CN1589955A - Intelligent control method of PTA particle size in fine terephthalic acid production device - Google Patents
Intelligent control method of PTA particle size in fine terephthalic acid production device Download PDFInfo
- Publication number
- CN1589955A CN1589955A CN 200410014996 CN200410014996A CN1589955A CN 1589955 A CN1589955 A CN 1589955A CN 200410014996 CN200410014996 CN 200410014996 CN 200410014996 A CN200410014996 A CN 200410014996A CN 1589955 A CN1589955 A CN 1589955A
- Authority
- CN
- China
- Prior art keywords
- particle diameter
- pta
- liquid level
- variable
- crystallizer
- 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
Links
Images
Landscapes
- Feedback Control In General (AREA)
Abstract
An intelligent method for controlling the granularity of PTA particles in refined terephthalic acid (PTA) production includes such steps as selecting the temp and discharge flow of hydrogenating reactor, and the liquid levels and temp of the first and the second crystallizers and using them as the input variables of the soft measuring system of granularity, creating the affection relation of primary operating parameters to PTA granularity, real-time continuous acquisition of procedure data as the soft measured values, determining the difference between settings and measured values, and automatically controlling the liquid levels and temp of the first and the second crystallizers.
Description
One, technical field
The invention belongs to chemical engineering and automation field, relate to pure terephthalic acid (hereinafter to be referred as PTA, i.e. Pure Terephthalic Acid) produce in the intelligence control method of PTA product cut size.
Two, background technology
PTA is the important source material of producing polyester (PET), adopt PTA direct esterification polycondensating process owing to produce PET more, therefore all multifactor (as crystal size, the foreign matter content etc.) of related PTA quality product to esterification, and influences bigger to the preparation of EG/PTA slurry and conveying etc.Secondly, introduce one of PTA product granularity index according to MIRES crystallization theory and relevant technologies data---median size increases, and helps PTA slurry stability and sludge forming performance.Along with the purposes of polyester more and more widely, the continuous increase of demand, what promoted greatly that PTA produces develops rapidly, simultaneously the quality of PTA product is had higher requirement.Therefore further investigation influences the principal element of PTA crystallisation process, and the PTA product cut size is stablized control, has realistic meaning to improving the PTA competitiveness of product in market.
The production technique of PTA is broadly divided into two classes: 1. two step method: at first with p-Xylol (PX) through atmospheric oxidation, make crude terephthalic acid (TA), and then TA be refined into PTA; 2. single stage method: PX only obtains terephthalic acid through peroxidation.Among the PTA that makes with two step method, its representational impurity---usually below 25ppm, and 4-CBA content is 200～300ppm among the PTA that single stage method makes to the content of carboxyl benzaldehyde (4-CBA).
In the two step method production technique, can be divided three classes again by the process for purification difference: the first kind is applied hydrofining methods such as Mitsui petro-chemical corporation, U.S. Amoco company, Britain ICI company; Second class is applied accurate oxidation style such as Mitsubishi changes into, Eastman Kodak company; The 3rd class is dimethyl terephthalate (DMT) (DMT) method for refining that Dynamitonobel company etc. implements.Wherein first kind method is because production cost is lower, quality product is more stable, is to produce be most widely used in the middle of the method for PTA a kind of at present.Accounting for the product of PTA product ultimate production more than 70% in the world all is to adopt this method to produce.The present invention is promptly at the hydrofining technology in the PTA two step method.
Usually, the hydrofining technology flow process is: on the technology the thick TA water behind the oxidation unit drying crystalline is dissolved into certain density slurry, is heated to and requires to deliver to hydrogenator after the solvent temperature.With Pd/C (palladium/carbon) is catalyzer, by catalytic hydrogenation reaction, makes that impurities is converted into water-soluble substances in the crude terephthalic acid.Hydrogenation reaction solution is sent to whizzer step by step after the decrease temperature and pressure and is separated in placed in-line crystallizer, the filter cake that obtains is again with the deionized water making beating, then after filtration and dry, makes fibre-grade pure terephthalic acid (PTA).
The major influence factors that influences the PTA particle diameter has: 1) slurry concentration; 2) mould temperature distributes; 3) residence time in the crystallizer; 4) agitator structure shape and stirring intensity; 5) pH value of solvent, impurity.Wherein, mould temperature distributes more main with the residence time, and they also are the significant parameters of regulating size in the actual production process.
In the PTA production process, with regard to present prior art, can't carry out online actual measurement to size, normally every day is by manually the mixed sample of multiple spot collection once being analyzed.Because manual analysis exists bigger analysis to lag behind, like this under the situation that the PTA particle diameter changes, often can't in time adjust the operation operating mode, perhaps, owing to there is more influence factor, only rely on the manual operation experience, can't quantitatively regulate simultaneously a plurality of operational variables, may cause " toning " or the regulating effect of product cut size slow, these all will directly have influence on quality product.
Three, summary of the invention
The objective of the invention is: the intelligence control method that a kind of PTA particle diameter is provided.Nerual network technique prediction median size is used in this invention; Utilize the soft observed value of this median size simultaneously, crystallizer pressure, liquid level, as regulated variable according to the adjusting priority of each variable, are realized the automatic adjustment to each regulated variable, to satisfy the control requirement of particle diameter.The present invention seeks under many influence factors condition, provide a kind of and in time the operation operating mode is adjusted, guarantee the more accurate adjusting of product cut size, thereby guarantee quality product by particle diameter control.
The object of the present invention is achieved like this: utilize charging, first mould temperature, first mould liquid level, second mould temperature of hydrogenation reaction actuator temperature, hydrogenator, the real-time image data of second mould liquid level, use nerual network technique prediction median size; Utilize the soft observed value of this median size simultaneously, get the first crystallizer pressure, first mould liquid level, the second crystallizer pressure, second mould liquid level as regulated variable, according to the adjusting priority of each variable, realize automatic adjustment, to satisfy the control requirement of particle diameter to each regulated variable.
At first, utilize nerual network technique, set up the soft measuring system of median size.The input variable of this system has 5, is respectively: the temperature difference between (1) hydrogenator and first crystallizer, Δ T
1The residence time of (2) first crystallizers, τ
1The temperature difference between (3) first crystallizers and second crystallizer, Δ T
2The residence time of (4) second crystallizers, τ
2(5) the PTA particle size analysis value in the preceding moment.
Simultaneously, consider the different time of different variablees,, need choose the current time input variable data of different unit time according to concrete full scale plant situation to grain diameter influence.As suppose that the soft observed value of particle diameter only considers the second crystallizer exit, when then choosing the data of hydrogenator correlated variables (as temperature and flow), need choose preceding two units of current time data constantly; When choosing the data of the first crystallizer correlated variables (as temperature), need choose the previous unit of current time data constantly; When choosing the data of the second crystallizer correlated variables, need choose the data of current time.
Utilize formula (1), obtain system's input variable:
Here, T
Reactpr(t-t
1) the preceding t of expression current time
1Temperature of reactor constantly, T
1st-crys(t-t
2) the preceding t of expression current time
2First mould temperature constantly, L
1st-crys(t-t
2) the preceding t of expression current time
2First mould liquid level constantly, F
Reactor(t-t
1) the preceding t of expression current time
1Reactor flow constantly, T
2nd-crys(t) second mould temperature of expression current time, L
2nd-crys(t) second mould liquid level of expression current time, G
T-1The particle size values of expression previous moment.
Utilize formula (2), the input variable of above-mentioned soft-sensing model is carried out normalized;
(2) in the formula, x
iBe the actual measured value of i input variable of soft-sensing model (being independent variable(s)), sx
iRepresent after i the input variable normalization method value as the neural network input,
Expression collects the variation range of i input variable, and the variation range of input independent variable(s) is after the normalization method
Here, the BP neural network of employing three-decker is set up the soft measuring system of particle diameter.In the soft measuring system of this neural network, the node number of input layer is i (i=5), and the hidden layer number of plies in middle layer is l (l=1～100), and each the number of hidden nodes is j (j=2～100), and the output layer node is k (k=1).Historical data by acquisition system input, output variable is trained network, after precision meets the demands, and record network structure and weights.
After obtaining the soft observed value of particle diameter, just can be according to the control set(ting)value requirement of particle diameter, selected operational variable is automatically adjusted, select first mould liquid level, the first crystallizer pressure, second mould liquid level, the second crystallizer pressure here as operational variable.
Owing to comprised the influence of first mould liquid level, first mould temperature, second mould liquid level, second mould temperature in the soft measuring system to particle diameter, and, direct linear dependence between each mould temperature and the pressure, promptly can change temperature by pressure parameter, and further influence particle diameter, therefore, can pass through these technology adjustable parameters, carry out the control of particle diameter.At first need to determine the adjusting priority of each regulated variable, be as supposition adjusting priority orders:
First mould liquid level＞first crystallizer the pressure＞second mould liquid level＞second crystallizer the pressure
Then in the regulate process of particle diameter, the adjusting of first mould liquid level will at first be carried out in allowing operating restraint, utilize soft measuring system to making linearization process near the current operating point in the regulate process, and the dichotomy that utilizes one dimension to optimize in the searching algorithm carries out choosing of the first mould liquid level Optimizing operation point, if first mould liquid level has arrived the regulation range border, then begin to carry out the adjusting of the first crystallizer pressure; By that analogy, all arrive the control border until the control requirement of satisfying particle diameter or regulated variable.
Calculate the regulated variable of flex point for soft measuring system occurring, promptly regulated variable does not satisfy the linear transformation principle with particle diameter, then gets and the most close operating point of the current operation operating mode of this variable.
Characteristics of the present invention: the intelligence control method that a kind of PTA particle diameter is provided.This method can be carried out particle diameter and carry out real-time soft measurement by other process variable (as flow, temperature, liquid level etc.) that easily detects.Simultaneously, adopt time variable control, can a plurality of major influence factors that influence particle diameter be automatically adjusted.Owing to adopt the soft measuring system of neural network to carry out the particle diameter real-time estimate, improved the precision of prediction and the fault freedom of particle diameter greatly, further guaranteed controlling performance.The present invention can generally be applicable to the production technique of utilizing the hydrofining method to produce PTA.
In addition, also can promptly introduce the soft measuring system that variablees such as slurry density, stir current, the 3rd mould temperature and liquid level, the 4th mould temperature and liquid level, the 5th mould temperature are set up the PTA particle diameter by the just amount that increases, method is consistent with the thinking of mentioning in this invention.And regulated variable equally also can expand to these newly-increased adjustable variables on original basis.
Four, description of drawings
PTA hydrofining technology FB(flow block) has been described among Fig. 1,
Wherein: corresponding numeral is 1 to enter the load flow of hydrogenator; 2. hydrogenation reaction actuator temperature indication; 3. first mould temperature indication; 4. first mould liquid level indication; 5. second mould temperature indication; 6. second mould liquid level indication.As shown in Figure 1, the artificial sample analysis of PTA product cut size records in discharging place of PTA product.
Fig. 2 is the soft measuring system block diagram of PTA granularity neural network, the x in 5 the input variables difference counterparty's formulas (1) shown in the figure
1～x
5The node of input layer is i (i=5), and the hidden node in middle layer is j (j=7), and the output layer node is k (k=1), and the input variable of soft measuring system is carried out normalized, and the normalization method scope of choosing here is [0.2 0.8].
Fig. 3 is a PTA particle diameter automatic control program block diagram.
Five, embodiment
The invention will be further described below in conjunction with accompanying drawing and by embodiment:
Load flow and temperature, first mould temperature and liquid level, second mould temperature and liquid level according to hydrogenator, and the particle size analysis value of previous moment, by computer system (as DCS system or real-time data base), obtain before the hydrogenator current time 1 hour load flow and temperature value, temperature and the level value of half hour before the first crystallizer current time, the temperature of the second crystallizer current time and level value.
By historgraphic data recording, choose 100 groups of data of above-mentioned variable, utilize equation (1) to obtain the input data of the soft measuring system of neural network, and carry out normalized, utilize the current manual analysis value of particle diameter to do target value.With 50 groups of neural network trainings,, trained and one group of weights that predicated error is less with 50 groups of prediction neural network generalization abilities; Here, the neural network input variable is 5, and choosing the hidden node number is 7, and output variable is 1.Can choose the different hidden layer numbers of plies and hidden node number according to requirement to soft measuring system precision; In general, the increase hidden layer number of plies in the proper range and the number of node thereof can the soft measuring system precision of corresponding raising.After training, obtain following one group of weights:
w
11＝-0.55531 w
12＝0.63814 w
13＝0.85929 w
14＝-0.13105 w
15＝0.4398
w
21＝0.34733 w
22＝0.57198 w
23＝0.13712 w
24＝-0.4362 w
25＝0.40703
w
31＝-0.52782 w
32＝0.26804 w
33＝-0.39269?w
34＝-0.018656?w
35＝0.83147
w
41＝0.11692 w
42＝2.2245 w
43＝-0.44579?w
44＝-0.1445 w
45＝0.68512
w
51＝-0.5777 w
52＝0.55799 w
53＝-0.46309?w
54＝-0.20686 w
55＝0.171
w
61＝0.32117 w
62＝1.0526 w
63＝-0.242 w
64＝0.066991 w
65＝0.87777
w
71＝-0.35918 w
72＝0.7199 w
73＝0.25523 w
74＝-0.2242 w
75＝-0.38332
ww
1＝-2.676 ww
2＝2.371 ww
3＝2.1013 ww
4＝-2.686 ww
5＝-2.2008
ww
6＝3.6055 ww
7＝3.6055
b
1＝-0.94256 b
2＝1.7425 b
3＝1.1609 b
4＝0.025712 b
5＝2.3402 b
6＝-0.19773
b
7＝0.32666
bpb
1＝-3.8373
(w wherein
IjBe the weights of i node to j node; Ww
jBe the weights of j node of middle hidden layer to k node of output layer; b
jThreshold value for j node of middle hidden layer; Bpb
kThreshold value for the output layer node)
It more than is the off-line simulation process, in the real time execution of device, then need go up the establishment that realizes the said process control language at the AM/APM of the DCS of PTA production equipment system (APPLICATION MODULE/ADVANCED PROCESS MANAGER), real-time, continuous acquisition by DCS systematic procedure data (referring to each input variable data in the soft measuring system here), the weights and the threshold value that train are brought into and calculated, just can obtain the real-time soft measurement predictor of PTA particle diameter; The soft measuring system output valve that obtain this moment is utilized (3) formula between [0.1,0.9], carry out anti-normalized, obtains the actual value of particle diameter.
G
t＝(sG
t-sG
min)/(sG
max-sG
min)*(G
max-G
min)+G
min (3)
In addition, for guarantee soft measuring system accurately, effectively the operation, utilize the manual analysis value carry out " rollings " correction.
As choose the corresponding value constantly of on-the-spot each variable, then obtain soft measuring system input variable x according to equation (1)
1～x
5Be respectively 38.6570,0.2399,28.4619,0.2352,100, after arriving between [11] after the normalization method, numerical value is respectively 0.0771, and-0.0202,0.3975 ,-0.0296,0, then calculate by the soft measuring system of neural network:
net1＝w
11*x
1+w
12*x
2+w
13*x
3+w
14*x
4+w
15*x
5+b
1 (4)
net2＝w
21*x
1+w
22*x
2+w
23*x
3+w
24*x
4+w
25*x
5+b
2 (5)
net3＝w
31*x
1+w
32*x
2+w
33*x
3+w
34*x
4+w
35*x
5+b
3 (6)
net4＝w
41*x
1+w
42*x
2+w
43*x
3+w
44*x
4+w
45*x
5+b
4 (7)
net5＝w
51*x
1+w
52*x
2+w
53*x
3+w
54*x
4+w
55*x
5+b
5 (8)
net6＝w
61*x
1+w
62*x
2+w
63*x
3+w
64*x
4+w
65*x
5+b
6 (9)
net7＝w
71*x
1+w
72*x
2+w
73*x
3+w
74*x
4+w
75*x
5+b
7 (10)
out1＝2/(1+exp(-2*net(1)))-1 (11)
out2＝2/(1+exp(-2*net(2)))-1 (12)
out3＝2/(1+exp(-2*net(3)))-1 (13)
out4＝2/(1+exp(-2*net(4)))-1 (14)
out5＝2/(1+exp(-2*net(5)))-1 (15)
out6＝2/(1+exp(-2*net(6)))-1 (16)
out7＝2/(1+exp(-2*net(7)))-1 (17)
net8＝ww
1*out1+ww
2*out2+ww
3*out3+ww
4*out4+ww
5*out5+ww
6*out6+ww
7*out7+bpb
1
(18)
out8＝net8 (19)
out9＝(out8-0.05)*150+70 (20)
When bringing weights, threshold value to (4)～(20) into, out9 is the particle diameter predictor that the soft measuring system of neural network calculates, and error is within ± 5% between grain diameter measurement value that obtains and the manual analysis value.
After obtaining the particle diameter predictor of soft measuring system, can control it.The adjustment priority of setting regulated variable is:
First mould liquid level＞first crystallizer the pressure＞second mould liquid level＞second crystallizer the pressure
If the set(ting)value of current first mould liquid level is 50%, allowing regulation range is [20 70]; The set(ting)value of the first crystallizer pressure is 3.6MPa, and allowing regulation range is [2.6 4.6]; The set(ting)value of second mould liquid level is 50%, and allowing regulation range is [20 70]; The set(ting)value of the second crystallizer pressure is 2.6MPa, and allowing regulation range is [1.6 3.6].
And, obtained the conversion relational expression of temperature and pressure:
T1=181+18*P1=245 ℃; (first mould temperature and pressure transformational relation)
T2=169+23*P2=228.8 ℃; (second mould temperature and pressure transformational relation)
Read current operation operating mode, calculating current particle diameter by soft measuring system is 120 μ m, if require particle diameter to be controlled within [100 110] μ m, illustrates that then current particle diameter does not meet the demands, and need regulate.
At first, if first mould liquid level satisfies near the condition of the available linearization of operating point, then can calculate first mould liquid level in the particle size values on 20%, 70% border, under the constant situation of other variable according to soft measuring system, its calculation result is respectively 90,140, and then explanation can be by only realizing the control requirement of particle diameter to the adjustment of first mould liquid level.Therefore, between liquid level regulation range [20 70], can utilize one dimension to optimize searching method,, seek an operating point and make the soft measuring system calculated value of particle diameter within [100 110] μ m as dichotomy, golden section method etc.Then, install in the bottom control loop realization particle diameter control requirement under the optimum value that liquid level is calculated.
If technology allows the regulation range of first crystallizer to narrow down, promptly only allow between [45 65], to regulate, then utilizing the particle size values on the operational boundaries that soft measuring system calculates is 112 and 125, only illustrates can't to realize the control requirement of particle diameter by the adjustment to first mould liquid level.Like this, just need operate under the situation of low limit value (45%), carry out adjusting the first crystallizer Pressure/Temperature at first mould liquid level.Control method is similar to the liquid level regulate process, and at first the particle diameter calculated value of calculating pressure on [2.6 4.6] operational boundaries if the particle size values that obtains then can be passed through optimization method within [100 110] μ m, sought suitable force value to satisfy the control requirement; If the particle size values that obtains does not still satisfy within [100 110] μ m, then carry out next variable, i.e. the adjustment of second mould liquid level.By that analogy, until satisfying particle diameter control requirement, perhaps all variable has all arrived operational boundaries.
In the process that above-mentioned variable is regulated, if the condition that certain variable does not satisfy near the available linearization of operating point occurs, then in two segment limits of flex point and operational boundaries, search for the adjusting parameter that meets the demands respectively, then the current operation operating mode of new operating point and this variable is compared, get the shortest operating point of both normal form distances as new technological operation point.
The condition of above-mentioned requirements all can satisfy, so this invention has universality in most PTA production equipment.
Claims (9)
1, a kind of intelligence control method of full scale plant PTA particle diameter, it is characterized in that utilizing charging, first mould temperature, first mould liquid level, second mould temperature of hydrogenation reaction actuator temperature, hydrogenator, the real-time image data of second mould liquid level, use nerual network technique prediction median size; Utilize the soft observed value of this median size simultaneously, get the first crystallizer pressure, first mould liquid level, the second crystallizer pressure, second mould liquid level as regulated variable;
At first, utilize nerual network technique, set up the soft measuring system of median size.The input variable of this system has 5, is respectively: the temperature difference between (1) hydrogenator and first crystallizer, Δ T
1The residence time of (2) first crystallizers, τ
1The temperature difference between (3) first crystallizers and second crystallizer, Δ T
2The residence time of (4) second crystallizers, τ
2(5) the PTA particle size analysis value in the preceding moment.
Utilize formula (1), obtain system's input variable:
Here, T
Reactor(t-t
1) the preceding t of expression current time
1Temperature of reactor constantly, T
Lst-crys(t-t
2) the preceding t of expression current time
2First mould temperature constantly, L
Lst-crys(t-t
2) the preceding t of expression current time
2First mould liquid level constantly, F
Reactor(t-t
1) the preceding t of expression current time
1Reactor flow constantly, T
2nd～crys(t) second mould temperature of expression current time, L
2nd-crys(t) second mould liquid level of expression current time, G
T-1The particle size values of expression previous moment;
Utilize formula (2), the input variable of above-mentioned soft-sensing model is carried out normalized; The input variable of neural network model is: Δ T
1(x
1, ℃), τ
1(x
2, mins), Δ T
2(x
3, ℃), τ
2(x
4, mins), G
T-1(x
5, μ m), and carry out normalized:
(2) in the formula, x
iBe the actual measured value of i input variable of soft-sensing model (being independent variable(s)), sx
iRepresent after i the input variable normalization method value as the neural network input,
Expression collects the variation range of i input variable, and the variation range of input independent variable(s) is after the normalization method
To collecting n1 group data, wherein every group of data comprise [x
1, x
2, x
3, x
4, x
5, G
t], after normalization method [sx
1, sx
2, sx
3, sx
4, sx
5, sG
t], form learning sample.To PTA particle diameter G
tNeural network model, with [sx
1, sx
2, sx
3, sx
4, sx
5] as the input of network, corresponding PTA particle diameter sG
tAs target value, training network; From the incidence relation of existing DCS system data and PTA particle diameter, set up model with nerual network technique; By real-time, continuous acquisition to above-mentioned model input variable, the weights and the threshold value that train are brought into and calculated, obtain the real-time estimate value of PTA particle diameter; Then, according to model a plurality of major influence factors that influence particle diameter are realized regulating automatically.
2, the intelligence control method of full scale plant PTA particle diameter as claimed in claim 1, it is characterized in that adopting the BP neural network of three-decker to set up the soft measuring system of particle diameter, in the soft measuring system of this neural network, the node number of input layer is i (i=5), the hidden layer number of plies in middle layer is l (l=1～100), each the number of hidden nodes is j (j=2～100), and the output layer node is k (k=1); Historical data by acquisition system input, output variable is trained network, after precision meets the demands, and record network structure and weights.
3, the intelligence control method of full scale plant PTA particle diameter as claimed in claim 1 is characterized in that the neural network flexible measurement method of PTA particle diameter is: the difference Δ T of the temperature in 1 unit moment before the temperature of choosing 2 unit time before the hydrogenator current time and the first crystallizer current time
1, 1 residence time τ constantly of unit before the first crystallizer current time
1, the difference Δ T of the temperature of the temperature of 2 unit time and the second crystallizer current time before the hydrogenator current time
2, the second crystallizer current time residence time τ
2And the particle size analysis value of previous moment, as the input variable of the soft measuring system of neural network, with the current particle diameter of real-time estimate.
4, by the intelligence control method of the described full scale plant PTA of claim 2 particle diameter, it is characterized in that soft measuring system predictor at set intervals with the manual analysis value to soft measuring system output carry out online " rollings " and optimize correction.
5, by the intelligence control method of the described full scale plant PTA of claim 1 particle diameter, it is characterized in that according to the soft measuring system of particle diameter, to first mould liquid level, pressure, the operational condition of second mould liquid level, pressure is regulated in real time, satisfying the control requirement of particle diameter, and according to the adjusting priority of each variable with allow regulation range to regulate successively step by step.
6,, it is characterized in that carrying out the linearization process of current operating point in regulation range according to soft measuring system by the intelligence control method of the described full scale plant PTA of claim 5 particle diameter; For the regulated variable that flex point occurs, get and the most close operating point of the current operation operating mode of this variable.
7,, it is characterized in that setting up the transforming relationship of crystallizer pressure and temperature by the intelligence control method of the described full scale plant PTA of claim 5 particle diameter.
8, by the intelligence control method of the described full scale plant PTA of claim 5 particle diameter, it is characterized in that determining the adjusting priority of each regulated variable,
First mould liquid level＞first crystallizer the pressure＞second mould liquid level＞second crystallizer the pressure.
9, by the intelligence control method of the described full scale plant PTA of claim 5 particle diameter, it is characterized in that in the regulate process of particle diameter, the adjusting of first mould liquid level will be allowed to carry out in the operating restraint, utilize soft measuring system to making linearization process near the current operating point in the regulate process, and the dichotomy that utilizes one dimension to optimize in the searching algorithm carries out choosing of the first mould liquid level Optimizing operation point, if first mould liquid level has arrived the regulation range border, then begin to carry out the adjusting of the first crystallizer pressure; By that analogy, all arrive the control border until the control requirement of satisfying particle diameter or regulated variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410014996 CN100475325C (en) | 2004-05-25 | 2004-05-25 | Intelligent control method of PTA particle diameter in fine terephthalic acid production device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410014996 CN100475325C (en) | 2004-05-25 | 2004-05-25 | Intelligent control method of PTA particle diameter in fine terephthalic acid production device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1589955A true CN1589955A (en) | 2005-03-09 |
| CN100475325C CN100475325C (en) | 2009-04-08 |
Family
ID=34600469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410014996 Expired - Fee Related CN100475325C (en) | 2004-05-25 | 2004-05-25 | Intelligent control method of PTA particle diameter in fine terephthalic acid production device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100475325C (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251747B (en) * | 2007-10-31 | 2012-04-25 | 华东理工大学 | Modelling method for industrial device model for dimethylbenzene oxidation reaction |
| CN102486632A (en) * | 2010-12-01 | 2012-06-06 | 中国石油化工股份有限公司 | On-line analyzing method of terephthalic acid crystal particle diameter in P-xylene oxidation process |
| CN102826388A (en) * | 2011-06-14 | 2012-12-19 | 逸盛大化石化有限公司 | Conveying method of CTA (cellulose triacetate fiber) powder in PTA (pure terephthalic acid) production device |
| CN101591240B (en) * | 2008-05-30 | 2013-07-10 | 华东理工大学 | Modeling method of crude terephthalic acid hydrofining reaction industrial device model |
| CN106777922A (en) * | 2016-11-30 | 2017-05-31 | 华东理工大学 | A kind of CTA hydrofinishings production process agent model modeling method |
| CN117740632A (en) * | 2024-02-21 | 2024-03-22 | 江苏嘉通能源有限公司 | PTA particle size dynamic soft measurement method based on differential evolution algorithm |
-
2004
- 2004-05-25 CN CN 200410014996 patent/CN100475325C/en not_active Expired - Fee Related
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251747B (en) * | 2007-10-31 | 2012-04-25 | 华东理工大学 | Modelling method for industrial device model for dimethylbenzene oxidation reaction |
| CN101591240B (en) * | 2008-05-30 | 2013-07-10 | 华东理工大学 | Modeling method of crude terephthalic acid hydrofining reaction industrial device model |
| CN102486632A (en) * | 2010-12-01 | 2012-06-06 | 中国石油化工股份有限公司 | On-line analyzing method of terephthalic acid crystal particle diameter in P-xylene oxidation process |
| CN102826388A (en) * | 2011-06-14 | 2012-12-19 | 逸盛大化石化有限公司 | Conveying method of CTA (cellulose triacetate fiber) powder in PTA (pure terephthalic acid) production device |
| CN106777922A (en) * | 2016-11-30 | 2017-05-31 | 华东理工大学 | A kind of CTA hydrofinishings production process agent model modeling method |
| CN117740632A (en) * | 2024-02-21 | 2024-03-22 | 江苏嘉通能源有限公司 | PTA particle size dynamic soft measurement method based on differential evolution algorithm |
| CN117740632B (en) * | 2024-02-21 | 2024-04-26 | 江苏嘉通能源有限公司 | PTA particle size dynamic soft measurement method based on differential evolution algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100475325C (en) | 2009-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE60010522T2 (en) | METHOD FOR THE CONTROLLED PRODUCTION OF POLYETHYLENE AND ITS COPOLYMERS | |
| EP1119800B1 (en) | A system for on line inference of physical and chemical properties and system for on line control | |
| CN1589955A (en) | Intelligent control method of PTA particle size in fine terephthalic acid production device | |
| WO1990010898A1 (en) | Supporting method and system for process control | |
| CN106802563B (en) | A kind of sewage procedure optimization control method based on drosophila optimization and LSSVM | |
| WO2021043712A1 (en) | System for planning, maintaining, managing and optimizing a production process | |
| CN106777922A (en) | A kind of CTA hydrofinishings production process agent model modeling method | |
| CN101963785B (en) | On-line control method for oxidation mother liquor filter process in purified terephthalic acid production | |
| JP2013544106A5 (en) | ||
| CN108934416A (en) | One kind being based on BP neural network combined harvester multi-parameter operation-control system and method | |
| US20220199195A1 (en) | Information processing apparatus, information processing method, and information processing program | |
| Thakur et al. | Process analytical technology (PAT) implementation for membrane operations in continuous manufacturing of mAbs: Model-based control of single-pass tangential flow ultrafiltration | |
| CN1230412C (en) | Establishing method of dynamic model of liquid phase catalytic paraxylene oxidizing reaction | |
| CN1036646C (en) | A method of controlling a hydrogormylation reaction | |
| DE102007026314A1 (en) | Fuel cell controlling and regulating method for electrical vehicles, involves determining one or multi dimensional connections between dynamic input variable and dynamic output variable in training algorithm | |
| CN109925992A (en) | It is a kind of based on multimode stage by stage continuously stir on-line monitoring method | |
| CN101591240A (en) | The modeling method of crude terephthalic acid hydrofining reaction industrial device model | |
| CN100345809C (en) | Process for the preparation of aldehydes | |
| CN102486632A (en) | On-line analyzing method of terephthalic acid crystal particle diameter in P-xylene oxidation process | |
| CN206240777U (en) | Feed Screen assemblies and feed screening machine | |
| CN101173952B (en) | Soft measuring method for confirming 4-carboxybenzaldehyde content | |
| CN1552753A (en) | Soft measure for polyester product quality target and process operating optimization method | |
| Banerjee et al. | Multi-objective optimization of continuous CHO cell clarification using acoustic wave separation | |
| CN108732940B (en) | Method and system for optimizing top circulating oil flow of catalytic cracking fractionator for gasoline and diesel cutting process | |
| CN117348614A (en) | Intelligent reservoir dispatching system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: SINOPEC YANGZI PETROCHEMICAL CO. Free format text: FORMER OWNER: YANGZI PETROCHEMICAL CO., LTD. Effective date: 20071214 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20071214 Address after: Xinhua Road, Dachang District, Jiangsu, Nanjing Province, China: 210048 Applicant after: Sinopec Yangzi Petrochemical Company Ltd. Address before: Xinhua Road, Dachang District, Jiangsu, Nanjing Province, China: 210048 Applicant before: Yangtze Petrochemical Industry Co., Ltd. |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090408 Termination date: 20170525 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |