub-SVM；
E. it repeats step b~d and obtains new model
Figure DEST_PATH_IMAGE108
With new sub- training set, until
Figure 440204DEST_PATH_IMAGE096
Secondary iteration is completed；
F. by acquisition
Figure 177216DEST_PATH_IMAGE096
A sub- supporting vector machine model is integrated, Model Weight are as follows:
Figure DEST_PATH_IMAGE110
, the integrated model that finally obtains are as follows:,
Figure DEST_PATH_IMAGE114
For obtained support vector machine ensembles model；
Step (2) reaches the initial value of established standards air quantity leeward unit frequency using the model prediction heat exchanger channel that step (1) is established according to the design standard air quantity in real exchanger channel, design vapour lock and ambient temperature conditions
Figure 893412DEST_PATH_IMAGE002
, due to by test condition and input and output and be not present strong mechanism relevance, blower is in original frequency
Figure 385573DEST_PATH_IMAGE002
Although requiring to be not much different with actual set standard air quantity, but it is unable to satisfy requirement；It needs to correct again and by automatically controlling actual air volume control at established standards air quantity；
The frequency initial value that step (3) is obtained according to step (2)
Figure 689515DEST_PATH_IMAGE002
, blower frequency is adjusted to from 0
Figure 964639DEST_PATH_IMAGE002
80%, be denoted as
Figure DEST_PATH_IMAGE116
, when frequency reaches
Figure 732743DEST_PATH_IMAGE002
80% and stable operation 10-20s after, frequency converter frequency is adjusted to 100%, and stablize 10-20s, standard actual measurement air quantity at this time is denoted as
Figure DEST_PATH_IMAGE118
, frequency converter frequency reaches
Figure 921465DEST_PATH_IMAGE002
80% and standard actual measurement air quantity when stable operation 10-20s be denoted as
Figure DEST_PATH_IMAGE120
；And in frequency from 80%
Figure 62597DEST_PATH_IMAGE002
To 100%And during stablizing 10-20s, with standard actual measurement air quantity, actual measurement vapour lock and the blower frequency after period 0.5-1s record heat exchanger channel conversion；Then, by frequency from 80%
Figure 456855DEST_PATH_IMAGE002
To 100%
Figure 469810DEST_PATH_IMAGE002
And 10-20s records obtained actual measurement air quantity to stabilization in the process, blower frequency is individually subtracted
Figure 352315DEST_PATH_IMAGE120
With；Air quantity changing value is obtained, is denoted as
Figure DEST_PATH_IMAGE122
And frequency change, it is denoted as
Figure DEST_PATH_IMAGE124
, wherein
Figure DEST_PATH_IMAGE126
WithRespectively indicate the of record
Figure DEST_PATH_IMAGE130
A air quantity variation and frequency change；
Step (4) flows Controlling model feature according to pipeline gas, establishes the transmission function between frequency and air quantity；Air quantity control system transmission function is considered as one order inertia and adds delay link, therefore can be set as the transmission function between frequency and standard air quantity
Figure DEST_PATH_IMAGE132
, wherein
Figure DEST_PATH_IMAGE134
Open loop gain cofficient, time constant and delay time are respectively indicated,
Figure DEST_PATH_IMAGE136
For plural number；It is obtained according to step (3)
Figure 140023DEST_PATH_IMAGE124
, willPositive initial value is assigned respectively, passes through transmission function
Figure 60892DEST_PATH_IMAGE132
Calculate the control amount at sampled point
Figure DEST_PATH_IMAGE138
Output under effect
Figure DEST_PATH_IMAGE140
, then with
Figure DEST_PATH_IMAGE142
For target, three parameters in transmission function are fitted using least-squares algorithm
Figure 720412DEST_PATH_IMAGE134
, obtain the specific transmission function in the heat exchanger channel；
Step (5), in order to ensure the adjusting air quantity to preset standard air quantity of fast and stable, adjusts the parameter of PID after step (4) establishes the transmission function of heat exchanger channel control characteristic；Using IATE integral performance as optimal index, with parameter in PID controller
Figure DEST_PATH_IMAGE144
、
Figure DEST_PATH_IMAGE146
、
Figure DEST_PATH_IMAGE148
For variable, with PID+
Figure DEST_PATH_IMAGE150
As open-loop transfer function, using closed loop transfer function, as constraint equation, with
Figure 83129DEST_PATH_IMAGE144
、
Figure 70677DEST_PATH_IMAGE146
、
Figure 357302DEST_PATH_IMAGE148
Be positive value be variable bound, optimized using nonlinear optimization solution technique, obtain optimal PID controller parameter
Figure 418799DEST_PATH_IMAGE144
、
Figure 765466DEST_PATH_IMAGE146
、
Figure 246169DEST_PATH_IMAGE148
Value, wherein
Figure 336485DEST_PATH_IMAGE144
、
Figure 518068DEST_PATH_IMAGE146
、
Figure 770058DEST_PATH_IMAGE148
Respectively indicate ratio, integral and differential parameter；
Step (6) obtains original frequency according to step (2)
Figure DEST_PATH_IMAGE152
And the frequency that step (3) obtains
Figure 732197DEST_PATH_IMAGE116
, approximate linear relationship parameter between air output and frequency is calculated, and determine the correction value of original frequency
Figure DEST_PATH_IMAGE154
, the adjusting of blower frequency is arrived
Figure 422942DEST_PATH_IMAGE154
, and stablize
Figure DEST_PATH_IMAGE156
Second；According to the principle of similitude, frequency
Figure DEST_PATH_IMAGE158
With air quantity
Figure 911561DEST_PATH_IMAGE118
There are approximation relations to be
Figure DEST_PATH_IMAGE160
；
Figure DEST_PATH_IMAGE162
With
Figure DEST_PATH_IMAGE164
The linear relationship parameter as needed；It will be in the frequency that step (3) obtain
Figure 459086DEST_PATH_IMAGE152
And
Figure 846205DEST_PATH_IMAGE116
Under air quantity, bring relation above formula into and find out parameter
Figure DEST_PATH_IMAGE166
With；It is obtaining
Figure 278323DEST_PATH_IMAGE166
With
Figure 168919DEST_PATH_IMAGE168
Afterwards, according to
Figure DEST_PATH_IMAGE170
Relationship, enable
Figure 28290DEST_PATH_IMAGE118
For established standards air quantity, can find out in established standards air quantity leeward unit frequency forecast value revision value
Figure 699443DEST_PATH_IMAGE154
；
Blower frequency is set in by step (7)Frequency point, etc. frequencies reach setting value and stabilizationAfter second；Then use incremental timestamp device that the actual measurement Boiler pressure control of heat exchanger channel at established standards air quantity, is obtained the vapour lock characteristic of the heat exchanger channel under established standards air quantity；The output form of PID are as follows:
Figure DEST_PATH_IMAGE172
；
Here three parameters of incremental timestamp device
Figure 913890DEST_PATH_IMAGE144
、
Figure 89917DEST_PATH_IMAGE146
、To obtain obtaining that three parameters in step (5)；Indicate the sampling period,
Figure DEST_PATH_IMAGE178
The error between the setting value and value of feedback of corresponding step number is represented,、
Figure DEST_PATH_IMAGE182
、
Figure DEST_PATH_IMAGE184
Respectively indicate the frequency values of current frequency value, frequency change and next step；By incremental timestamp, actual measurement airflow value can be controlled automatically at established standards airflow value, control precision within 0.5%；Then it is at this time the vapour lock characteristic under established standards air quantity by the vapour lock characteristic that measurement obtains, by actual measurement vapour lock and designs the comparison between vapour lock, heat exchanger vapour lock characteristic performance index can be obtained.
CN201210129566.5A 2012-04-28 2012-04-28 Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger Active CN102645315B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210129566.5A CN102645315B (en) 2012-04-28 2012-04-28 Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210129566.5A CN102645315B (en) 2012-04-28 2012-04-28 Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger

Publications (2)

Publication Number Publication Date
CN102645315A true CN102645315A (en) 2012-08-22
CN102645315B CN102645315B (en) 2014-08-13

Family

ID=46658277

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210129566.5A Active CN102645315B (en) 2012-04-28 2012-04-28 Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger

Country Status (1)

Country Link
CN (1) CN102645315B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103245481A (en) * 2013-05-07 2013-08-14 杭州电子科技大学 Frequency conversion technology based detection method for air resistance characteristic of large-scale variable-load heat exchanger
CN103822758A (en) * 2014-03-06 2014-05-28 中国石油大学(北京) Online diagnosis and selective control method and device for leakage current unusual service conditions of heat exchanger
CN111067515A (en) * 2019-12-11 2020-04-28 中国人民解放军军事科学院军事医学研究院 Intelligent airbag helmet system based on closed-loop control technology

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5105372A (en) * 1987-10-31 1992-04-14 Rolls-Royce Plc Data processing system using a kalman filter
US5301101A (en) * 1990-06-21 1994-04-05 Honeywell Inc. Receding horizon based adaptive control having means for minimizing operating costs
CN101344761A (en) * 2008-08-15 2009-01-14 哈尔滨工业大学 Control object modeling method for energy-saving reconstruction of thermal power station
CN102122132A (en) * 2010-01-11 2011-07-13 北京航空航天大学 Intelligent control system for environmental simulation system based on a fuzzy neural network
CN102411660A (en) * 2011-11-30 2012-04-11 哈尔滨工业大学 Heat supply process modeling method based on typical test signals and least squares

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5105372A (en) * 1987-10-31 1992-04-14 Rolls-Royce Plc Data processing system using a kalman filter
US5301101A (en) * 1990-06-21 1994-04-05 Honeywell Inc. Receding horizon based adaptive control having means for minimizing operating costs
CN101344761A (en) * 2008-08-15 2009-01-14 哈尔滨工业大学 Control object modeling method for energy-saving reconstruction of thermal power station
CN102122132A (en) * 2010-01-11 2011-07-13 北京航空航天大学 Intelligent control system for environmental simulation system based on a fuzzy neural network
CN102411660A (en) * 2011-11-30 2012-04-11 哈尔滨工业大学 Heat supply process modeling method based on typical test signals and least squares

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
徐丁等: "板翅式换热器翅片表面流动特性测试系统", 《杭州电子科技大学学报》, vol. 30, no. 4, 31 August 2010 (2010-08-31) *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103245481A (en) * 2013-05-07 2013-08-14 杭州电子科技大学 Frequency conversion technology based detection method for air resistance characteristic of large-scale variable-load heat exchanger
CN103245481B (en) * 2013-05-07 2015-07-15 杭州电子科技大学 Frequency conversion technology based detection method for air resistance characteristic of large-scale variable-load heat exchanger
CN103822758A (en) * 2014-03-06 2014-05-28 中国石油大学(北京) Online diagnosis and selective control method and device for leakage current unusual service conditions of heat exchanger
CN103822758B (en) * 2014-03-06 2016-05-04 中国石油大学(北京) The unusual service condition inline diagnosis of heat exchanger leakage current and Selective Control method and device
CN111067515A (en) * 2019-12-11 2020-04-28 中国人民解放军军事科学院军事医学研究院 Intelligent airbag helmet system based on closed-loop control technology
CN111067515B (en) * 2019-12-11 2022-03-29 中国人民解放军军事科学院军事医学研究院 Intelligent airbag helmet system based on closed-loop control technology

Also Published As

Publication number Publication date
CN102645315B (en) 2014-08-13

Similar Documents

Publication Publication Date Title
CN109583585B (en) Construction method of power station boiler wall temperature prediction neural network model
CN102486390B (en) Method for steam pipe network of iron and steel enterprise to correct metering data
CN101025424B (en) Climate room for high precision dynamic detection for large-sized VOC detection and its detecting apparatus
CN107632524B (en) Communication machine room temperature model prediction control method and system
CN107299170B (en) A kind of blast-melted quality robust flexible measurement method
CN109165418B (en) Room temperature measuring method based on household calorimeter data
CN113821984B (en) Heating furnace billet temperature calculation method based on time domain convolution model
CN202939322U (en) Device for calibrating humidity dynamic response characteristics of sonde with double-flow method
CN102645315A (en) Automatic, fast and accurate detection method for air resistance characteristics of large heat exchanger
CN109519957B (en) A kind of ultra-supercritical boiler closed loop optimized control method of combustion
CN102998720A (en) Method and device for calibrating dynamic response characteristic of sonde humidity by double flow method
CN104714572A (en) High-precision constant temperature control method for microwave radiometer
CN105528000A (en) Intelligent temperature control meter for aircraft
Wu et al. Integrated soft sensing of coke-oven temperature
CN108106679B (en) Method and system for measuring inlet air volume of coal mill of power station
CN104950670A (en) Integrated multi-model method for controlling CSTRs (continuous stirred tank reactors)
CN102879421A (en) Measuring device and measuring method of heat-conductivity efficiency
CN103322807B (en) Novel burning-through point forecasting method
CN102305676B (en) Automatic calibrating apparatus on flow of heat energy meter
CN103245481B (en) Frequency conversion technology based detection method for air resistance characteristic of large-scale variable-load heat exchanger
CN103092815B (en) To the method transferring function by calibration in monitoring device
CN104199296B (en) Linear regression performance evaluation method with forgetting factor
CN112985692A (en) Atmospheric pressure sensor error calibration method integrating polynomial and learning model
CN103413029B (en) There is the moving horizon estimation method of multi-speed sample CSTR
Li et al. The research of temperature compensation for thermopile sensor based on improved PSO-BP algorithm

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