JPH08127602A - Continuous polymerization of polymer and apparatus therefor - Google Patents

Abstract

PURPOSE: To shorten the exchanging time of a brand and improve the quality of the brand. CONSTITUTION: This apparatus for continuously producing a polymer comprises a controlling part 3 for changing a value of a controllable factor based on a pattern for each brand in exchanging the brand in continuous polymerization of the polymer, a sensing part 16 for always sensing the value of a dependent factor changing dependently upon the value of the controllable factor and a fuzzy controlling part 11 capable of carrying out the fuzzy control so as to approximate the values of the controllable factor and the dependent factor to the target values corresponding to the exchanged brand after the passage of a prescribed time from the changed value of the controllable factor.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for continuously polymerizing a polymer which automatically switches brands in the continuous polymerization of the polymer.

[0002]

Polymers are produced by polymerization. Polymerization refers to the binding of two or more molecules of one compound to another compound having a multiple of the molecular weight. Further, the polymer is continuously polymerized in the polymerization vessel, and a plurality of types of brands whose molecular weight and density are changed are produced.

In such a continuous polymerization method of a polymer, when the brand A is switched to the brand B, the composition of the polymerization vessel is swift and stable in order to minimize transition products (products lost due to brand switching). Need to switch.

For example, hydrogen / ethylene (change of molecular weight),
The ratio of the gas supply amount such as propylene / ethylene (molecular weight / density change) is the composition of the polymerization vessel and is an index of transition (brand switching).

As a conventional polymer continuous polymerization method, when the amount of ethylene is kept constant, the amount of catalyst, the amount of hydrogen, the amount of propylene, etc. are adjusted to bring the gas chromatography ratio close to the target value of the transition brand. In order to bring the composition to the target value quickly and stably,
The amount of hydrogen and the amount of propylene are overacted for a certain time,
After that, there is a method of returning the composition to the target value.

[0006]

However, in the conventional method, the brand of the polymer is switched manually by the operator, and in this case, it largely depends on the skill and feeling of the operator. Therefore, the time required for the transition varies depending on the individual difference. Also,
Due to individual differences, it was difficult to stabilize the quality of transition brand values.

The present invention provides a continuous polymerizing method and apparatus for a polymer, which can shorten the switching time and improve the quality of the brand value by automatically switching the brands in the continuous polymerization of the polymer. is there.

[0008]

The present invention adopts the following means in order to solve the above problems. FIG. 1 is a principle view showing a polymer continuous polymerization apparatus of the present invention.

<Summary of Polymer Continuous Polymerization Apparatus of the Present Invention>
The polymer continuous polymerization apparatus of the present invention is a polymer continuous polymerization apparatus for continuously polymerizing a polymer by continuously changing the composition of a polymerization vessel to produce different kinds of brands, and the brands in the continuous polymerization of the polymer. At the time of switching, the control unit 3 that changes the value of the control factor based on the model pattern for each brand, the detection unit 16 that constantly detects the value of the dependent factor that changes depending on the value of the control factor, and the control factor of the control factor. A fuzzy control unit 11 for performing fuzzy control so that the value of the control factor and the value of the dependent factor approach a target value corresponding to the switched brand after a lapse of a predetermined time from the time when the value is changed. . The control unit 3
Inputs the value of the dependent factor detected by the detection unit 16 and outputs the values of the dependent factor and the control factor to the fuzzy control unit 11 after the lapse of the predetermined time (corresponding to claim 15).

Each constituent element will be described below. (Control Unit) The control unit 3 changes the value of the control factor on the basis of the model pattern for each brand at the time of brand switching in continuous polymer polymerization, and is, for example, a distributed computer system. (Detection Unit) The detection unit 16 constantly detects the value of the dependent factor that changes depending on the value of the control factor, and is, for example, a data collection unit. (Fuzzy control unit) The fuzzy control unit 11 sets the value of the control factor and the value of the dependent factor to a target value corresponding to the switched brand after a predetermined time has elapsed since the value of the control factor was changed. The fuzzy control is performed so as to approach each other, and is, for example, a fuzzy control management unit.

<Additional Structure in the Polymer Continuous Polymerization Apparatus of the Present Invention> The polymer continuous polymerization apparatus of the present invention comprises the above-mentioned essential constituent elements. Even when the constituent elements are specifically as follows: To establish.

The additional component is that the fuzzy control unit obtains the value of the control factor by performing a fuzzy operation on the numerical value of the dependent factor detected by the detection unit, and the control unit obtains the value. Is to change the value of the control factor (corresponding to claim 16).

Another additional component is that the operation is performed by a computer (corresponding to claim 17). Another additional component is that the predetermined time is a time for starting the switching of brands and setting the target value to the value of the switched brand after over-action of the control factor (claim). 18).

Another additional component is that only the pressure of the polymerization vessel is controlled as the dependent factor by changing the control factor during the predetermined time period (corresponding to claim 19). ).

Another additional component is that the model pattern is a pattern calculated from past manufacturing results (corresponding to claim 20). The polymer may be polyolefin (corresponding to claim 21), and the polyolefin may be polyethylene (corresponding to claim 22) or polyolefin wax (corresponding to claim 23).

The control factor is a hydrogen feed amount (claim 2
4), a catalyst feed amount (corresponding to claim 25), and a propylene feed amount (corresponding to claim 26). The dependent factor is the ratio of hydrogen to ethylene (Claim 27)
(Corresponding to claim 28) and the pressure in the polymerization vessel corresponding to claim 28).

FIG. 2 is a principle flow chart showing the continuous polymerizing method of the present invention. <Summary of Polymer Continuous Polymerization Method of the Present Invention> The polymer continuous polymerization method of the present invention is a continuous polymerization method of a polymer in which continuous polymerization of a polymer is carried out by continuously changing the composition of a polymerization vessel to produce different kinds of brands in order. The step S1 of changing the value of the control factor based on the model pattern of each brand at the time of brand switching in the continuous polymerization of the polymer, and the value of the dependent factor that changes depending on the value of the control factor are always detected. Detection step S2, and fuzzy so that the value of the control factor and the value of the dependent factor are close to the target value corresponding to the switched brand after a predetermined time has elapsed since the value of the control factor was changed. And fuzzy control step S3 for performing control (corresponding to claim 1).

<Additional Structure in the Polymer Continuous Polymerization Method of the Present Invention> The polymer continuous polymerization method of the present invention comprises the above-mentioned essential steps. However, even if the steps are specifically as follows: .

The additional process is that the fuzzy control step S3 obtains the value of the control factor by performing a fuzzy operation on the numerical value of the dependent factor detected in the detection step S2, and the control step S1 It is to change the value of the obtained control factor (corresponding to claim 2).

Another additional step is characterized in that the calculation is executed by a computer (corresponding to claim 3). Another additional step is that the predetermined time is a time for starting the switching of the brand and setting the target value to the value of the brand switched after the control factor is over-actioned (claim 4). Corresponding to).

Another additional step is that only the pressure of the polymerization vessel is controlled as the dependent factor by changing the control factor during the predetermined time period (corresponding to claim 5). .

Another additional step is that the model pattern is a pattern calculated from past manufacturing results (corresponding to claim 6). The polymer may be polyolefin (corresponding to claim 7), and the polyolefin may be polyethylene (corresponding to claim 8) or polyolefin wax (corresponding to claim 9).

The control factor is the hydrogen feed amount (claim 1
0), a catalyst feed amount (corresponding to claim 11), and a propylene feed amount (corresponding to claim 12). The dependent factor is the ratio of hydrogen to ethylene (claim 13).
(Corresponding to claim 14) and the pressure in the polymerization vessel corresponding to claim 14).

[0024]

According to the polymer continuous polymerization method of the present invention, in the control step S1, the value of the control factor is changed based on the model pattern at the time of brand switching in the continuous polymerization of the polymer, and in the detection step S2, the value of the control factor is changed. Always detect the value of the dependent dependent factor.

Further, in the fuzzy control step S3, the value of the control factor and the value of the dependent factor are set to a target value corresponding to the switched brand after a predetermined time has elapsed since the value of the control factor was changed. Fuzzy control is performed so as to approach.

Since the transition polymerization is automatically performed by using the model pattern and by fuzzy control, the switching time can be shortened and the quality of the brand value can be improved.

Further, the value of the control factor is obtained by calculating the numerical value of the detected dependent factor by the fuzzy calculation section, and the value of the obtained control factor is changed. That is, it is possible to approach the target value quickly and stably by the feedback control.

Furthermore, during the predetermined time period, since the brand switching is started and the control factor is over-actioned, the target value is set to the switched brand value, so that the fuzzy control can be performed quickly and stably. The control factor and the dependent factor can be brought close to the target value.

Further, by changing the control factor during the predetermined time period, only the pressure of the polymerization vessel is controlled as the dependent factor and the physical property is not controlled, so that the transition time can be shortened.

Further, since the model pattern is a pattern calculated from past manufacturing results, and this pattern is an optimized pattern to some extent, the transition time can be shortened.

[0031]

EXAMPLES Examples of the continuous polymerizing method and apparatus of the present invention will be described below with reference to the drawings.

Examples of the polymer continuous polymerization method and apparatus of the present invention will be described. FIG. 3 is a hardware block diagram showing a polymer continuous polymerization apparatus according to an embodiment of the present invention. The continuous polymerizer for polymers continuously polymerizes the polymer by continuously changing the composition of the polymerizer to sequentially produce different types of brands.

The polymer continuous polymerization apparatus comprises a model algorithm control (MAC) unit 1 and a distributed control system (DCS) unit 3 connected to the MAC unit 1. The MAC unit 1 performs model pattern setting (to the DCS unit) and fuzzy control, and manages model pattern parameters and fuzzy parameters, brand condition files, and fuzzy control. DCS unit 3
Is a program control by model pattern,
It manages the entire device by a sequence and controls the sequence.

The MAC section 1 will be described later with data in which the action portion is patterned by the test sequence created by the DCS section 3. What is the model pattern?
It refers to a pattern obtained by setting the manipulated variable of propylene.

The method of patterning the transition operation based on the experience of a skilled operator and constructing a model for each brand is similar to model predictive control. In order to achieve optimization, it is preferable to use a controller that uses model predictive control.

As a typical method of model predictive control,
For example, MAC (Model Algorithmic Control) and
There is DMC (Dynamic Matrix Control). In this typical method, a dynamic characteristic model of a process is brought into a control algorithm as shown in FIG. 4, and the behavior Y _P (t), Y _P (t + 1), ... It is controlled so as to reach a desired value within a finite time while predicting from the operation amount of.

Fuzzy control in the MAC unit 1 is as follows.
Since the use of the model pattern alone cannot control the physical properties that change every moment, it is used to control the physical properties to the target values. Fuzzy control is a control that outputs an operation output specified based on ambiguous information. Ambiguous information is information that cannot be completely specified. Fuzzy control realizes human-like control operation by changing human know-how into membership functions and rules.

Further, in the polymer continuous polymerization apparatus, the feedback value of the process is input to the fuzzy calculation unit 14 which will be described later, and the calculation result is reflected again in the process. That is, the polymer continuous polymerization device is a model pattern of the operation record for each brand, control to control in a sequence,
This is a device having an indirect advance control for performing fuzzy calculation at a higher level and setting and controlling the result in the DCS unit 3.

The DCS section 3 is composed of CENTUM-V, and has a COPSV (CENTUM OPERATORS STATION) 4 and a COP.
SC (CENTUM OPERATORS CONSOLE) 5, CFCD (CENTUM
FIELD CONTROL STATION) 2 part 6 is provided.

Although not shown, the MAC unit 1 includes a hard disk drive (HDD) and a random access memory (R).
AM) disc, RS71 card, software. Although not shown, the software includes a line controller BASIC, an I / O card, an interactive graphic UTY, a trend library, a database UTY, and M.
S-DOS conversion UTY, DOS SHEEL, FUZZ
It consists of Y "AdMAS".

Further, a catalyst, hydrogen, ethylene, a monomer, etc. are inputted to the polymerization vessel 8 in order to polymerize the polymer. A catalyst feed amount, a hydrogen feed amount, a propylene feed amount, and the like are input to the CFCD 2 section 6. Also, the CFCD2 part 6
Controls the pressure of the polymerization vessel 8 and the gas chromatography ratio. The CFCD unit 2 includes a batch setting device 35 described later.

The polymerization vessel 8 carries out continuous polymerization of a polymer, and the polymer is, for example, polyolefin.
Polyolefin is a polymer obtained by addition polymerization of hydrocarbon (olefin) having one carbon-carbon double bond in the molecule. The polyolefin is, for example, low-density polyethylene synthesized by radical polymerization. The polyolefin may also be, for example, a polyolefin wax.

Next, a software configuration of the polymer continuous polymerization apparatus will be described. FIG. 5 is a diagram showing a software configuration of the MAC unit 1. FIG. 6 is a diagram showing a software configuration of the DCS unit 3.

The software of the device is the MA shown in FIG.
It is composed of the C software unit 10 and the DCS sequence unit 30 shown in FIG. The MAC software unit 10 includes a transition automation system unit 10a and an application unit 10b.

The transition automation system unit 10
a is a fuzzy control management unit 11 connected to the fuzzy controller 18 and the database 9, a trend (recording) monitor unit 12 connected to the fuzzy control management unit 11, a database input / output unit 13 connected to the database 9,
Fuzzy operation unit 1 connected to fuzzy control management unit 11
4 is provided.

Further, the transition automation system unit 10a is connected to the process simulation unit 15, the data collection unit 16 connected to the CFCD2 unit 6 and the fuzzy control management unit 11, the CFCD2 unit 6 and the fuzzy control management unit 11. And a data setting unit 17 for

The fuzzy control management section 11 manages data input / output to / from the fuzzy controller 18 and brand condition parameters, and performs fuzzy control such as data processing and calculation and pressure control of model patterns. That is, the fuzzy control management unit 11 determines the value of the control factor (hydrogen feed amount,
Fuzzy control so that dependent factors (gas ratio, pressure) that change depending on the value of the control factor after a lapse of a predetermined time from the change of the catalyst feed amount) approaches the target value corresponding to the switched brand. I do.

The trend monitor section 12 displays the transition status in real time as a trend, and displays the content and status of the fuzzy control. The database input / output unit 1
Reference numeral 3 manages input / output of a data file 9a in a database 9 of a brand, the constructed model pattern parameter, and brand operating condition data, and inputs or changes data.

The process simulation section 15 is
Although not shown, it is provided in a simple simulator 21, which will be described later, and has a simple model formula for process simulation and a chemical model formula for estimating polymer physical properties and verifying fuzzy outputs.

The data collection unit 16 collects the data from the DCS unit 3 by the communication software activated on the line control and stores it in the common area. That is, the data collection unit 16 functions as a detection unit and detects the values of the dependent factors (gas chromatography ratio, pressure).

Although not shown, the fuzzy operation unit 14 is provided in the fuzzy controller 18, is a part of the fuzzy package "AdMAS", is supplied by a subprogram, and is accessed by the fuzzy control management unit 11 to perform fuzzy operation. To do. That is, the fuzzy operation unit 14 obtains the value of the control factor by performing a fuzzy operation on the numerical values of the dependent factors collected by the data collection unit 16.

The data setting unit 17 sets the model pattern data of the brand and the output value from the fuzzy calculation unit 14 to CF.
It is set in the batch setter 35 in the CD2 unit 6. The application unit 10b includes a fuzzy package (not shown), a simple simulator 21, a transition menu unit (not shown), a transition data collection unit 22, a transition setting unit 23, a file selection unit 24, a data analysis trend unit 25, and a DOS shell package (not shown). Prepare

The fuzzy package constructs fuzzy rules, membership functions and the like. The fuzzy rule has an antecedent part proposition and a consequent part proposition, each of which has a fuzzy set. FIG. 7 shows an example of the fuzzy rule. In FIG. 7, the rule is
Two antecedent variables x is A, y is B and one antecedent variable z
has is C. Where A1, A2, B1, B2, C
1 and C2 are fuzzy sets.

The membership function is defined to quantify the fuzzy rule proposition. In Fig. 9, the membership function applied fuzzy reasoning to the driving logic of the vehicle considering the distance and speed.

The fuzzy operation unit 14 obtains the degree of conformity of the antecedent part of each rule with respect to a given input, obtains the inference result of each rule based on the obtained degree of conformity, and further The final inference result is obtained from the rule inference result.
FIG. 9 shows an inference process of the rule shown in FIG. 7 by the fuzzy operation unit 14.

The simple simulator 21 inputs the data from the transition data file 9a to the fuzzy controller 18, displays the output value in the trend monitor section 12 as a trend, and sets the fuzzy control membership function and rule based on the result. adjust.

The transition menu section is a transition-related menu software. The transition data collection unit 22 stores 88 data from 1 hour before the transition start to the transition end in the data file 9a in the database 9. The transition setting unit 23 sets a trend issue and a data collection start time.

The file selection section 24 selects the transition data file 9a to be displayed. The data analysis trend part 25 is
The trend of the data file 9a of the designated brand is displayed.

The DOS shell package is a DOS support system that supports the development of application software for transition automation, and edits the disk of the MAC unit 1 and converts files.

Next, the DCS sequence unit 30 is connected to the MAC unit 1 by the system management sequence unit 3
1, a parameter setting sequence unit 3 connected to the system management sequence unit 31 and the batch setter 35
2. Fuzzy control sequence unit 3 connected to the system management sequence unit 31 and the batch setting unit 35
3. The model pattern sequence section 34 is connected to the system management sequence section 31.

The fuzzy control sequence section 33 and the model pattern sequence section 34 are adapted to control the superimposing unit 8. The system management sequence section 3
1 manages the selection of the brand, the selection of the used sequence, the start and stop of the sequence, the selection of the model pattern parameter, and the communication with the MAC unit 1. The stock is managed by an 8-bit code, and the code expressed in binary is 10
Handled by the stock number in decimal notation.

The parameter setting sequence section 32 is
After selecting a brand, the MAC starts when the sequence starts.
The model pattern parameter is set in the batch setter 35 in the DCS unit 3 by communicating with the unit 1.

The parameter setting sequence section 32 is
The same value is set as the model pattern parameter in the two batch setters 35 separately. The parameter setting sequence section 32 compares the values set in the setter 18, and if they are the same value, recognizes that the communication has been normally performed,
A value is set in a program setter (not shown). When the model pattern parameters of the two batch setters 18 are different, a message is sent to the MAC unit 1 as a communication error and communication is performed again.

The fuzzy control sequence unit 33 uses M
The fuzzy operation value set from the data setting unit 17 of the AC unit 1 is managed, and the batch setting unit 35 writes the value to the control tag. The fuzzy control sequence unit 33 is activated after a predetermined time has elapsed from the setting of the model pattern parameter, and manages the communication between the MAC unit 1 and the DCS unit 3.

The model pattern sequence section 34
This is a sequence for executing patterning, and optimizes patterning and constructs model patterns for issues for which fuzzy parameters are not constructed. The model pattern sequence unit 34 changes the value of the control factor based on the model pattern of each brand when switching brands in the continuous polymerization of polymers.

Next, the polymer continuous polymerization method realized by the polymer continuous polymerization apparatus will be described. FIG. 10 is a flowchart showing the polymer continuous polymerization method of the example. FIG. 11 is a diagram showing control of the polymer polymerization transition.

First, the MAC unit 1 builds a model pattern (step 100). For example, 18 brands were patterned by a patterning test pattern sequence. In this apparatus, patterning was carried out for each of the three factors, that is, hydrogen feed amount, catalyst feed amount, and propylene feed amount for each brand. FIG. 12 shows an example of a hydrogen model pattern. As shown in FIG. 12, the hydrogen pattern is set to 30 Nm3 / H for 25 minutes and 120 Nm3 / H for 30 minutes during the overaction.

However, the activity of the catalyst constantly changes, and it is difficult to predetermine the feed amount. For this reason, the rate of change in activity is factored from the actual results, and the MAC unit 1 instantaneously calculates the activity at the time of transition to obtain DC.
The method is set in the S section 3.

Since the constructed model pattern has been optimized to some extent based on past results, the polymerization transition time can be shortened. For example, the transition work (including feed-down and up) takes about 12 times on average.
Time. An automated system of transition polymerization reduces the transition time to 10 hours, reducing it by 2 hours.

Next, the operator's station (OP
S) This is done on the screen and the message is transmitted from the DCS unit 3 to the MAC unit 1 (step 101).

The MAC unit 1 receives the message from the DCS unit 3 and sends a confirmation message to the DCS unit 3 (step 102). Next, the MAC unit 1 determines whether the message from the DCS unit 3 is correct (step 10).
3). Here, if the message from the DCS unit 3 is correct, the data setting unit 17 in the MAC unit 1 sets the model pattern parameter in the batch setter 35 in the DCS unit 3 (step 104). Here, for example, the same model pattern parameter is set to the two batch setters 35.
Set to.

Next, the DCS unit 3 determines whether or not the set model pattern parameter has been correctly communicated by comparing the flag with two identical model pattern parameters (step 105).

Here, when the model pattern parameters are correctly communicated, it is determined whether or not the brand is fuzzy set (step 106). If the brand is fuzzy set, the system management sequence section 31 selects the fuzzy control sequence section 33 and activates the fuzzy control sequence section 33 (step 107).

When the fuzzy control sequence section 33 is selected, a transition operation is started with a pre-defined model pattern, and after a lapse of a predetermined time, the mode is switched to fuzzy control (step 108).

Specifically, in the initial state at the time of transition, an over action is taken to change the feed condition to the next brand condition according to the model pattern, and after the predetermined time has elapsed, the fuzzy control causes the gas flow rate ratio H shown in FIG.
/ E Control to match the target value.

Here, the predetermined time is the time for setting the control target value to the value of the next brand after starting the transition and performing the overaction, and is set for each brand. In this case, the fuzzy control manager 11 is
In the fuzzy control period, as shown in (4), gas chromatography (H / E) control and pressure control are performed by the control algorithm. The control algorithm can be changed even during operation, and more precise adjustment can be performed. The control cycle is divided into gas chromatography control and pressure control, and an arbitrary control cycle can be set for each. For example, gas chromatography control is performed every 3 minutes and pressure control is performed every 1 minute.

Next, when the physical properties reach the target values and become stable,
The transition ends, and normal operation starts (step 109). Fuzzy control is continued even in normal operation, and stabilization control is performed during normal operation.

The fuzzy control sequence is continued until the feed up, down and the next transition. However, when the fuzzy control sequence is arbitrarily stopped from the management table, the control is switched to the normal manual operation.

On the other hand, for a brand for which fuzzy settings have not been made, the system management sequence section 31 selects the model pattern sequence section 34 and activates the model pattern sequence section 34 (step 110).

The model pattern sequence section 34 starts the model pattern control shown in FIG. 11 at the time of overaction (step 111). In the pattern correction in this case, the pressure after n minutes is estimated and the pressure is controlled.

During the predetermined time period according to the model pattern, since the over-action occurs, the physical properties are not controlled and only the pressure of the polymerization vessel is controlled. Higher MAC
The data collection unit 16 of the unit 1 collects, for example, five points of pressure data for each minute, and outputs the rate of change of the collected pressure data (step 112). Pressure change for 1 minute dp1 = P-P1, dp2 = P1-P2, dp3 = P2-P3, dp4 = P3-
P4, dp5 = P5-P6 Then, the data collection unit 16 estimates the pressure data after one minute has passed from the present time based on the change rate data (step 113).

Estimated pressure after 1 minute AVPS = SUM (dp1, dp2, dp3, dp
4, dp5) / 5 + polymerizer pressure D201PS data collection unit 16 determines whether or not the estimated pressure value has reached a pressure point (upper limit value / lower limit value) predetermined in the brand condition (step 114).

The data collecting unit 16 controls the pressure by increasing or decreasing the amount of the catalyst at the time point (step 115). It should be noted that one operation amount of the catalyst amount is the difference amount between the first point and the second point of the model pattern from the actual results.

When the pressure point defined by the brand condition is reached,
The MAC unit 1 sends the upper limit pressure signal and the lower limit pressure signal to the DCS unit 3.
To control the sequence (step 11).
6). During this period, gas chromatography control is not performed, so
Manual operation is required to enter the stable state. The patterned sequence is not used for automation, but for optimization of the model pattern.

As described above, according to the embodiment, the polymerization transition of the polymer is automated by the construction of the model pattern and the fuzzy control. In addition, the pattern of the test sequence that has built the model pattern is MA by the program control by the model pattern.
It can be automatically set from the C section 1 and can be used for optimizing future brand-specific patterns.

Therefore, it is possible to develop a fuzzy parameter construction method for any brand other than the representative brand to be automated. As the automation priority issues, there are 18 issues from issue A to issue C including representative issues whose model patterns have already been constructed. The brand B is generated from the brand A in the first step, and the brand C is generated from the brand B in the second step. In this way, 18 different types
A brand has been generated.

Further, although all brands are 59 brands, the polymer continuous polymerization apparatus can support 60 brands. By changing the control factor during the predetermined period of time, only the pressure of the polymerization vessel is controlled as the dependent factor and the physical property is not controlled, so that the transition time can be shortened.

[0088]

According to the present invention, in the control step, the value of the control factor is changed based on the model pattern when the brand is switched in the continuous polymerization of the polymer, and from the time when the value of the control factor is changed in the detection step. After a lapse of a predetermined time, the value of the dependent factor dependent on the value of the control factor is detected.

Further, in the fuzzy control step, fuzzy control is performed so that the value of the control factor and the value of the dependent factor come close to the target value corresponding to the switched brand. That is, since the transition superposition is automatically performed by using the model pattern and by fuzzy control, the switching time can be shortened and the quality of the brand value can be improved.

Further, the value of the control factor is obtained by calculating the numerical value of the detected dependent factor by the fuzzy calculation section, and the value of the obtained control factor is changed. That is, it is possible to approach the target value quickly and stably by the feedback control.

Furthermore, during the period of the predetermined time, the target value is set to the value of the switched brand after the brand switching is started and the control factor is over-actioned, so that the fuzzy control can be performed quickly and stably. The control factor and the dependent factor can be brought close to the target value.

Further, by changing the control factor during the predetermined time period, only the pressure of the polymerization vessel is controlled as the dependent factor and the physical property is not controlled, so that the transition time can be shortened.

Furthermore, since the model pattern is a pattern calculated from past manufacturing results, and this pattern is an optimized pattern to some extent, the transition time can be shortened.

[Brief description of drawings]

FIG. 1 is a principle view showing a polymer continuous polymerization apparatus of the present invention.

FIG. 2 is a principle flowchart showing a polymer continuous polymerization method of the present invention.

FIG. 3 is a hardware configuration diagram showing a polymer continuous polymerization apparatus according to an example of the present invention.

FIG. 4 is a conceptual diagram of model predictive control.

FIG. 5 is a diagram showing a software configuration of a MAC unit of the embodiment.

FIG. 6 is a diagram showing a software configuration of a DCS unit of the embodiment.

FIG. 7 is a diagram showing rules.

FIG. 8 is a diagram showing a membership function.

FIG. 9 is a diagram showing an example of an inference process.

FIG. 10 is a flowchart showing a polymer continuous polymerization method of Example.

FIG. 11 is a diagram showing control of a polymer polymerization transition.

FIG. 12 is a diagram showing an example of a hydrogen model pattern.

[Explanation of symbols]

 1 ... MAC part 3 ... DCS part 4 ... COPSV 5 ... COPSC 6 ... CFCD2 7 ... Input 8 ... Polymer 9 ... Database 9a ... Data file 10 ... MAC software section 11 ... Fuzzy Control management unit 12 Trend monitor unit 13 Database input / output unit 16 Data collection unit 17 Data setting unit 18 Fuzzy controller 21 Simple simulator 22 Transition data collection unit 23 Transition setting Part 24 File selection part 25 Data trend part 31 System management sequence part 32 Parameter setting sequence part 33 Fuzzy control sequence part 34 Model pattern sequence part 35 Batch setter

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yasunori Kishishita 6-1-2 Waki, Waki-cho, Kuga-gun, Yamaguchi Mitsui Petrochemical Industry Co., Ltd. (72) Misao Mizumoto Kazu-gun, Yamaguchi Kimachi Waki 6-1-2 Mitsui Petrochemical Industry Co., Ltd. (72) Inventor Homura Kawamura 6-1-2 Waki, Waki-cho, Kuga-gun, Yamaguchi Prefecture Mitsui Petrochemical Industry Co., Ltd. (72) Invention Inoue Atsuro 6-12, Waki, Waki-cho, Kuga-gun, Yamaguchi Mitsui Petrochemical Industry Co., Ltd. Within Kogyo Co., Ltd. (72) Inventor Nobuaki Sako, 6-1-2, Waki, Waki-cho, Kuga-gun, Yamaguchi Mitsui Chemical Engineering Co., Ltd. (72) Inventor, Yukihiro Hasegawa 3 Chikusa coast, Ichihara Address Mitsui Petrochemical Engineering Co., Ltd.

Claims (28)

[Claims] 1. A method for continuously polymerizing a polymer, which comprises continuously polymerizing a polymer by continuously changing the composition of a polymerization vessel to produce different kinds of brands in sequence, wherein each brand is selected when the brands are changed in the continuous polymerization of the polymer. Control step for changing the value of the control factor based on the model pattern of, a detection step for constantly detecting the value of the dependent factor that changes depending on the value of the control factor, and when the value of the control factor is changed A fuzzy control step of performing fuzzy control so that the value of the control factor and the value of the dependent factor approach a target value corresponding to the switched brand after a predetermined time has passed from Continuous polymerization method. 2. The fuzzy control step obtains the value of the control factor by performing a fuzzy operation on the numerical value of the dependent factor detected in the detection step, and the control step obtains the value of the obtained control factor. The method for continuously polymerizing a polymer according to claim 1, wherein the method is changed. 3. The method for continuously polymerizing a polymer according to claim 2, wherein the calculation is performed by a computer. 4. The predetermined time period according to claim 1, wherein the target value is set to the value of the switched brand after the brand switching is started and the control factor is over-actioned. Polymerization method of continuous polymer. 5. The method for continuously polymerizing a polymer according to claim 1, wherein only the pressure of the polymerization vessel is controlled as the dependent factor by changing the control factor during the predetermined time period. . 6. The model pattern is a pattern calculated from past manufacturing results.
Alternatively, the continuous polymerization method of the polymer according to claim 3. 7. The continuous polymerization method for a polymer according to claim 1, wherein the polymer is a polyolefin. 8. The method for continuously polymerizing a polymer according to claim 7, wherein the polyolefin is polyethylene. 9. The method for continuously polymerizing a polymer according to claim 7, wherein the polyolefin is a polyolefin wax. 10. The continuous polymerization method for polymers according to claim 7, wherein the control factor is a hydrogen feed amount. 11. The method for continuously polymerizing a polymer according to claim 7, wherein the control factor is a catalyst feed amount. 12. The method for continuously polymerizing a polymer according to claim 7, wherein the control factor is a propylene feed amount. 13. The method for continuous polymerization of a polymer according to claim 7, wherein the dependent factor is a ratio of hydrogen to ethylene. 14. The method according to claim 7, wherein the dependent factor is a pressure in the polymerization vessel.
A method for continuous polymerization of a polymer according to one of the claims. 15. A continuous polymer polymerization apparatus for continuously polymerizing a polymer by continuously changing the composition of a polymerization vessel to produce different kinds of brands in order, wherein each brand is selected when the brands are changed in the continuous polymerization of the polymer. When the value of the control factor is changed, a control unit that changes the value of the control factor based on the model pattern of No. 1, a detection unit that constantly detects the value of the dependent factor that changes depending on the value of the control factor, And a fuzzy control unit that performs fuzzy control so that the value of the dependent factor and the value of the dependent factor approach a target value corresponding to the switched brand after a lapse of a predetermined time from The value of the dependent factor detected by the section is input, and after the lapse of the predetermined time, the values of the dependent factor and the control factor are output to the fuzzy control section. Of the continuous polymerization apparatus. 16. The fuzzy control unit obtains the value of the control factor by performing a fuzzy operation on the numerical value of the dependent factor detected by the detection unit, and the control unit obtains the value of the obtained control factor. 16. The continuous polymerization apparatus for polymer according to claim 15, wherein the continuous polymerization apparatus is varied. 17. The continuous polymerization apparatus for polymer according to claim 16, wherein the calculation is performed by a computer. 18. The method according to claim 15, wherein the predetermined time period is a time period in which the target value is set to the value of the switched brand after the brand switching is started and the control factor is over-actioned. Polymer continuous polymerization equipment. 19. The method according to claim 1, wherein only the pressure of the polymerization vessel is controlled as the dependent factor by changing the control factor during the predetermined time period.
5. A continuous polymerization apparatus for polymer according to item 5. 20. The polymer continuous polymerization apparatus according to claim 16 or 17, wherein the model pattern is a pattern calculated from past production records. 21. The polymer continuous polymerization apparatus according to claim 1, wherein the polymer is a polyolefin. 22. The continuous polymer polymerization apparatus according to claim 21, wherein the polyolefin is polyethylene. 23. The continuous polymer polymerization apparatus according to claim 21, wherein the polyolefin is a polyolefin wax. 24. The polymer continuous polymerization apparatus according to any one of claims 21 to 23, wherein the control factor is a hydrogen feed amount. 25. The polymer continuous polymerization apparatus according to claim 21, wherein the control factor is a catalyst feed amount. 26. The polymer continuous polymerization apparatus according to any one of claims 21 to 23, wherein the control factor is a propylene feed amount. 27. The continuous polymerization apparatus for polymer according to claim 21, wherein the dependent factor is a ratio of hydrogen to ethylene. 28. The continuous polymerization apparatus for polymer according to claim 21, wherein the dependent factor is the pressure in the polymerization vessel.